Attachment 71007 Details for Bug 57888 – amnesia output with RADEON_DEBUG=fp

amnesia output with RADEON_DEBUG=fp

amnesia-fp-debug (text/plain), 1.11 MB, created by Scott Moreau on 2012-12-04 18:34:36 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Scott Moreau

Created: 2012-12-04 18:34:36 UTC

Size: 1.11 MB

patch

obsolete

>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], COLOR, COLOR
>DCL OUT[0], COLOR
>  0: MOV OUT[0], IN[0]
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=2*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], COLOR, COLOR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TXP TEMP[0], IN[1].xyyw, SAMP[0], RECT
>  1: MUL TEMP[0].xyz, TEMP[0], IN[0]
>  2: MOV TEMP[0].xyz, TEMP[0].xyzx
>  3: MOV TEMP[0].w, IN[0].wwww
>  4: MOV OUT[0], TEMP[0]
>  5: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0].xyz, temp[0], input[0];
>  2: MOV temp[0].xyz, temp[0].xyzx;
>  3: MOV temp[0].w, input[0].wwww;
>  4: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0].xyz, temp[0], input[0];
>  2: MOV temp[0].xyz, temp[0].xyzx;
>  3: MOV temp[0].w, input[0].wwww;
>  4: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0].xyz, temp[0], input[0];
>  2: MOV temp[0].xyz, temp[0].xyzx;
>  3: MOV temp[0].w, input[0].wwww;
>  4: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0].xyz, temp[0], input[0];
>  2: MOV temp[0].xyz, temp[0].xyzx;
>  3: MOV temp[0].w, input[0].wwww;
>  4: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0].xyz, temp[0], input[0];
>  2: MOV temp[0].xyz, temp[0].xyzx;
>  3: MOV temp[0].w, input[0].wwww;
>  4: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[1], input[1].xyyw, const[0];
>  1: TXP temp[0], temp[1], 2D[0];
>  2: MUL temp[0].xyz, temp[0], input[0];
>  3: MOV temp[0].xyz, temp[0].xyzx;
>  4: MOV temp[0].w, input[0].wwww;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[1], input[1].xyyw, const[0];
>  1: TXP temp[0], temp[1], 2D[0];
>  2: MUL temp[0].xyz, temp[0], input[0];
>  3: MOV temp[0].xyz, temp[0].xyzx;
>  4: MOV temp[0].w, input[0].wwww;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[1].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[0].xyz, temp[1].xy_w, 2D[0];
>  2: MUL temp[0].xyz, temp[0].xyz_, input[0].xyz_;
>  3: MOV temp[0].xyz, temp[0].xyz_;
>  4: MOV temp[0].w, input[0].___w;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[1].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[0].xyz, temp[1].xy_w, 2D[0];
>  2: MUL temp[0].xyz, temp[0].xyz_, input[0].xyz_;
>  3: MOV temp[0].xyz, temp[0].xyz_;
>  4: MOV temp[0].w, input[0].___w;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3].xyz, temp[2].xy_w, 2D[0];
>  2: MUL temp[4].xyz, temp[3].xyz_, input[0].xyz_;
>  3: MOV temp[5].xyz, temp[4].xyz_;
>  4: MOV temp[5].w, input[0].___w;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3].xyz, temp[2].xy_w, 2D[0];
>  2: MUL temp[4].xyz, temp[3].xyz_, input[0].xyz_;
>  3: MOV temp[5].xyz, temp[4].xyz_;
>  4: MOV temp[5].w, input[0].___w;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3].xyz, temp[2].xy_w, 2D[0];
>  2: MUL temp[4].xyz, temp[3].xyz_, input[0].xyz_;
>  3: MOV temp[5].xyz, temp[4].xyz_;
>  4: MOV temp[5].w, input[0].___w;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3].xyz, temp[2].xy_w, 2D[0];
>  2: MUL temp[4].xyz, temp[3].xyz_, input[0].xyz_;
>  3: MOV temp[5].xyz, temp[4].xyz_;
>  4: MOV temp[5].w, input[0].___w;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: TXP temp[3].xyz, temp[2].xy_w, 2D[0];
>  2: src0.xyz = temp[3], src1.xyz = input[0]
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>  3: src0.xyz = temp[4]
>     MAD temp[5].xyz, src0.xyz, src0.111, src0.000
>  4: src0.w = input[0]
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[3].xyz, temp[2].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = input[0], src1.xyz = input[0] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[4]
>     MAD temp[5].xyz, src0.xyz, src0.111, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[3].xyz, temp[2].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = input[0], src1.xyz = input[0] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[4]
>     MAD temp[5].xyz, src0.xyz, src0.111, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>     MAD temp[1].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[1].xyz, temp[1].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[1], src0.w = input[0], src1.xyz = input[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
>  5: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=3*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00000000)
>  0: xyz:  t1  c0  t0  bias-> t1.xy                (01840801)
>       w:  t1  c0  t0  bias-> t1.w                 (00840801)
>     xyz:   t1.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t1.w     c0.w      0.0    op: 00040509
>NODE 1: alu_offset: 1, tex_offset: 0, alu_end: 2, tex_end: 0  (code_addr: 00400081)
>  TEX:
>    TXP t1, t1, texture[0]   (00018041)
>  1: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t0.xyz      0.0    op: 00050200 
>       w:     t0.w      1.0      0.0    op: 00040889
>  2: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], COLOR, COLOR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TXP TEMP[0], IN[1].xyyw, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[0]
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xy_w, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xy_w, 2D[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TXP temp[1], input[1].xy_w, 2D[0];
>  1: MUL temp[2], temp[1], input[0];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TXP temp[1], input[1].xy_w, 2D[0];
>  1: MUL temp[2], temp[1], input[0];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TXP temp[1], input[1].xy_w, 2D[0];
>  1: MUL temp[2], temp[1], input[0];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TXP temp[1], input[1].xy_w, 2D[0];
>  1: MUL temp[2], temp[1], input[0];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TXP temp[1], input[1].xy_w, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[0], src1.w = input[0]
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TXP temp[1], input[1].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TXP temp[1], input[1].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TXP temp[1], input[1].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=4*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TXP t1, t1, texture[0]   (00018041)
>  0: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t1.xyz   t0.xyz      0.0    op: 00050200 
>       w:     t1.w     t0.w      0.0    op: 00040509
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment programFRAG
>DCL IN[0], COLOR, PERSPECTIVE
>DCL OUT[0], COLOR
>  0: MOV OUT[0], IN[0]
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=5*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxx
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=6*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      1.0      1.0      0.0    op: 00050a95 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MOV OUT[0], TEMP[0]
>  2: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=7*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxx
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=8*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      1.0      1.0      0.0    op: 00050a95 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], COLOR, COLOR
>DCL OUT[0], COLOR
>  0: MOV OUT[0], IN[0]
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=9*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL IN[3], GENERIC[14], PERSPECTIVE
>DCL IN[4], GENERIC[15], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {   -0.5000,     0.5000,   255.0000,     0.0039}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[2], 2D
>  1: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[1], 2D
>  2: ADD TEMP[2].xyz, TEMP[1].xyzz, IMM[0].xxxx
>  3: MUL TEMP[3].xyz, TEMP[2].xxxx, IN[3].xyzz
>  4: MAD TEMP[3].xyz, TEMP[2].yyyy, IN[2].xyzz, TEMP[3].xyzz
>  5: MAD TEMP[3].xyz, TEMP[2].zzzz, IN[4].xyzz, TEMP[3].xyzz
>  6: DP3 TEMP[4].x, TEMP[3].xyzz, TEMP[3].xyzz
>  7: RSQ TEMP[4].x, TEMP[4].xxxx
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].yyyy, IMM[0].yyyy
> 10: MUL TEMP[3].x, IN[1].xxxx, IMM[0].zzzz
> 11: FLR TEMP[4].x, TEMP[3].xxxx
> 12: MOV TEMP[5].x, TEMP[4].xxxx
> 13: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 14: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 15: FLR TEMP[4].x, TEMP[3].xxxx
> 16: MOV TEMP[5].y, TEMP[4].xxxx
> 17: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 18: MOV TEMP[5].z, TEMP[3].xxxx
> 19: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].wwww
> 20: MOV TEMP[2].xyz, TEMP[5].xyzx
> 21: TEX TEMP[3].xy, IN[0].xyyy, SAMP[0], 2D
> 22: MOV TEMP[1].w, TEMP[3].xxxx
> 23: MOV TEMP[2].w, TEMP[3].yyyy
> 24: MOV OUT[0], TEMP[0]
> 25: MOV OUT[1], TEMP[1]
> 26: MOV OUT[2], TEMP[2]
> 27: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, temp[4].xxxx;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FLR temp[4].x, temp[3].xxxx;
> 13: MOV temp[5].x, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 16: FLR temp[4].x, temp[3].xxxx;
> 17: MOV temp[5].y, temp[4].xxxx;
> 18: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 19: MOV temp[5].z, temp[3].xxxx;
> 20: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 21: MOV temp[2].xyz, temp[5].xyzx;
> 22: TEX temp[7], input[0].xyyy, 2D[0];
> 23: MOV temp[3].xy, temp[7];
> 24: MOV temp[1].w, temp[3].xxxx;
> 25: MOV temp[2].w, temp[3].yyyy;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, |temp[4].xxxx|;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FRC temp[4].x, temp[3].xxxx;
> 13: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 14: MOV temp[5].x, temp[4].xxxx;
> 15: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 16: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 17: FRC temp[4].x, temp[3].xxxx;
> 18: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 19: MOV temp[5].y, temp[4].xxxx;
> 20: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 21: MOV temp[5].z, temp[3].xxxx;
> 22: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 23: MOV temp[2].xyz, temp[5].xyzx;
> 24: TEX temp[7], input[0].xyyy, 2D[0];
> 25: MOV temp[3].xy, temp[7];
> 26: MOV temp[1].w, temp[3].xxxx;
> 27: MOV temp[2].w, temp[3].yyyy;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[10].xyz, temp[9].xyz_;
>  3: ADD temp[11].xyz, temp[10].xyz_, const[0].xxx_;
>  4: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  5: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  6: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  7: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  8: RSQ temp[16].x, |temp[15].x___|;
>  9: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
> 10: MAD temp[18].xyz, temp[17].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[19].x, input[1].x___, const[0].z___;
> 12: FRC temp[20].x, temp[19].x___;
> 13: ADD temp[21].x, temp[19].x___, -temp[20].x___;
> 14: MOV temp[22].x, temp[21].x___;
> 15: ADD temp[23].x, temp[19].x___, -temp[21].x___;
> 16: MUL temp[24].x, temp[23].x___, const[0].z___;
> 17: FRC temp[25].x, temp[24].x___;
> 18: ADD temp[26].x, temp[24].x___, -temp[25].x___;
> 19: MOV temp[22].y, temp[26]._x__;
> 20: ADD temp[27].x, temp[24].x___, -temp[26].x___;
> 21: MOV temp[28].z, temp[27].__x_;
> 22: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 23: MOV temp[29].xyz, temp[28].xyz_;
> 24: TEX temp[30].xy, input[0].xy__, 2D[0];
> 25: MOV temp[31].xy, temp[30].xy__;
> 26: MOV temp[18].w, temp[31].___x;
> 27: MOV temp[29].w, temp[31].___y;
> 28: MOV output[0], temp[8];
> 29: MOV output[1], temp[18];
> 30: MOV output[2], temp[29];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: src0.xyz = temp[9]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>  3: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  4: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
>  5: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
>  6: src0.xyz = temp[14]
>     DP3 temp[15].x, src0.xyz, src0.xyz
>  7: src0.xyz = temp[15]
>     REPL_ALPHA temp[16].x
>     RSQ, |src0.x|
>  8: src0.xyz = temp[14], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src1.xxx, src0.000
>  9: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 10: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
> 11: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
> 12: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
> 13: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].x, src1.x__, src0.111, -srcp.x__
> 14: src0.xyz = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.x__, src1.z__, src0.000
> 15: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 16: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
> 17: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[27].x, src1.x__, src0.111, -srcp.x__
> 18: src0.xyz = temp[27]
>     MAD temp[28].z, src0.__x, src0.111, src0.000
> 19: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: src0.xyz = temp[30]
>     MAD temp[18].w, src0.x, src0.1, src0.0
> 23: src0.xyz = temp[30]
>     MAD temp[29].w, src0.y, src0.1, src0.0
> 24: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 25: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 26: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.xyz = temp[23], src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.xyz = temp[27], src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[5], input[0].xy__, 2D[2];
>  2: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[0].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[0] SEM_WAIT
>     MAD temp[1].x, src0.x__, src1.z__, src0.0__
>     MAD temp[1].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[6], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[0].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[1]
>     FRC temp[6].x, src0.x__
>  7: src0.xyz = temp[0], src1.xyz = input[3]
>     MAD temp[3].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].x, srcp.x__, src0.11_, src0.00_
>  9: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[0], src1.xyz = input[2], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.z__, src0.0__
> 12: src0.xyz = temp[0], src1.xyz = input[4], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[2]
>     FRC temp[1].x, src0.x__
> 14: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[0], srcp.xyz = (src1 - src0)
>     DP3 temp[1].y, src2.xyz, src2.xyz
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].y, srcp._x_, src0.11_, src0.00_
>     RSQ temp[3].w, |src2.y|
> 16: src0.w = temp[2]
>     MAD temp[1].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[0], src0.w = temp[3]
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[7], src0.w = const[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[1]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=10*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 19, tex_end: 2  (code_addr: 004404c0)
>  TEX:
>    TEX t5, t0, texture[2]   (00009140)
>    TEX t6, t0, texture[1]   (00008980)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t0.x      1.0      0.0    op: 00040886
>  1: xyz:  t6  t0  t0  bias-> t0.xyz               (03800006)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t6.xyz      1.0     -0.5    op: 000d8a80 
>       w:     t0.y      1.0      0.0    op: 00040884
>  2: xyz:  t1  t0  t0  bias-> t6.x                 (00980001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 04804080 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t0  t3  t0  bias-> t3.xyz               (038c00c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t3.xyz      0.0    op: 00050201 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t6  t1  t0   sub-> t7.x                 (009c0046)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  5: xyz:  t6  t1  t0   sub->                      (00000046)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t6.xxx   t6.xxx   t6.xxx    op: 00204081 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  t2  t3  bias-> t1.xyz               (03843080)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.yyy   t2.xyz   t3.xyz    op: 00020202 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.zzz      0.0    op: 0005038c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0  t4  t1  bias-> t0.xyz               (03801100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.zzz   t4.xyz   t1.xyz    op: 00020203 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  9: xyz:  t2  t0  t0  bias-> t1.x                 (00840002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 10: xyz:  t1  t2  t0   sub-> t1.y                 (01040081)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz   t0.xyz   t1.xxx    op: 00a04408 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 11: xyz:  t1  t2  t1   sub-> t7.y                 (011c1081)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:   |t1.y|     t1.x     t1.x    op: 05800047
> 12: xyz:  t0  t0  t0  bias-> t1.z                 (02040000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t0.xyz   t3.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 14: xyz:  t7  t0  t0  bias-> t1.xy                (01840007)
>       w:  c0  t0  t0  bias->                      (00000020)
>     xyz:   t7.xyz   c0.www      0.0    op: 00050600 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      0.5      0.5    op: 00058b00 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t1  t0  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 17: xyz:  t5  t0  t0  bias-> o0.xyz               (1c000005)
>       w:  t5  t0  t0  bias-> o0.w                 (01000005)
>     xyz:   t5.xyz      1.0      0.0    op: 00050a80 
>       w:     t5.w      1.0      0.0    op: 00040889
> 18: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t1  t0  t0  bias-> o1.w                 (03000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
> 19: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t0  t0  t0  bias-> o2.w                 (05000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    0.5000,   255.0000,     0.0039,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[1], 2D
>  1: DP3 TEMP[1].x, IN[2].xyzz, IN[2].xyzz
>  2: RSQ TEMP[2].x, TEMP[1].xxxx
>  3: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[2].xxxx
>  4: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].xxxx, IMM[0].xxxx
>  5: MUL TEMP[3].x, IN[1].xxxx, IMM[0].yyyy
>  6: FLR TEMP[4].x, TEMP[3].xxxx
>  7: MOV TEMP[5].x, TEMP[4].xxxx
>  8: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
>  9: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].yyyy
> 10: FLR TEMP[4].x, TEMP[3].xxxx
> 11: MOV TEMP[5].y, TEMP[4].xxxx
> 12: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 13: MOV TEMP[5].z, TEMP[3].xxxx
> 14: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].zzzz
> 15: MOV TEMP[2].xyz, TEMP[5].xyzx
> 16: TEX TEMP[3].xy, IN[0].xyyy, SAMP[0], 2D
> 17: MOV TEMP[1].w, TEMP[3].xxxx
> 18: MOV TEMP[2].w, TEMP[3].yyyy
> 19: MOV OUT[0], TEMP[0]
> 20: MOV OUT[1], TEMP[1]
> 21: MOV OUT[2], TEMP[2]
> 22: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[6], input[0].xyyy, 2D[0];
> 17: MOV temp[3].xy, temp[6];
> 18: MOV temp[1].w, temp[3].xxxx;
> 19: MOV temp[2].w, temp[3].yyyy;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, |temp[1].xxxx|;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FRC temp[4].x, temp[3].xxxx;
>  7: ADD temp[4].x, temp[3].xxxx, -temp[4];
>  8: MOV temp[5].x, temp[4].xxxx;
>  9: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 10: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 11: FRC temp[4].x, temp[3].xxxx;
> 12: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 13: MOV temp[5].y, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MOV temp[5].z, temp[3].xxxx;
> 16: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 17: MOV temp[2].xyz, temp[5].xyzx;
> 18: TEX temp[6], input[0].xyyy, 2D[0];
> 19: MOV temp[3].xy, temp[6];
> 20: MOV temp[1].w, temp[3].xxxx;
> 21: MOV temp[2].w, temp[3].yyyy;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: TEX temp[6].xy, input[0].xy__, 2D[0];
> 19: MOV temp[3].xy, temp[6].xy__;
> 20: MOV temp[1].w, temp[3].___x;
> 21: MOV temp[2].w, temp[3].___y;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: TEX temp[6].xy, input[0].xy__, 2D[0];
> 19: MOV temp[3].xy, temp[6].xy__;
> 20: MOV temp[1].w, temp[3].___x;
> 21: MOV temp[2].w, temp[3].___y;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: ADD temp[14].x, temp[12].x___, -temp[13].x___;
>  8: MOV temp[15].x, temp[14].x___;
>  9: ADD temp[16].x, temp[12].x___, -temp[14].x___;
> 10: MUL temp[17].x, temp[16].x___, const[0].y___;
> 11: FRC temp[18].x, temp[17].x___;
> 12: ADD temp[19].x, temp[17].x___, -temp[18].x___;
> 13: MOV temp[15].y, temp[19]._x__;
> 14: ADD temp[20].x, temp[17].x___, -temp[19].x___;
> 15: MOV temp[21].z, temp[20].__x_;
> 16: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 17: MOV temp[22].xyz, temp[21].xyz_;
> 18: TEX temp[23].xy, input[0].xy__, 2D[0];
> 19: MOV temp[24].xy, temp[23].xy__;
> 20: MOV temp[11].w, temp[24].___x;
> 21: MOV temp[22].w, temp[24].___y;
> 22: MOV output[0], temp[7];
> 23: MOV output[1], temp[11];
> 24: MOV output[2], temp[22];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: src0.xyz = input[2]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>  2: src0.xyz = temp[8]
>     REPL_ALPHA temp[9].x
>     RSQ, |src0.x|
>  3: src0.xyz = input[2], src1.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src1.xxx, src0.000
>  4: src0.xyz = temp[10]
>     MAD temp[11].xyz, src0.xyz, src0.HHH, src0.HHH
>  5: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>  6: src0.xyz = temp[12]
>     FRC temp[13].x, src0.x__
>  7: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[16].x, src1.x__, src0.111, -srcp.x__
>  9: src0.xyz = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.x__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].x, src1.x__, src0.111, -srcp.x__
> 13: src0.xyz = temp[20]
>     MAD temp[21].z, src0.__x, src0.111, src0.000
> 14: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: src0.xyz = temp[23]
>     MAD temp[11].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[23]
>     MAD temp[22].w, src0.y, src0.1, src0.0
> 19: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 21: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[23].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[23] SEM_WAIT
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>     MAD temp[22].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[23]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>     MAD temp[11].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[12], src1.xyz = temp[8]
>     FRC temp[13].x, src0.x__
>     RSQ temp[9].w, |src1.x|
>  6: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  7: src0.xyz = input[2], src0.w = temp[9], src1.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], src2.xyz = temp[10], srcp.xyz = (src1 - src0)
>     MAD temp[11].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[16].w, src1.x, src0.1, -srcp.x
>  9: src0.xyz = temp[16], src0.w = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.w__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.xyz = temp[20], src0.w = temp[20]
>     MAD temp[21].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[23].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[23] SEM_WAIT
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>     MAD temp[22].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[23]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>     MAD temp[11].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[12], src1.xyz = temp[8]
>     FRC temp[13].x, src0.x__
>     RSQ temp[9].w, |src1.x|
>  6: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  7: src0.xyz = input[2], src0.w = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], src2.xyz = temp[10], srcp.xyz = (src1 - src0)
>     MAD temp[11].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[16].w, src1.x, src0.1, -srcp.x
>  9: src0.w = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.w__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.w = temp[20]
>     MAD temp[21].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[3], input[0].xy__, 2D[1];
>  2: TEX temp[0].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[0] SEM_WAIT
>     MAD temp[1].x, src0.x__, src1.y__, src0.0__
>     MAD temp[1].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[0]
>     DP3 temp[0].x, src0.xyz, src0.xyz
>     MAD temp[0].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[1], src1.xyz = temp[0]
>     FRC temp[0].x, src0.x__
>     RSQ temp[2].w, |src1.x|
>  6: src0.xyz = temp[0], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].x, srcp.x__, src0.11_, src0.00_
>  7: src0.xyz = input[2], src0.w = temp[2]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[0], src1.xyz = temp[1], src2.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[0].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
>  9: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[1].x, src0.w__, src1.y__, src0.0__
> 10: src0.xyz = temp[1]
>     FRC temp[2].x, src0.x__
> 11: src0.xyz = temp[2], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].y, srcp._x_, src0.11_, src0.00_
> 12: src0.xyz = temp[2], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[4], src1.xyz = const[0]
>     MAD temp[1].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.w = temp[2]
>     MAD temp[1].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[1]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[3], src0.w = temp[3]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[1], src0.w = temp[1]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=11*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 15, tex_end: 1  (code_addr: 004203c0)
>  TEX:
>    TEX t3, t0, texture[1]   (000088c0)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.yyy      0.0    op: 00050300 
>       w:     t0.y      1.0      0.0    op: 00040887
>  1: xyz:  t2  t0  t0  bias-> t0.x                 (00800002)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:     t0.x      1.0      0.0    op: 00040883
>  2: xyz:  t1  t0  t0  bias-> t0.x                 (00800001)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:   |t0.x|     t1.x     t1.x    op: 05800043
>  3: xyz:  t0  t1  t0   sub-> t4.x                 (00900040)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.xyz   t2.www      0.0    op: 00050600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  5: xyz:  t0  t1  t2   sub-> t0.xyz               (03802040)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xyz      0.5      0.5    op: 00258b08 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  c0  t0  bias-> t1.x                 (00840800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.yyy      0.0    op: 0005030c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t1  t0  t0  bias-> t2.x                 (00880001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  8: xyz:  t2  t1  t0   sub-> t4.y                 (01100042)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  9: xyz:  t2  t1  t0   sub->                      (00000042)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xxx   t2.xxx   t2.xxx    op: 00204081 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
> 10: xyz:  t4  c0  t0  bias-> t1.xy                (01840804)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> t1.z                 (02040000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t1  t0  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 13: xyz:  t3  t0  t0  bias-> o0.xyz               (1c000003)
>       w:  t3  t0  t0  bias-> o0.w                 (01000003)
>     xyz:   t3.xyz      1.0      0.0    op: 00050a80 
>       w:     t3.w      1.0      0.0    op: 00040889
> 14: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t0  t0  t0  bias-> o1.w                 (03000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
> 15: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t1  t0  t0  bias-> o2.w                 (05000001)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    0.5000,   255.0000,     0.0039,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: DP3 TEMP[1].x, IN[2].xyzz, IN[2].xyzz
>  2: RSQ TEMP[2].x, TEMP[1].xxxx
>  3: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[2].xxxx
>  4: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].xxxx, IMM[0].xxxx
>  5: MUL TEMP[3].x, IN[1].xxxx, IMM[0].yyyy
>  6: FLR TEMP[4].x, TEMP[3].xxxx
>  7: MOV TEMP[5].x, TEMP[4].xxxx
>  8: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
>  9: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].yyyy
> 10: FLR TEMP[4].x, TEMP[3].xxxx
> 11: MOV TEMP[5].y, TEMP[4].xxxx
> 12: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 13: MOV TEMP[5].z, TEMP[3].xxxx
> 14: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].zzzz
> 15: MOV TEMP[2].xyz, TEMP[5].xyzx
> 16: MOV TEMP[1].w, IMM[0].wwww
> 17: MOV TEMP[2].w, IMM[0].wwww
> 18: MOV OUT[0], TEMP[0]
> 19: MOV OUT[1], TEMP[1]
> 20: MOV OUT[2], TEMP[2]
> 21: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, |temp[1].xxxx|;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FRC temp[4].x, temp[3].xxxx;
>  7: ADD temp[4].x, temp[3].xxxx, -temp[4];
>  8: MOV temp[5].x, temp[4].xxxx;
>  9: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 10: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 11: FRC temp[4].x, temp[3].xxxx;
> 12: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 13: MOV temp[5].y, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MOV temp[5].z, temp[3].xxxx;
> 16: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 17: MOV temp[2].xyz, temp[5].xyzx;
> 18: MOV temp[1].w, const[0].wwww;
> 19: MOV temp[2].w, const[0].wwww;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: ADD temp[13].x, temp[11].x___, -temp[12].x___;
>  8: MOV temp[14].x, temp[13].x___;
>  9: ADD temp[15].x, temp[11].x___, -temp[13].x___;
> 10: MUL temp[16].x, temp[15].x___, const[0].y___;
> 11: FRC temp[17].x, temp[16].x___;
> 12: ADD temp[18].x, temp[16].x___, -temp[17].x___;
> 13: MOV temp[14].y, temp[18]._x__;
> 14: ADD temp[19].x, temp[16].x___, -temp[18].x___;
> 15: MOV temp[20].z, temp[19].__x_;
> 16: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 17: MOV temp[21].xyz, temp[20].xyz_;
> 18: MOV temp[10].w, const[0].___w;
> 19: MOV temp[21].w, const[0].___w;
> 20: MOV output[0], temp[6];
> 21: MOV output[1], temp[10];
> 22: MOV output[2], temp[21];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>  2: src0.xyz = temp[7]
>     REPL_ALPHA temp[8].x
>     RSQ, |src0.x|
>  3: src0.xyz = input[2], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src1.xxx, src0.000
>  4: src0.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.HHH, src0.HHH
>  5: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>  6: src0.xyz = temp[11]
>     FRC temp[12].x, src0.x__
>  7: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  8: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, src1.x__, src0.111, -srcp.x__
>  9: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.x__, src1.y__, src0.000
> 10: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].x, src1.x__, src0.111, -srcp.x__
> 13: src0.xyz = temp[19]
>     MAD temp[20].z, src0.__x, src0.111, src0.000
> 14: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 15: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 16: 
>     MAD temp[10].w, src0.0, src0.1, src0.0
> 17: 
>     MAD temp[21].w, src0.0, src0.1, src0.0
> 18: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 19: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.xyz = temp[15], src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.xyz = temp[19], src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[1].x, src0.x__, src1.y__, src0.0__
>     MAD temp[1].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[1].y, src0.xyz, src0.xyz
>     MAD temp[2].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[1], src1.xyz = temp[1]
>     FRC temp[3].x, src0.x__
>     RSQ temp[3].w, |src1.y|
>  5: src0.xyz = temp[3], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].x, srcp.x__, src0.11_, src0.00_
>  6: src0.xyz = input[2], src0.w = temp[3]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[3], src1.xyz = temp[1], src2.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[1].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[3], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.y__, src0.0__
>  9: src0.xyz = temp[2]
>     FRC temp[3].x, src0.x__
> 10: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[4].y, srcp._x_, src0.11_, src0.00_
> 11: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[4], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[3]
>     MAD temp[2].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[1], src0.w = temp[2]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[2], src0.w = temp[1]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=12*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 15, tex_end: 0  (code_addr: 004003c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.yyy      0.0    op: 00050300 
>       w:      0.0      1.0      0.0    op: 00040890
>  1: xyz:  t2  t0  t0  bias-> t1.y                 (01040002)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:      0.0      1.0      0.0    op: 00040890
>  2: xyz:  t1  t1  t0  bias-> t3.x                 (008c0041)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:   |t1.y|     t1.x     t1.x    op: 05800044
>  3: xyz:  t3  t1  t0   sub-> t4.x                 (00900043)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  4: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t2.xyz   t3.www      0.0    op: 00050600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  5: xyz:  t3  t1  t2   sub-> t1.xyz               (03842043)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t2.xyz      0.5      0.5    op: 00258b08 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www   c0.yyy      0.0    op: 0005030c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t2  t0  t0  bias-> t3.x                 (008c0002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  8: xyz:  t3  t2  t0   sub-> t4.y                 (01100083)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  9: xyz:  t3  t2  t0   sub->                      (00000083)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t3.xxx   t3.xxx   t3.xxx    op: 00204081 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 10: xyz:  t4  c0  t0  bias-> t2.xy                (01880804)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> t2.z                 (02080000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
> 14: xyz:  t1  t0  t0  bias-> o1.xyz               (3c000001)
>       w:  t2  t0  t0  bias-> o1.w                 (03000002)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.w      1.0      0.0    op: 00040889
> 15: xyz:  t2  t0  t0  bias-> o2.xyz               (5c000002)
>       w:  t1  t0  t0  bias-> o2.w                 (05000001)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[1]
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1]
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=13*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t0.xyz   t1.xyz      0.0    op: 00050200 
>       w:     t0.w     t1.w      0.0    op: 00040509
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[2]
>DCL CONST[0]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[2].xxxx, CONST[2].yyyy
>  2: TEX TEMP[1].xyz, TEMP[0].xyyy, SAMP[0], RECT
>  3: MUL TEMP[1].xyz, TEMP[1].xyzz, CONST[0].xyzz
>  4: MOV OUT[0], TEMP[1]
>  5: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[0];
>  6: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  7: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[0];
>  6: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  7: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[0];
>  6: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  7: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[0];
>  6: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  7: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[0];
>  6: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  7: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: MUL temp[3], temp[0].xyyy, const[5];
>  6: TEX temp[4], temp[3], 2D[0];
>  7: MOV temp[1].xyz, temp[4];
>  8: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  9: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].wwww;
>  1: MUL temp[2].xyz, input[0], temp[2].wwww;
>  2: MAD temp[2].xyz, temp[2].xyz0, const[3].xyz0, const[4].xyz0;
>  3: MOV temp[0], temp[2];
>  4: MAD temp[0].y, temp[2], const[2].xxxx, const[2].yyyy;
>  5: MUL temp[3], temp[0].xyyy, const[5];
>  6: TEX temp[4], temp[3], 2D[0];
>  7: MOV temp[1].xyz, temp[4];
>  8: MUL temp[1].xyz, temp[1].xyzz, const[0].xyzz;
>  9: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].___w;
>  1: MUL temp[2].xy, input[0].xy__, temp[2].ww__;
>  2: MAD temp[2].xy, temp[2].xy__, const[3].xy__, const[4].xy__;
>  3: MOV temp[0].x, temp[2].x___;
>  4: MAD temp[0].y, temp[2]._y__, const[2]._x__, const[2]._y__;
>  5: MUL temp[3].xy, temp[0].xy__, const[5].xy__;
>  6: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  7: MOV temp[1].xyz, temp[4].xyz_;
>  8: MUL temp[1].xyz, temp[1].xyz_, const[0].xyz_;
>  9: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[2].w, input[0].___w;
>  1: MUL temp[2].xy, input[0].xy__, temp[2].ww__;
>  2: MAD temp[2].xy, temp[2].xy__, const[3].xy__, const[4].xy__;
>  3: MOV temp[0].x, temp[2].x___;
>  4: MAD temp[0].y, temp[2]._y__, const[2]._x__, const[2]._y__;
>  5: MUL temp[3].xy, temp[0].xy__, const[5].xy__;
>  6: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  7: MOV temp[1].xyz, temp[4].xyz_;
>  8: MUL temp[1].xyz, temp[1].xyz_, const[0].xyz_;
>  9: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[7].xy, temp[6].xy__, const[3].xy__, const[4].xy__;
>  3: MOV temp[8].x, temp[7].x___;
>  4: MAD temp[8].y, temp[7]._y__, const[2]._x__, const[2]._y__;
>  5: MUL temp[9].xy, temp[8].xy__, const[5].xy__;
>  6: TEX temp[10].xyz, temp[9].xy__, 2D[0];
>  7: MOV temp[11].xyz, temp[10].xyz_;
>  8: MUL temp[12].xyz, temp[11].xyz_, const[0].xyz_;
>  9: MOV output[0], temp[12];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[7].xy, temp[6].xy__, const[3].xy__, const[4].xy__;
>  3: MOV temp[8].x, temp[7].x___;
>  4: MAD temp[8].y, temp[7]._y__, const[2]._x__, const[2]._y__;
>  5: MUL temp[9].xy, temp[8].xy__, const[5].xy__;
>  6: TEX temp[10].xyz, temp[9].xy__, 2D[0];
>  7: MUL temp[12].xyz, temp[10].xyz_, const[0].xyz_;
>  8: MOV output[0], temp[12];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[7].xy, temp[6].xy__, const[3].xy__, const[4].xy__;
>  3: MOV temp[8].x, temp[7].x___;
>  4: MAD temp[8].y, temp[7]._y__, const[2]._x__, const[2]._y__;
>  5: MUL temp[9].xy, temp[8].xy__, const[5].xy__;
>  6: TEX temp[10].xyz, temp[9].xy__, 2D[0];
>  7: MUL temp[12].xyz, temp[10].xyz_, const[0].xyz_;
>  8: MOV output[0], temp[12];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[7].xy, temp[6].xy__, const[2].xy__, const[3].xy__;
>  3: MOV temp[8].x, temp[7].x___;
>  4: MAD temp[8].y, temp[7]._y__, const[1]._x__, const[1]._y__;
>  5: MUL temp[9].xy, temp[8].xy__, const[4].xy__;
>  6: TEX temp[10].xyz, temp[9].xy__, 2D[0];
>  7: MUL temp[12].xyz, temp[10].xyz_, const[0].xyz_;
>  8: MOV output[0], temp[12];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[5].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[5]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[6], src1.xyz = const[2], src2.xyz = const[3]
>     MAD temp[7].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[7]
>     MAD temp[8].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[7], src1.xyz = const[1]
>     MAD temp[8].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[8], src1.xyz = const[4]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  6: TEX temp[10].xyz, temp[9].xy__, 2D[0];
>  7: src0.xyz = temp[10], src1.xyz = const[0]
>     MAD temp[12].xyz, src0.xyz, src1.xyz, src0.000
>  8: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[5].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[5]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[6], src1.xyz = const[2], src2.xyz = const[3]
>     MAD temp[7].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[7], src1.xyz = const[1]
>     MAD temp[8].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[7]
>     MAD temp[8].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[8], src1.xyz = const[4]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[10].xyz, temp[9].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[10], src1.xyz = const[0] SEM_WAIT
>     MAD temp[12].xyz, src0.xyz, src1.xyz, src0.000
>  9: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[5].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[5]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[6], src1.xyz = const[2], src2.xyz = const[3]
>     MAD temp[7].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[7], src1.xyz = const[1]
>     MAD temp[8].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[7]
>     MAD temp[8].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[8], src1.xyz = const[4]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[10].xyz, temp[9].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[10], src1.xyz = const[0] SEM_WAIT
>     MAD temp[12].xyz, src0.xyz, src1.xyz, src0.000
>  9: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[1].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[1]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src0.00_
>  2: src0.xyz = temp[0], src1.xyz = const[2], src2.xyz = const[3]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[0], src1.xyz = const[1]
>     MAD temp[1].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[0]
>     MAD temp[1].x, src0.x__, src0.11_, src0.00_
>  5: src0.xyz = temp[1], src1.xyz = const[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  6: BEGIN_TEX;
>  7: TEX temp[0].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[0], src1.xyz = const[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>  9: src0.xyz = temp[0], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=14*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 5, tex_end: 0  (code_addr: 00000140)
>  0: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.w     t0.x     t0.x    op: 05000009
>  1: xyz:  t0  t0  t0  bias-> t0.xy                (01800000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xyz   t1.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  2: xyz:  t0  c2  c3  bias-> t0.xy                (01823880)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c2.xyz   c3.xyz    op: 00020200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  c1  t0  bias-> t1.y                 (01040840)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c1.xxx   c1.xyz    op: 00010280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t0  t0  t0  bias-> t1.x                 (00840000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  5: xyz:  t1  c4  t0  bias-> t0.xy                (01800901)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 1: alu_offset: 6, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400046)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  6: xyz:  t0  c0  t0  bias-> t0.xyz               (03800800)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w: t12  t0  t0  bias-> o0.w                 (0100000c)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:    t12.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxx
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=15*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      1.0      1.0      0.0    op: 00050a95 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[7..8]
>DCL CONST[4..6]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>DCL TEMP[6], LOCAL
>DCL TEMP[7], LOCAL
>DCL TEMP[8], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     2.0000}
>IMM[1] FLT32 {   -1.0000,     0.0000,    10.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[8].xxxx, CONST[8].yyyy
>  2: TEX TEMP[1], TEMP[0].xyyy, SAMP[2], RECT
>  3: TEX TEMP[2], TEMP[0].xyyy, SAMP[1], RECT
>  4: RCP TEMP[3].x, IN[1].zzzz
>  5: MUL TEMP[3].xy, IN[1].xyyy, TEMP[3].xxxx
>  6: MOV TEMP[3].z, CONST[4].xxxx
>  7: DP3 TEMP[4].x, TEMP[2].xyzz, IMM[0].xyzz
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: ADD TEMP[4].xyz, CONST[7].xyzz, -TEMP[3].xyzz
> 10: MUL TEMP[4].xyz, TEMP[4].xyzz, CONST[6].xxxx
> 11: DP3 TEMP[5].x, TEMP[4].xyzz, TEMP[4].xyzz
> 12: RSQ TEMP[5].x, TEMP[5].xxxx
> 13: MUL TEMP[5].xyz, TEMP[4].xyzz, TEMP[5].xxxx
> 14: MAD TEMP[6].xyz, TEMP[1].xyzz, IMM[0].wwww, IMM[1].xxxx
> 15: DP3 TEMP[7].x, TEMP[6].xyzz, TEMP[6].xyzz
> 16: RSQ TEMP[7].x, TEMP[7].xxxx
> 17: MUL TEMP[6].xyz, TEMP[6].xyzz, TEMP[7].xxxx
> 18: MOV TEMP[3].xyz, -TEMP[3].xyzx
> 19: DP3 TEMP[7].x, TEMP[3].xyzz, TEMP[3].xyzz
> 20: RSQ TEMP[7].x, TEMP[7].xxxx
> 21: MAD TEMP[3].xyz, TEMP[3].xyzz, TEMP[7].xxxx, TEMP[5].xyzz
> 22: TEX TEMP[7].xyz, TEMP[0].xyyy, SAMP[3], RECT
> 23: MUL TEMP[7].xyz, TEMP[7].xyzz, CONST[5].xyzz
> 24: DP3 TEMP[5].x, TEMP[5].xyzz, TEMP[6].xyzz
> 25: MAX TEMP[5].x, TEMP[5].xxxx, IMM[1].yyyy
> 26: MUL TEMP[1].x, CONST[5].wwww, TEMP[1].wwww
> 27: DP3 TEMP[8].x, TEMP[3].xyzz, TEMP[3].xyzz
> 28: RSQ TEMP[8].x, TEMP[8].xxxx
> 29: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[8].xxxx
> 30: DP3_SAT TEMP[3].x, TEMP[3].xyzz, TEMP[6].xyzz
> 31: MUL TEMP[2].x, TEMP[2].wwww, IMM[1].zzzz
> 32: EX2 TEMP[2].x, TEMP[2].xxxx
> 33: ADD TEMP[2].x, TEMP[2].xxxx, IMM[0].xxxx
> 34: POW TEMP[2].x, TEMP[3].xxxx, TEMP[2].xxxx
> 35: MUL TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
> 36: MUL TEMP[1].xyz, TEMP[1].xxxx, CONST[5].xyzz
> 37: MAD TEMP[1].xyz, TEMP[7].xyzz, TEMP[5].xxxx, TEMP[1].xyzz
> 38: DP3 TEMP[2].x, TEMP[4].xyzz, TEMP[4].xyzz
> 39: TEX TEMP[2].x, TEMP[2].xxxx, SAMP[0], 1D
> 40: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[2].xxxx
> 41: MOV OUT[0], TEMP[1]
> 42: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[4].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 18: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 19: RSQ temp[7].x, temp[7].xxxx;
> 20: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 21: MOV temp[3].xyz, -temp[3].xyzx;
> 22: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 23: RSQ temp[7].x, temp[7].xxxx;
> 24: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 25: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 26: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 27: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 28: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 29: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 30: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 31: RSQ temp[8].x, temp[8].xxxx;
> 32: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 33: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 34: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 35: EX2 temp[2].x, temp[2].xxxx;
> 36: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 37: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 38: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 39: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 40: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 41: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 42: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 43: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 44: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[4].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 18: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 19: RSQ temp[7].x, temp[7].xxxx;
> 20: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 21: MOV temp[3].xyz, -temp[3].xyzx;
> 22: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 23: RSQ temp[7].x, temp[7].xxxx;
> 24: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 25: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 26: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 27: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 28: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 29: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 30: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 31: RSQ temp[8].x, temp[8].xxxx;
> 32: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 33: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 34: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 35: EX2 temp[2].x, temp[2].xxxx;
> 36: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 37: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 38: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 39: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 40: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 41: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 42: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 43: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 44: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[4].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 18: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 19: RSQ temp[7].x, temp[7].xxxx;
> 20: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 21: MOV temp[3].xyz, -temp[3].xyzx;
> 22: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 23: RSQ temp[7].x, temp[7].xxxx;
> 24: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 25: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 26: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 27: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 28: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 29: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 30: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 31: RSQ temp[8].x, temp[8].xxxx;
> 32: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 33: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 34: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 35: EX2 temp[2].x, temp[2].xxxx;
> 36: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 37: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 38: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 39: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 40: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 41: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 42: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 43: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 44: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[4].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 18: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 19: RSQ temp[7].x, temp[7].xxxx;
> 20: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 21: MOV temp[3].xyz, -temp[3].xyzx;
> 22: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 23: RSQ temp[7].x, temp[7].xxxx;
> 24: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 25: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 26: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 27: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 28: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 29: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 30: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 31: RSQ temp[8].x, temp[8].xxxx;
> 32: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 33: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 34: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 35: EX2 temp[2].x, temp[2].xxxx;
> 36: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 37: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 38: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 39: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 40: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 41: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 42: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 43: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 44: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[4].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 18: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 19: RSQ temp[7].x, temp[7].xxxx;
> 20: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 21: MOV temp[3].xyz, -temp[3].xyzx;
> 22: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 23: RSQ temp[7].x, temp[7].xxxx;
> 24: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 25: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 26: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 27: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 28: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 29: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 30: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 31: RSQ temp[8].x, temp[8].xxxx;
> 32: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 33: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 34: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 35: EX2 temp[2].x, temp[2].xxxx;
> 36: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 37: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 38: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 39: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 40: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 41: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 42: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 43: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 44: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: MUL temp[10], temp[0].xyyy, const[13];
>  6: TEX temp[1], temp[10], 2D[2];
>  7: MUL temp[11], temp[0].xyyy, const[14];
>  8: TEX temp[2], temp[11], 2D[1];
>  9: RCP temp[3].x, input[1].zzzz;
> 10: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
> 11: MOV temp[3].z, const[4].xxxx;
> 12: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 13: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 14: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 15: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 16: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 17: RSQ temp[5].x, temp[5].xxxx;
> 18: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 19: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 21: RSQ temp[7].x, temp[7].xxxx;
> 22: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 23: MOV temp[3].xyz, -temp[3].xyzx;
> 24: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 25: RSQ temp[7].x, temp[7].xxxx;
> 26: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 27: MUL temp[12], temp[0].xyyy, const[15];
> 28: TEX temp[13], temp[12], 2D[3];
> 29: MOV temp[7].xyz, temp[13];
> 30: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 33: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 34: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[8].x, temp[8].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 44: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 46: TEX temp[14], temp[2].xxxx, 1D[0];
> 47: MOV temp[2].x, temp[14];
> 48: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 49: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[8].xxxx, const[8].yyyy;
>  5: MUL temp[10], temp[0].xyyy, const[13];
>  6: TEX temp[1], temp[10], 2D[2];
>  7: MUL temp[11], temp[0].xyyy, const[14];
>  8: TEX temp[2], temp[11], 2D[1];
>  9: RCP temp[3].x, input[1].zzzz;
> 10: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
> 11: MOV temp[3].z, const[4].xxxx;
> 12: DP3 temp[4].x, temp[2].xyzz, const[9].xyzz;
> 13: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 14: ADD temp[4].xyz, const[7].xyzz, -temp[3].xyzz;
> 15: MUL temp[4].xyz, temp[4].xyzz, const[6].xxxx;
> 16: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 17: RSQ temp[5].x, |temp[5].xxxx|;
> 18: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 19: MAD temp[6].xyz, temp[1].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 21: RSQ temp[7].x, |temp[7].xxxx|;
> 22: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 23: MOV temp[3].xyz, -temp[3].xyzx;
> 24: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 25: RSQ temp[7].x, |temp[7].xxxx|;
> 26: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 27: MUL temp[12], temp[0].xyyy, const[15];
> 28: TEX temp[13], temp[12], 2D[3];
> 29: MOV temp[7].xyz, temp[13];
> 30: MUL temp[7].xyz, temp[7].xyzz, const[5].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[10].yyyy;
> 33: MUL temp[1].x, const[5].wwww, temp[1].wwww;
> 34: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[8].x, |temp[8].xxxx|;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[10].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[9].xxxx;
> 41: LG2 temp[15].w, temp[3].xxxx;
> 42: MUL temp[15].w, temp[15].wwww, temp[2].xxxx;
> 43: EX2 temp[2].x, temp[15].wwww;
> 44: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 45: MUL temp[1].xyz, temp[1].xxxx, const[5].xyzz;
> 46: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[14], temp[2].xxxx, 1D[0];
> 49: MOV temp[2].x, temp[14];
> 50: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 51: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[9].ww__;
>  2: MAD temp[9].xy, temp[9].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[0].x, temp[9].x___;
>  4: MAD temp[0].y, temp[9]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[10].xy, temp[0].xy__, const[13].xy__;
>  6: TEX temp[1], temp[10].xy__, 2D[2];
>  7: MUL temp[11].xy, temp[0].xy__, const[14].xy__;
>  8: TEX temp[2], temp[11].xy__, 2D[1];
>  9: RCP temp[3].x, input[1].z___;
> 10: MUL temp[3].xy, input[1].xy__, temp[3].xx__;
> 11: MOV temp[3].z, const[4].__x_;
> 12: DP3 temp[4].x, temp[2].xyz_, const[9].xyz_;
> 13: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 14: ADD temp[4].xyz, const[7].xyz_, -temp[3].xyz_;
> 15: MUL temp[4].xyz, temp[4].xyz_, const[6].xxx_;
> 16: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 17: RSQ temp[5].x, |temp[5].x___|;
> 18: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 19: MAD temp[6].xyz, temp[1].xyz_, const[9].www_, const[10].xxx_;
> 20: DP3 temp[7].x, temp[6].xyz_, temp[6].xyz_;
> 21: RSQ temp[7].x, |temp[7].x___|;
> 22: MUL temp[6].xyz, temp[6].xyz_, temp[7].xxx_;
> 23: MOV temp[3].xyz, -temp[3].xyz_;
> 24: DP3 temp[7].x, temp[3].xyz_, temp[3].xyz_;
> 25: RSQ temp[7].x, |temp[7].x___|;
> 26: MAD temp[3].xyz, temp[3].xyz_, temp[7].xxx_, temp[5].xyz_;
> 27: MUL temp[12].xy, temp[0].xy__, const[15].xy__;
> 28: TEX temp[13].xyz, temp[12].xy__, 2D[3];
> 29: MOV temp[7].xyz, temp[13].xyz_;
> 30: MUL temp[7].xyz, temp[7].xyz_, const[5].xyz_;
> 31: DP3 temp[5].x, temp[5].xyz_, temp[6].xyz_;
> 32: MAX temp[5].x, temp[5].x___, const[10].y___;
> 33: MUL temp[1].x, const[5].w___, temp[1].w___;
> 34: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 35: RSQ temp[8].x, |temp[8].x___|;
> 36: MUL temp[3].xyz, temp[3].xyz_, temp[8].xxx_;
> 37: DP3_SAT temp[3].x, temp[3].xyz_, temp[6].xyz_;
> 38: MUL temp[2].x, temp[2].w___, const[10].z___;
> 39: EX2 temp[2].x, temp[2].x___;
> 40: ADD temp[2].x, temp[2].x___, const[9].x___;
> 41: LG2 temp[15].w, temp[3].___x;
> 42: MUL temp[15].w, temp[15].___w, temp[2].___x;
> 43: EX2 temp[2].x, temp[15].w___;
> 44: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 45: MUL temp[1].xyz, temp[1].xxx_, const[5].xyz_;
> 46: MAD temp[1].xyz, temp[7].xyz_, temp[5].xxx_, temp[1].xyz_;
> 47: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 48: TEX temp[14].x, temp[2].x___, 1D[0];
> 49: MOV temp[2].x, temp[14].x___;
> 50: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 51: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[9].ww__;
>  2: MAD temp[9].xy, temp[9].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[0].x, temp[9].x___;
>  4: MAD temp[0].y, temp[9]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[10].xy, temp[0].xy__, const[13].xy__;
>  6: TEX temp[1], temp[10].xy__, 2D[2];
>  7: MUL temp[11].xy, temp[0].xy__, const[14].xy__;
>  8: TEX temp[2], temp[11].xy__, 2D[1];
>  9: RCP temp[3].x, input[1].z___;
> 10: MUL temp[3].xy, input[1].xy__, temp[3].xx__;
> 11: MOV temp[3].z, const[4].__x_;
> 12: DP3 temp[4].x, temp[2].xyz_, const[9].xyz_;
> 13: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 14: ADD temp[4].xyz, const[7].xyz_, -temp[3].xyz_;
> 15: MUL temp[4].xyz, temp[4].xyz_, const[6].xxx_;
> 16: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 17: RSQ temp[5].x, |temp[5].x___|;
> 18: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 19: MAD temp[6].xyz, temp[1].xyz_, const[9].www_, const[10].xxx_;
> 20: DP3 temp[7].x, temp[6].xyz_, temp[6].xyz_;
> 21: RSQ temp[7].x, |temp[7].x___|;
> 22: MUL temp[6].xyz, temp[6].xyz_, temp[7].xxx_;
> 23: MOV temp[3].xyz, -temp[3].xyz_;
> 24: DP3 temp[7].x, temp[3].xyz_, temp[3].xyz_;
> 25: RSQ temp[7].x, |temp[7].x___|;
> 26: MAD temp[3].xyz, temp[3].xyz_, temp[7].xxx_, temp[5].xyz_;
> 27: MUL temp[12].xy, temp[0].xy__, const[15].xy__;
> 28: TEX temp[13].xyz, temp[12].xy__, 2D[3];
> 29: MOV temp[7].xyz, temp[13].xyz_;
> 30: MUL temp[7].xyz, temp[7].xyz_, const[5].xyz_;
> 31: DP3 temp[5].x, temp[5].xyz_, temp[6].xyz_;
> 32: MAX temp[5].x, temp[5].x___, const[10].y___;
> 33: MUL temp[1].x, const[5].w___, temp[1].w___;
> 34: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 35: RSQ temp[8].x, |temp[8].x___|;
> 36: MUL temp[3].xyz, temp[3].xyz_, temp[8].xxx_;
> 37: DP3_SAT temp[3].x, temp[3].xyz_, temp[6].xyz_;
> 38: MUL temp[2].x, temp[2].w___, const[10].z___;
> 39: EX2 temp[2].x, temp[2].x___;
> 40: ADD temp[2].x, temp[2].x___, const[9].x___;
> 41: LG2 temp[15].w, temp[3].___x;
> 42: MUL temp[15].w, temp[15].___w, temp[2].___x;
> 43: EX2 temp[2].x, temp[15].w___;
> 44: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 45: MUL temp[1].xyz, temp[1].xxx_, const[5].xyz_;
> 46: MAD temp[1].xyz, temp[7].xyz_, temp[5].xxx_, temp[1].xyz_;
> 47: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 48: TEX temp[14].x, temp[2].x___, 1D[0];
> 49: MOV temp[2].x, temp[14].x___;
> 50: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 51: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[13].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[14].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: RCP temp[24].x, input[1].z___;
> 10: MUL temp[25].xy, input[1].xy__, temp[24].xx__;
> 11: MOV temp[25].z, const[4].__x_;
> 12: DP3 temp[26].x, temp[23].xyz_, const[9].xyz_;
> 13: MUL temp[27].xyz, temp[25].xyz_, temp[26].xxx_;
> 14: ADD temp[28].xyz, const[7].xyz_, -temp[27].xyz_;
> 15: MUL temp[29].xyz, temp[28].xyz_, const[6].xxx_;
> 16: DP3 temp[30].x, temp[29].xyz_, temp[29].xyz_;
> 17: RSQ temp[31].x, |temp[30].x___|;
> 18: MUL temp[32].xyz, temp[29].xyz_, temp[31].xxx_;
> 19: MAD temp[33].xyz, temp[21].xyz_, const[9].www_, const[10].xxx_;
> 20: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 21: RSQ temp[35].x, |temp[34].x___|;
> 22: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 23: MOV temp[37].xyz, -temp[27].xyz_;
> 24: DP3 temp[38].x, temp[37].xyz_, temp[37].xyz_;
> 25: RSQ temp[39].x, |temp[38].x___|;
> 26: MAD temp[40].xyz, temp[37].xyz_, temp[39].xxx_, temp[32].xyz_;
> 27: MUL temp[41].xy, temp[19].xy__, const[15].xy__;
> 28: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 29: MOV temp[43].xyz, temp[42].xyz_;
> 30: MUL temp[44].xyz, temp[43].xyz_, const[5].xyz_;
> 31: DP3 temp[45].x, temp[32].xyz_, temp[36].xyz_;
> 32: MAX temp[46].x, temp[45].x___, const[10].y___;
> 33: MUL temp[47].x, const[5].w___, temp[21].w___;
> 34: DP3 temp[48].x, temp[40].xyz_, temp[40].xyz_;
> 35: RSQ temp[49].x, |temp[48].x___|;
> 36: MUL temp[50].xyz, temp[40].xyz_, temp[49].xxx_;
> 37: DP3_SAT temp[51].x, temp[50].xyz_, temp[36].xyz_;
> 38: MUL temp[52].x, temp[23].w___, const[10].z___;
> 39: EX2 temp[53].x, temp[52].x___;
> 40: ADD temp[54].x, temp[53].x___, const[9].x___;
> 41: LG2 temp[55].w, temp[51].___x;
> 42: MUL temp[56].w, temp[55].___w, temp[54].___x;
> 43: EX2 temp[57].x, temp[56].w___;
> 44: MUL temp[58].x, temp[47].x___, temp[57].x___;
> 45: MUL temp[59].xyz, temp[58].xxx_, const[5].xyz_;
> 46: MAD temp[60].xyz, temp[44].xyz_, temp[46].xxx_, temp[59].xyz_;
> 47: DP3 temp[61].x, temp[29].xyz_, temp[29].xyz_;
> 48: TEX temp[62].x, temp[61].x___, 1D[0];
> 49: MOV temp[63].x, temp[62].x___;
> 50: MUL temp[21].xyz, temp[60].xyz_, temp[63].xxx_;
> 51: MOV output[0], temp[21];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[13].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[14].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: RCP temp[24].x, input[1].z___;
> 10: MUL temp[25].xy, input[1].xy__, temp[24].xx__;
> 11: MOV temp[25].z, const[4].__x_;
> 12: DP3 temp[26].x, temp[23].xyz_, const[9].xyz_;
> 13: MUL temp[27].xyz, temp[25].xyz_, temp[26].xxx_;
> 14: MUL temp[29].xyz, (const[7] - temp[27]).xyz_, const[6].xxx_;
> 15: DP3 temp[30].x, temp[29].xyz_, temp[29].xyz_;
> 16: RSQ temp[31].x, |temp[30].x___|;
> 17: MUL temp[32].xyz, temp[29].xyz_, temp[31].xxx_;
> 18: MAD temp[33].xyz, temp[21].xyz_, const[9].www_, none.-1-1-1_;
> 19: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 20: RSQ temp[35].x, |temp[34].x___|;
> 21: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 22: DP3 temp[38].x, temp[27].-x-y-z_, temp[27].-x-y-z_;
> 23: RSQ temp[39].x, |temp[38].x___|;
> 24: MAD temp[40].xyz, temp[27].-x-y-z_, temp[39].xxx_, temp[32].xyz_;
> 25: MUL temp[41].xy, temp[19].xy__, const[15].xy__;
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 27: MUL temp[44].xyz, temp[42].xyz_, const[5].xyz_;
> 28: DP3 temp[45].x, temp[32].xyz_, temp[36].xyz_;
> 29: MAX temp[46].x, temp[45].x___, none.0___;
> 30: MUL temp[47].x, const[5].w___, temp[21].w___;
> 31: DP3 temp[48].x, temp[40].xyz_, temp[40].xyz_;
> 32: RSQ temp[49].x, |temp[48].x___|;
> 33: MUL temp[50].xyz, temp[40].xyz_, temp[49].xxx_;
> 34: DP3_SAT temp[51].x, temp[50].xyz_, temp[36].xyz_;
> 35: MUL temp[52].x, temp[23].w___, const[10].z___;
> 36: EX2 temp[53].x, temp[52].x___;
> 37: ADD temp[54].x, temp[53].x___, none.1___;
> 38: LG2 temp[55].w, temp[51].___x;
> 39: MUL temp[56].w, temp[55].___w, temp[54].___x;
> 40: EX2 temp[57].x, temp[56].w___;
> 41: MUL temp[58].x, temp[47].x___, temp[57].x___;
> 42: MUL temp[59].xyz, temp[58].xxx_, const[5].xyz_;
> 43: MAD temp[60].xyz, temp[44].xyz_, temp[46].xxx_, temp[59].xyz_;
> 44: DP3 temp[61].x, temp[29].xyz_, temp[29].xyz_;
> 45: TEX temp[62].x, temp[61].x___, 1D[0];
> 46: MUL temp[21].xyz, temp[60].xyz_, temp[62].xxx_;
> 47: MOV output[0], temp[21];
>CONST[9] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[10] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[13].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[14].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: RCP temp[24].x, input[1].z___;
> 10: MUL temp[25].xy, input[1].xy__, temp[24].xx__;
> 11: MOV temp[25].z, const[4].__x_;
> 12: DP3 temp[26].x, temp[23].xyz_, const[9].xyz_;
> 13: MUL temp[27].xyz, temp[25].xyz_, temp[26].xxx_;
> 14: MUL temp[29].xyz, (const[7] - temp[27]).xyz_, const[6].xxx_;
> 15: DP3 temp[30].x, temp[29].xyz_, temp[29].xyz_;
> 16: RSQ temp[31].x, |temp[30].x___|;
> 17: MUL temp[32].xyz, temp[29].xyz_, temp[31].xxx_;
> 18: MAD temp[33].xyz, temp[21].xyz_, const[9].www_, none.-1-1-1_;
> 19: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 20: RSQ temp[35].x, |temp[34].x___|;
> 21: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 22: DP3 temp[38].x, temp[27].-x-y-z_, temp[27].-x-y-z_;
> 23: RSQ temp[39].x, |temp[38].x___|;
> 24: MAD temp[40].xyz, temp[27].-x-y-z_, temp[39].xxx_, temp[32].xyz_;
> 25: MUL temp[41].xy, temp[19].xy__, const[15].xy__;
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 27: MUL temp[44].xyz, temp[42].xyz_, const[5].xyz_;
> 28: DP3 temp[45].x, temp[32].xyz_, temp[36].xyz_;
> 29: MAX temp[46].x, temp[45].x___, none.0___;
> 30: MUL temp[47].x, const[5].w___, temp[21].w___;
> 31: DP3 temp[48].x, temp[40].xyz_, temp[40].xyz_;
> 32: RSQ temp[49].x, |temp[48].x___|;
> 33: MUL temp[50].xyz, temp[40].xyz_, temp[49].xxx_;
> 34: DP3_SAT temp[51].x, temp[50].xyz_, temp[36].xyz_;
> 35: MUL temp[52].x, temp[23].w___, const[10].z___;
> 36: EX2 temp[53].x, temp[52].x___;
> 37: ADD temp[54].x, temp[53].x___, none.1___;
> 38: LG2 temp[55].w, temp[51].___x;
> 39: MUL temp[56].w, temp[55].___w, temp[54].___x;
> 40: EX2 temp[57].x, temp[56].w___;
> 41: MUL temp[58].x, temp[47].x___, temp[57].x___;
> 42: MUL temp[59].xyz, temp[58].xxx_, const[5].xyz_;
> 43: MAD temp[60].xyz, temp[44].xyz_, temp[46].xxx_, temp[59].xyz_;
> 44: DP3 temp[61].x, temp[29].xyz_, temp[29].xyz_;
> 45: TEX temp[62].x, temp[61].x___, 1D[0];
> 46: MUL temp[21].xyz, temp[60].xyz_, temp[62].xxx_;
> 47: MOV output[0], temp[21];
>CONST[5] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[6] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[4]._x__, const[4]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[9].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[10].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: RCP temp[24].x, input[1].z___;
> 10: MUL temp[25].xy, input[1].xy__, temp[24].xx__;
> 11: MOV temp[25].z, const[0].__x_;
> 12: DP3 temp[26].x, temp[23].xyz_, const[5].xyz_;
> 13: MUL temp[27].xyz, temp[25].xyz_, temp[26].xxx_;
> 14: MUL temp[29].xyz, (const[3] - temp[27]).xyz_, const[2].xxx_;
> 15: DP3 temp[30].x, temp[29].xyz_, temp[29].xyz_;
> 16: RSQ temp[31].x, |temp[30].x___|;
> 17: MUL temp[32].xyz, temp[29].xyz_, temp[31].xxx_;
> 18: MAD temp[33].xyz, temp[21].xyz_, const[5].www_, none.-1-1-1_;
> 19: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 20: RSQ temp[35].x, |temp[34].x___|;
> 21: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 22: DP3 temp[38].x, temp[27].-x-y-z_, temp[27].-x-y-z_;
> 23: RSQ temp[39].x, |temp[38].x___|;
> 24: MAD temp[40].xyz, temp[27].-x-y-z_, temp[39].xxx_, temp[32].xyz_;
> 25: MUL temp[41].xy, temp[19].xy__, const[11].xy__;
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 27: MUL temp[44].xyz, temp[42].xyz_, const[1].xyz_;
> 28: DP3 temp[45].x, temp[32].xyz_, temp[36].xyz_;
> 29: MAX temp[46].x, temp[45].x___, none.0___;
> 30: MUL temp[47].x, const[1].w___, temp[21].w___;
> 31: DP3 temp[48].x, temp[40].xyz_, temp[40].xyz_;
> 32: RSQ temp[49].x, |temp[48].x___|;
> 33: MUL temp[50].xyz, temp[40].xyz_, temp[49].xxx_;
> 34: DP3_SAT temp[51].x, temp[50].xyz_, temp[36].xyz_;
> 35: MUL temp[52].x, temp[23].w___, const[6].z___;
> 36: EX2 temp[53].x, temp[52].x___;
> 37: ADD temp[54].x, temp[53].x___, none.1___;
> 38: LG2 temp[55].w, temp[51].___x;
> 39: MUL temp[56].w, temp[55].___w, temp[54].___x;
> 40: EX2 temp[57].x, temp[56].w___;
> 41: MUL temp[58].x, temp[47].x___, temp[57].x___;
> 42: MUL temp[59].xyz, temp[58].xxx_, const[1].xyz_;
> 43: MAD temp[60].xyz, temp[44].xyz_, temp[46].xxx_, temp[59].xyz_;
> 44: DP3 temp[61].x, temp[29].xyz_, temp[29].xyz_;
> 45: TEX temp[62].x, temp[61].x___, 1D[0];
> 46: MUL temp[21].xyz, temp[60].xyz_, temp[62].xxx_;
> 47: MOV output[0], temp[21];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[17], src1.xyz = const[7], src2.xyz = const[8]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[18], src1.xyz = const[4]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: src0.xyz = input[1]
>     REPL_ALPHA temp[24].x
>     RCP, src0.z
> 10: src0.xyz = input[1], src1.xyz = temp[24]
>     MAD temp[25].xy, src0.xy_, src1.xx_, src0.000
> 11: src0.xyz = const[0]
>     MAD temp[25].z, src0.__x, src0.111, src0.000
> 12: src0.xyz = temp[23], src1.xyz = const[5]
>     DP3 temp[26].x, src0.xyz, src1.xyz
> 13: src0.xyz = temp[25], src1.xyz = temp[26]
>     MAD temp[27].xyz, src0.xyz, src1.xxx, src0.000
> 14: src0.xyz = temp[27], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[29].xyz, srcp.xyz, src2.xxx, src0.000
> 15: src0.xyz = temp[29]
>     DP3 temp[30].x, src0.xyz, src0.xyz
> 16: src0.xyz = temp[30]
>     REPL_ALPHA temp[31].x
>     RSQ, |src0.x|
> 17: src0.xyz = temp[29], src1.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src0.000
> 18: src0.xyz = temp[21], src0.w = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.www, -src0.111
> 19: src0.xyz = temp[33]
>     DP3 temp[34].x, src0.xyz, src0.xyz
> 20: src0.xyz = temp[34]
>     REPL_ALPHA temp[35].x
>     RSQ, |src0.x|
> 21: src0.xyz = temp[33], src1.xyz = temp[35]
>     MAD temp[36].xyz, src0.xyz, src1.xxx, src0.000
> 22: src0.xyz = temp[27]
>     DP3 temp[38].x, -src0.xyz, -src0.xyz
> 23: src0.xyz = temp[38]
>     REPL_ALPHA temp[39].x
>     RSQ, |src0.x|
> 24: src0.xyz = temp[27], src1.xyz = temp[39], src2.xyz = temp[32]
>     MAD temp[40].xyz, -src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[19], src1.xyz = const[11]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 27: src0.xyz = temp[42], src1.xyz = const[1]
>     MAD temp[44].xyz, src0.xyz, src1.xyz, src0.000
> 28: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[45].x, src0.xyz, src1.xyz
> 29: src0.xyz = temp[45]
>     MAX temp[46].x, src0.x__, src0.0__
> 30: src0.w = const[1], src1.w = temp[21]
>     MAD temp[47].x, src0.w__, src1.w__, src0.000
> 31: src0.xyz = temp[40]
>     DP3 temp[48].x, src0.xyz, src0.xyz
> 32: src0.xyz = temp[48]
>     REPL_ALPHA temp[49].x
>     RSQ, |src0.x|
> 33: src0.xyz = temp[40], src1.xyz = temp[49]
>     MAD temp[50].xyz, src0.xyz, src1.xxx, src0.000
> 34: src0.xyz = temp[50], src1.xyz = temp[36]
>     DP3_SAT temp[51].x, src0.xyz, src1.xyz
> 35: src0.xyz = const[6], src0.w = temp[23]
>     MAD temp[52].x, src0.w__, src0.z__, src0.000
> 36: src0.xyz = temp[52]
>     REPL_ALPHA temp[53].x
>     EX2, src0.x
> 37: src0.xyz = temp[53]
>     MAD temp[54].x, src0.x__, src0.111, src0.1__
> 38: src0.xyz = temp[51]
>     LG2 temp[55].w, src0.x
> 39: src0.xyz = temp[54], src0.w = temp[55]
>     MAD temp[56].w, src0.w, src0.x, src0.0
> 40: src0.w = temp[56]
>     REPL_ALPHA temp[57].x
>     EX2, src0.w
> 41: src0.xyz = temp[47], src1.xyz = temp[57]
>     MAD temp[58].x, src0.x__, src1.x__, src0.000
> 42: src0.xyz = temp[58], src1.xyz = const[1]
>     MAD temp[59].xyz, src0.xxx, src1.xyz, src0.000
> 43: src0.xyz = temp[44], src1.xyz = temp[46], src2.xyz = temp[59]
>     MAD temp[60].xyz, src0.xyz, src1.xxx, src2.xyz
> 44: src0.xyz = temp[29]
>     DP3 temp[61].x, src0.xyz, src0.xyz
> 45: TEX temp[62].x, temp[61].x___, 1D[0];
> 46: src0.xyz = temp[60], src1.xyz = temp[62]
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 47: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[0]
>     MAD temp[25].z, src0.__x, src0.111, src0.000
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16], src1.xyz = input[1]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[24].w, src1.z
>  2: src0.xyz = temp[17], src1.xyz = const[7], src2.xyz = const[8]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[24], src1.xyz = temp[24]
>     MAD temp[25].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[18], src1.xyz = const[4]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[19], src1.xyz = const[11]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 11: TEX temp[23], temp[22].xy__, 2D[1];
> 12: TEX temp[21], temp[20].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[23], src0.w = const[1], src1.xyz = const[5], src1.w = temp[21] SEM_WAIT
>     DP3 temp[26].x, src0.xyz, src1.xyz
>     MAD temp[47].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[42], src0.w = temp[23], src1.xyz = const[1], src2.xyz = const[6]
>     MAD temp[44].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[52].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[21], src0.w = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[25], src1.xyz = temp[26]
>     MAD temp[27].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[52], src0.w = temp[52]
>     REPL_ALPHA temp[53].x
>     EX2, src0.w
> 18: src0.xyz = temp[33], src1.xyz = temp[53]
>     DP3 temp[34].x, src0.xyz, src0.xyz
>     MAD temp[54].w, src1.x, src0.1, src0.1
> 19: src0.xyz = temp[27], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[29].xyz, srcp.xyz, src2.xxx, src0.000
> 20: src0.xyz = temp[27]
>     DP3 temp[38].x, -src0.xyz, -src0.xyz
> 21: src0.xyz = temp[29], src1.xyz = temp[38]
>     DP3 temp[30].x, src0.xyz, src0.xyz
>     RSQ temp[39].w, |src1.x|
> 22: src0.xyz = temp[29], src1.xyz = temp[34]
>     DP3 temp[61].x, src0.xyz, src0.xyz
>     RSQ temp[35].w, |src1.x|
> 23: src0.xyz = temp[33], src0.w = temp[35], src1.xyz = temp[35], src2.xyz = temp[30]
>     MAD temp[36].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[31].w, |src2.x|
> 24: src0.xyz = temp[29], src0.w = temp[31], src1.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src0.www, src0.000
> 25: src0.xyz = temp[27], src0.w = temp[39], src1.xyz = temp[39], src2.xyz = temp[32]
>     MAD temp[40].xyz, -src0.xyz, src0.www, src2.xyz
> 26: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[45].x, src0.xyz, src1.xyz
> 27: src0.xyz = temp[40], src1.xyz = temp[45]
>     DP3 temp[48].x, src0.xyz, src0.xyz
>     MAX temp[46].w, src1.x, src0.0
> 28: src0.xyz = temp[48]
>     REPL_ALPHA temp[49].x
>     RSQ, |src0.x|
> 29: src0.xyz = temp[40], src1.xyz = temp[49]
>     MAD temp[50].xyz, src0.xyz, src1.xxx, src0.000
> 30: src0.xyz = temp[50], src1.xyz = temp[36]
>     DP3_SAT temp[51].x, src0.xyz, src1.xyz
> 31: src0.xyz = temp[51]
>     LG2 temp[55].w, src0.x
> 32: src0.xyz = temp[54], src0.w = temp[55], src1.w = temp[54]
>     MAD temp[56].w, src0.w, src1.w, src0.0
> 33: src0.w = temp[56]
>     REPL_ALPHA temp[57].x
>     EX2, src0.w
> 34: src0.xyz = temp[47], src0.w = temp[47], src1.xyz = temp[57]
>     MAD temp[58].x, src0.w__, src1.x__, src0.000
> 35: src0.xyz = temp[58], src1.xyz = const[1]
>     MAD temp[59].xyz, src0.xxx, src1.xyz, src0.000
> 36: src0.xyz = temp[44], src0.w = temp[46], src1.xyz = temp[46], src2.xyz = temp[59]
>     MAD temp[60].xyz, src0.xyz, src0.www, src2.xyz
> 37: BEGIN_TEX;
> 38: TEX temp[62].x, temp[61].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[60], src1.xyz = temp[62] SEM_WAIT
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 40: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[0]
>     MAD temp[25].z, src0.__x, src0.111, src0.000
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16], src1.xyz = input[1]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[24].w, src1.z
>  2: src0.xyz = temp[17], src1.xyz = const[7], src2.xyz = const[8]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[24]
>     MAD temp[25].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[18], src1.xyz = const[4]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[19], src1.xyz = const[11]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[42].xyz, temp[41].xy__, 2D[3];
> 11: TEX temp[23], temp[22].xy__, 2D[1];
> 12: TEX temp[21], temp[20].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[23], src0.w = const[1], src1.xyz = const[5], src1.w = temp[21] SEM_WAIT
>     DP3 temp[26].x, src0.xyz, src1.xyz
>     MAD temp[47].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[42], src0.w = temp[23], src1.xyz = const[1], src2.xyz = const[6]
>     MAD temp[44].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[52].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[21], src0.w = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[25], src1.xyz = temp[26]
>     MAD temp[27].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.w = temp[52]
>     REPL_ALPHA temp[53].x
>     EX2, src0.w
> 18: src0.xyz = temp[33], src1.xyz = temp[53]
>     DP3 temp[34].x, src0.xyz, src0.xyz
>     MAD temp[54].w, src1.x, src0.1, src0.1
> 19: src0.xyz = temp[27], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[29].xyz, srcp.xyz, src2.xxx, src0.000
> 20: src0.xyz = temp[27]
>     DP3 temp[38].x, -src0.xyz, -src0.xyz
> 21: src0.xyz = temp[29], src1.xyz = temp[38]
>     DP3 temp[30].x, src0.xyz, src0.xyz
>     RSQ temp[39].w, |src1.x|
> 22: src0.xyz = temp[29], src1.xyz = temp[34]
>     DP3 temp[61].x, src0.xyz, src0.xyz
>     RSQ temp[35].w, |src1.x|
> 23: src0.xyz = temp[33], src0.w = temp[35], src2.xyz = temp[30]
>     MAD temp[36].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[31].w, |src2.x|
> 24: src0.xyz = temp[29], src0.w = temp[31]
>     MAD temp[32].xyz, src0.xyz, src0.www, src0.000
> 25: src0.xyz = temp[27], src0.w = temp[39], src2.xyz = temp[32]
>     MAD temp[40].xyz, -src0.xyz, src0.www, src2.xyz
> 26: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[45].x, src0.xyz, src1.xyz
> 27: src0.xyz = temp[40], src1.xyz = temp[45]
>     DP3 temp[48].x, src0.xyz, src0.xyz
>     MAX temp[46].w, src1.x, src0.0
> 28: src0.xyz = temp[48]
>     REPL_ALPHA temp[49].x
>     RSQ, |src0.x|
> 29: src0.xyz = temp[40], src1.xyz = temp[49]
>     MAD temp[50].xyz, src0.xyz, src1.xxx, src0.000
> 30: src0.xyz = temp[50], src1.xyz = temp[36]
>     DP3_SAT temp[51].x, src0.xyz, src1.xyz
> 31: src0.xyz = temp[51]
>     LG2 temp[55].w, src0.x
> 32: src0.w = temp[55], src1.w = temp[54]
>     MAD temp[56].w, src0.w, src1.w, src0.0
> 33: src0.w = temp[56]
>     REPL_ALPHA temp[57].x
>     EX2, src0.w
> 34: src0.w = temp[47], src1.xyz = temp[57]
>     MAD temp[58].x, src0.w__, src1.x__, src0.000
> 35: src0.xyz = temp[58], src1.xyz = const[1]
>     MAD temp[59].xyz, src0.xxx, src1.xyz, src0.000
> 36: src0.xyz = temp[44], src0.w = temp[46], src2.xyz = temp[59]
>     MAD temp[60].xyz, src0.xyz, src0.www, src2.xyz
> 37: BEGIN_TEX;
> 38: TEX temp[62].x, temp[61].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[60], src1.xyz = temp[62] SEM_WAIT
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 40: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[1]
>     MAD temp[2].z, src0.__x, src0.111, src0.000
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0], src1.xyz = input[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.00_
>     RCP temp[0].w, src1.z
>  2: src0.xyz = temp[1], src1.xyz = const[7], src2.xyz = const[8]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[0], src0.w = temp[0]
>     MAD temp[2].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[1], src1.xyz = const[4]
>     MAD temp[0].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[1]
>     MAD temp[0].x, src0.x__, src0.11_, src0.00_
>  6: src0.xyz = temp[0], src1.xyz = const[11]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[0], src1.xyz = const[10]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[0], src1.xyz = const[9]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  9: BEGIN_TEX;
> 10: TEX temp[1].xyz, temp[1].xy__, 2D[3];
> 11: TEX temp[3], temp[3].xy__, 2D[1];
> 12: TEX temp[0], temp[0].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[3], src0.w = const[1], src1.xyz = const[5], src1.w = temp[0] SEM_WAIT
>     DP3 temp[4].x, src0.xyz, src1.xyz
>     MAD temp[2].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[1], src0.w = temp[3], src1.xyz = const[1], src2.xyz = const[6]
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[0], src0.w = const[5]
>     MAD temp[3].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[2], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.w = temp[1]
>     REPL_ALPHA temp[4].x
>     EX2, src0.w
> 18: src0.xyz = temp[3], src1.xyz = temp[4]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     MAD temp[1].w, src1.x, src0.1, src0.1
> 19: src0.xyz = temp[2], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[5].xyz, srcp.xyz, src2.xxx, src0.000
> 20: src0.xyz = temp[2]
>     DP3 temp[4].y, -src0.xyz, -src0.xyz
> 21: src0.xyz = temp[5], src1.xyz = temp[4]
>     DP3 temp[4].y, src0.xyz, src0.xyz
>     RSQ temp[3].w, |src1.y|
> 22: src0.xyz = temp[5], src1.xyz = temp[4]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     RSQ temp[4].w, |src1.x|
> 23: src0.xyz = temp[3], src0.w = temp[4], src2.xyz = temp[4]
>     MAD temp[3].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[4].w, |src2.y|
> 24: src0.xyz = temp[5], src0.w = temp[4]
>     MAD temp[5].xyz, src0.xyz, src0.www, src0.000
> 25: src0.xyz = temp[2], src0.w = temp[3], src2.xyz = temp[5]
>     MAD temp[2].xyz, -src0.xyz, src0.www, src2.xyz
> 26: src0.xyz = temp[5], src1.xyz = temp[3]
>     DP3 temp[4].y, src0.xyz, src1.xyz
> 27: src0.xyz = temp[2], src1.xyz = temp[4]
>     DP3 temp[4].y, src0.xyz, src0.xyz
>     MAX temp[3].w, src1.y, src0.0
> 28: src0.xyz = temp[4]
>     REPL_ALPHA temp[4].y
>     RSQ, |src0.y|
> 29: src0.xyz = temp[2], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.yyy, src0.000
> 30: src0.xyz = temp[2], src1.xyz = temp[3]
>     DP3_SAT temp[2].x, src0.xyz, src1.xyz
> 31: src0.xyz = temp[2]
>     LG2 temp[4].w, src0.x
> 32: src0.w = temp[4], src1.w = temp[1]
>     MAD temp[1].w, src0.w, src1.w, src0.0
> 33: src0.w = temp[1]
>     REPL_ALPHA temp[2].x
>     EX2, src0.w
> 34: src0.w = temp[2], src1.xyz = temp[2]
>     MAD temp[2].x, src0.w__, src1.x__, src0.0__
> 35: src0.xyz = temp[2], src1.xyz = const[1]
>     MAD temp[2].xyz, src0.xxx, src1.xyz, src0.000
> 36: src0.xyz = temp[1], src0.w = temp[3], src2.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src0.www, src2.xyz
> 37: BEGIN_TEX;
> 38: TEX temp[2].x, temp[4].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[1], src1.xyz = temp[2] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src1.xxx, src0.000
> 40: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=16*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 8, tex_end: 0  (code_addr: 00000200)
>  0: xyz:  c0  t0  t0  bias-> t2.z                 (02080020)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   c0.xxx      1.0      0.0    op: 00050a81 
>       w:     t1.w     c0.x     c0.x    op: 05000009
>  1: xyz:  t1  t0  t0  bias-> t1.xy                (01840001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.z     t1.x     t1.x    op: 05000005
>  2: xyz:  t1  c7  c8  bias-> t1.xy                (018689c1)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c7.xyz   c8.xyz    op: 00020200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t0  t0  t0  bias-> t2.xy                (01880000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t1  c4  t0  bias-> t0.y                 (01000901)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c4.xxx   c4.xyz    op: 00010280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  5: xyz:  t1  t0  t0  bias-> t0.x                 (00800001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  6: xyz:  t0 c11  t0  bias-> t1.xy                (01840ac0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c11.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0 c10  t0  bias-> t3.xy                (018c0a80)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c10.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0  c9  t0  bias-> t0.xy                (01800a40)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c9.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 9, tex_offset: 0, alu_end: 23, tex_end: 2  (code_addr: 000405c9)
>  TEX:
>    TEX t1, t1, texture[3]   (00009841)
>    TEX t3, t3, texture[1]   (000088c3)
>    TEX t0, t0, texture[2]   (00009000)
>  9: xyz:  t3  c5  t0  bias-> t4.x                 (00900943)
>       w:  c1  t0  t0  bias-> t2.w                 (00880021)
>     xyz:   t3.xyz   c5.xyz   t3.xxx    op: 00804200 
>       w:     c1.w     t0.w      0.0    op: 00040509
> 10: xyz:  t1  c1  c6  bias-> t1.xyz               (03866841)
>       w:  t3  t0  t0  bias-> t1.w                 (00840003)
>     xyz:   t1.xyz   c1.xyz      0.0    op: 00050200 
>       w:     t3.w     c6.z      0.0    op: 00040409
> 11: xyz:  t0  t0  t0  bias-> t3.xyz               (038c0000)
>       w:  c5  t0  t0  bias->                      (00000025)
>     xyz:   t0.xyz   c5.www     -1.0    op: 000d4600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t2  t4  t0  bias-> t2.xyz               (03880102)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz: t32  t0  t0  bias-> t4.x                 (00900000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:  t32.xxx  t32.xxx  t32.xxx    op: 05004081 
>       w:     t1.w    t32.x    t32.x    op: 04000009
> 14: xyz:  t3  t4  t0  bias-> t4.x                 (00900103)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t4.x      1.0      1.0    op: 00044883
> 15: xyz:  t2  c3  c2   sub-> t5.xyz               (039628c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz   c2.xxx      0.0    op: 0025048f 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 16: xyz:  t2  t0  t0  bias-> t4.y                 (01100002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:  -t2.xyz  -t2.xyz   t2.xxx    op: 00805020 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 17: xyz:  t5  t4  t0  bias-> t4.y                 (01100105)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:   |t4.y|     t5.x     t5.x    op: 05800044
> 18: xyz:  t5  t4  t0  bias-> t4.x                 (00900105)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:   |t4.x|     t5.x     t5.x    op: 05800043
> 19: xyz:  t3 t32  t4  bias-> t3.xyz               (038c4003)
>       w:  t4  t0  t0  bias-> t4.w                 (00900004)
>     xyz:   t3.xyz   t4.www      0.0    op: 00050600 
>       w:   |t4.y|     t3.x     t3.x    op: 05800047
> 20: xyz:  t5  t0  t0  bias-> t5.xyz               (03940005)
>       w:  t4  t0  t0  bias->                      (00000004)
>     xyz:   t5.xyz   t4.www      0.0    op: 00050600 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 21: xyz:  t2 t32  t5  bias-> t2.xyz               (03885002)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:  -t2.xyz   t3.www   t5.xyz    op: 00020620 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 22: xyz:  t5  t3  t0  bias-> t4.y                 (011000c5)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t3.xyz   t5.xxx    op: 00804200 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 23: xyz:  t2  t4  t0  bias-> t4.y                 (01100102)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:     t4.y      0.0     t2.x    op: 01800804
> 24: xyz:  t4  t0  t0  bias-> t4.y                 (01100004)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xxx   t4.xxx   t4.xxx    op: 05004081 
>       w:   |t4.y|     t4.x     t4.x    op: 05800041
> 25: xyz:  t2  t4  t0  bias-> t2.xyz               (03880102)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t4.yyy      0.0    op: 00050300 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 26: xyz:  t2  t3  t0  bias-> t2.x                 (008800c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz   t2.xxx    op: 40804200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 27: xyz:  t2  t0  t0  bias->                      (00000002)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t2.xxx   t2.xxx   t2.xxx    op: 00004081 
>       w:     t2.x     t2.x     t2.x    op: 04800000
> 28: xyz: t32  t0  t0  bias->                      (00000000)
>       w:  t4  t1  t0  bias-> t1.w                 (00840044)
>     xyz:  t32.xxx  t32.xxx  t32.xxx    op: 00004081 
>       w:     t4.w     t1.w      0.0    op: 00040509
> 29: xyz:  t0  t0  t0  bias-> t2.x                 (00880000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 30: xyz: t32  t2  t0  bias-> t2.x                 (00880080)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   t2.xyz      0.0    op: 0005020c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
> 31: xyz:  t2  c1  t0  bias-> t2.xyz               (03880842)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xxx   c1.xyz      0.0    op: 00050201 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 32: xyz:  t1 t32  t2  bias-> t1.xyz               (03842001)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t1.xyz   t3.www   t2.xyz    op: 00020600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 2: alu_offset: 33, tex_offset: 3, alu_end: 1, tex_end: 0  (code_addr: 00403061)
>  TEX:
>    TEX t2, t4, texture[0]   (00008084)
> 33: xyz:  t1  t2  t0  bias-> t1.xyz               (03840081)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t2.xxx      0.0    op: 00050280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 34: xyz:  t1  t0  t0  bias-> o0.xyz               (1c000001)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], CONST[0].xxxx
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = const[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[0].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=17*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  c0  t0  bias-> t0.xyz               (03800800)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xyz   c0.xxx      0.0    op: 00050280 
>       w:     t0.w     c0.x      0.0    op: 00040189
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[5]
>DCL CONST[0..3]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[5].xxxx, CONST[5].yyyy
>  2: MOV TEMP[1].xyz, CONST[1].xyzx
>  3: TEX TEMP[2].xyz, TEMP[0].xyyy, SAMP[0], RECT
>  4: DP3 TEMP[2].x, TEMP[2].xyzz, IMM[0].xyzz
>  5: MAD TEMP[2].x, -TEMP[2].xxxx, CONST[3].xxxx, IN[1].zzzz
>  6: ADD TEMP[2].x, TEMP[2].xxxx, -CONST[2].xxxx
>  7: MIN TEMP[2].x, TEMP[2].xxxx, CONST[2].yyyy
>  8: RCP TEMP[3].x, CONST[2].yyyy
>  9: MUL TEMP[2].x, TEMP[2].xxxx, TEMP[3].xxxx
> 10: MAX TEMP[2].x, TEMP[2].xxxx, IMM[0].wwww
> 11: POW TEMP[2].x, TEMP[2].xxxx, CONST[0].xxxx
> 12: MUL TEMP[2].x, TEMP[2].xxxx, CONST[1].wwww
> 13: MOV TEMP[1].w, TEMP[2].xxxx
> 14: MOV OUT[0], TEMP[1]
> 15: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
>  9: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 10: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 11: RCP temp[3].x, const[2].yyyy;
> 12: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 13: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 14: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 16: MOV temp[1].w, temp[2].xxxx;
> 17: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
>  9: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 10: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 11: RCP temp[3].x, const[2].yyyy;
> 12: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 13: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 14: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 16: MOV temp[1].w, temp[2].xxxx;
> 17: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
>  9: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 10: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 11: RCP temp[3].x, const[2].yyyy;
> 12: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 13: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 14: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 16: MOV temp[1].w, temp[2].xxxx;
> 17: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
>  9: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 10: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 11: RCP temp[3].x, const[2].yyyy;
> 12: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 13: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 14: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 16: MOV temp[1].w, temp[2].xxxx;
> 17: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
>  9: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 10: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 11: RCP temp[3].x, const[2].yyyy;
> 12: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 13: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 14: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 16: MOV temp[1].w, temp[2].xxxx;
> 17: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: MUL temp[5], temp[0].xyyy, const[9];
>  7: TEX temp[6], temp[5], 2D[0];
>  8: MOV temp[2].xyz, temp[6];
>  9: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
> 10: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
> 11: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 12: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 13: RCP temp[3].x, const[2].yyyy;
> 14: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 15: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 16: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 17: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 18: MOV temp[1].w, temp[2].xxxx;
> 19: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: MUL temp[5], temp[0].xyyy, const[9];
>  7: TEX temp[6], temp[5], 2D[0];
>  8: MOV temp[2].xyz, temp[6];
>  9: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
> 10: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, input[1].zzzz;
> 11: ADD temp[2].x, temp[2].xxxx, -const[2].xxxx;
> 12: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 13: RCP temp[3].x, const[2].yyyy;
> 14: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 15: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 16: LG2 temp[7].w, temp[2].xxxx;
> 17: MUL temp[7].w, temp[7].wwww, const[0].xxxx;
> 18: EX2 temp[2].x, temp[7].wwww;
> 19: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 20: MOV temp[1].w, temp[2].xxxx;
> 21: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].___w;
>  1: MUL temp[4].xy, input[0].xy__, temp[4].ww__;
>  2: MAD temp[4].xy, temp[4].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[0].x, temp[4].x___;
>  4: MAD temp[0].y, temp[4]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[1].xyz, const[1].xyz_;
>  6: MUL temp[5].xy, temp[0].xy__, const[9].xy__;
>  7: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  8: MOV temp[2].xyz, temp[6].xyz_;
>  9: DP3 temp[2].x, temp[2].xyz_, const[6].xyz_;
> 10: MAD temp[2].x, -temp[2].x___, const[3].x___, input[1].z___;
> 11: ADD temp[2].x, temp[2].x___, -const[2].x___;
> 12: MIN temp[2].x, temp[2].x___, const[2].y___;
> 13: RCP temp[3].x, const[2].y___;
> 14: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 15: MAX temp[2].x, temp[2].x___, const[6].w___;
> 16: LG2 temp[7].w, temp[2].___x;
> 17: MUL temp[7].w, temp[7].___w, const[0].___x;
> 18: EX2 temp[2].x, temp[7].w___;
> 19: MUL temp[2].x, temp[2].x___, const[1].w___;
> 20: MOV temp[1].w, temp[2].___x;
> 21: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].___w;
>  1: MUL temp[4].xy, input[0].xy__, temp[4].ww__;
>  2: MAD temp[4].xy, temp[4].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[0].x, temp[4].x___;
>  4: MAD temp[0].y, temp[4]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[1].xyz, const[1].xyz_;
>  6: MUL temp[5].xy, temp[0].xy__, const[9].xy__;
>  7: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  8: MOV temp[2].xyz, temp[6].xyz_;
>  9: DP3 temp[2].x, temp[2].xyz_, const[6].xyz_;
> 10: MAD temp[2].x, -temp[2].x___, const[3].x___, input[1].z___;
> 11: ADD temp[2].x, temp[2].x___, -const[2].x___;
> 12: MIN temp[2].x, temp[2].x___, const[2].y___;
> 13: RCP temp[3].x, const[2].y___;
> 14: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 15: MAX temp[2].x, temp[2].x___, const[6].w___;
> 16: LG2 temp[7].w, temp[2].___x;
> 17: MUL temp[7].w, temp[7].___w, const[0].___x;
> 18: EX2 temp[2].x, temp[7].w___;
> 19: MUL temp[2].x, temp[2].x___, const[1].w___;
> 20: MOV temp[1].w, temp[2].___x;
> 21: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: MOV temp[15].xyz, temp[14].xyz_;
>  9: DP3 temp[16].x, temp[15].xyz_, const[6].xyz_;
> 10: MAD temp[17].x, -temp[16].x___, const[3].x___, input[1].z___;
> 11: ADD temp[18].x, temp[17].x___, -const[2].x___;
> 12: MIN temp[19].x, temp[18].x___, const[2].y___;
> 13: RCP temp[20].x, const[2].y___;
> 14: MUL temp[21].x, temp[19].x___, temp[20].x___;
> 15: MAX temp[22].x, temp[21].x___, const[6].w___;
> 16: LG2 temp[23].w, temp[22].___x;
> 17: MUL temp[24].w, temp[23].___w, const[0].___x;
> 18: EX2 temp[25].x, temp[24].w___;
> 19: MUL temp[26].x, temp[25].x___, const[1].w___;
> 20: MOV temp[12].w, temp[26].___x;
> 21: MOV output[0], temp[12];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[6].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, input[1].z___;
> 10: MIN temp[19].x, (temp[17] - const[2]).x___, const[2].y___;
> 11: RCP temp[20].x, const[2].y___;
> 12: MUL temp[21].x, temp[19].x___, temp[20].x___;
> 13: MAX temp[22].x, temp[21].x___, none.0___;
> 14: LG2 temp[23].w, temp[22].___x;
> 15: MUL temp[24].w, temp[23].___w, const[0].___x;
> 16: EX2 temp[25].x, temp[24].w___;
> 17: MUL temp[26].x, temp[25].x___, const[1].w___;
> 18: MOV temp[12].w, temp[26].___x;
> 19: MOV output[0], temp[12];
>CONST[6] = {     1.0000     0.0039     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[6].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, input[1].z___;
> 10: MIN temp[19].x, (temp[17] - const[2]).x___, const[2].y___;
> 11: RCP temp[20].x, const[2].y___;
> 12: MUL temp[21].x, temp[19].x___, temp[20].x___;
> 13: MAX temp[22].x, temp[21].x___, none.0___;
> 14: LG2 temp[23].w, temp[22].___x;
> 15: MUL temp[24].w, temp[23].___w, const[0].___x;
> 16: EX2 temp[25].x, temp[24].w___;
> 17: MUL temp[26].x, temp[25].x___, const[1].w___;
> 18: MOV temp[12].w, temp[26].___x;
> 19: MOV output[0], temp[12];
>CONST[5] = {     1.0000     0.0039     0.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[6].xy__, const[7].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[4]._x__, const[4]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[8].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[5].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, input[1].z___;
> 10: MIN temp[19].x, (temp[17] - const[2]).x___, const[2].y___;
> 11: RCP temp[20].x, const[2].y___;
> 12: MUL temp[21].x, temp[19].x___, temp[20].x___;
> 13: MAX temp[22].x, temp[21].x___, none.0___;
> 14: LG2 temp[23].w, temp[22].___x;
> 15: MUL temp[24].w, temp[23].___w, const[0].___x;
> 16: EX2 temp[25].x, temp[24].w___;
> 17: MUL temp[26].x, temp[25].x___, const[1].w___;
> 18: MOV temp[12].w, temp[26].___x;
> 19: MOV output[0], temp[12];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = const[1]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>  6: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: src0.xyz = temp[14], src1.xyz = const[5]
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = input[1]
>     MAD temp[17].x, -src0.x__, src1.x__, src2.z__
> 10: src0.xyz = const[2], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MIN temp[19].x, srcp.x__, src0.y__
> 11: src0.xyz = const[2]
>     REPL_ALPHA temp[20].x
>     RCP, src0.y
> 12: src0.xyz = temp[19], src1.xyz = temp[20]
>     MAD temp[21].x, src0.x__, src1.x__, src0.000
> 13: src0.xyz = temp[21]
>     MAX temp[22].x, src0.x__, src0.0__
> 14: src0.xyz = temp[22]
>     LG2 temp[23].w, src0.x
> 15: src0.xyz = const[0], src0.w = temp[23]
>     MAD temp[24].w, src0.w, src0.x, src0.0
> 16: src0.w = temp[24]
>     REPL_ALPHA temp[25].x
>     EX2, src0.w
> 17: src0.xyz = temp[25], src0.w = const[1]
>     MAD temp[26].x, src0.x__, src0.w__, src0.000
> 18: src0.xyz = temp[26]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 19: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[0]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[20].w, src1.y
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[14], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = input[1]
>     MAD temp[17].x, -src0.x__, src1.x__, src2.z__
> 10: src0.xyz = const[2], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MIN temp[19].x, srcp.x__, src0.y__
> 11: src0.xyz = temp[19], src0.w = temp[20], src1.xyz = temp[20]
>     MAD temp[21].x, src0.x__, src0.w__, src0.000
> 12: src0.xyz = temp[21]
>     MAX temp[22].x, src0.x__, src0.0__
> 13: src0.xyz = temp[22]
>     LG2 temp[23].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[23]
>     MAD temp[24].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[24]
>     REPL_ALPHA temp[25].x
>     EX2, src0.w
> 16: src0.xyz = temp[25], src0.w = const[1]
>     MAD temp[26].x, src0.x__, src0.w__, src0.000
> 17: src0.xyz = temp[26]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[0]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[20].w, src1.y
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[14], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = input[1]
>     MAD temp[17].x, -src0.x__, src1.x__, src2.z__
> 10: src0.xyz = const[2], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MIN temp[19].x, srcp.x__, src0.y__
> 11: src0.xyz = temp[19], src0.w = temp[20]
>     MAD temp[21].x, src0.x__, src0.w__, src0.000
> 12: src0.xyz = temp[21]
>     MAX temp[22].x, src0.x__, src0.0__
> 13: src0.xyz = temp[22]
>     LG2 temp[23].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[23]
>     MAD temp[24].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[24]
>     REPL_ALPHA temp[25].x
>     EX2, src0.w
> 16: src0.xyz = temp[25], src0.w = const[1]
>     MAD temp[26].x, src0.x__, src0.w__, src0.000
> 17: src0.xyz = temp[26]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[1]
>     MAD temp[2].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0], src1.xyz = const[2]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src0.00_
>     RCP temp[0].w, src1.y
>  2: src0.xyz = temp[0], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[1].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[0]
>     MAD temp[1].x, src0.x__, src0.11_, src0.00_
>  5: src0.xyz = temp[1], src1.xyz = const[8]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  6: BEGIN_TEX;
>  7: TEX temp[1].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[1], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[0].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[0], src1.xyz = const[3], src2.xyz = input[0]
>     MAD temp[0].x, -src0.x__, src1.x__, src2.z__
> 10: src0.xyz = const[2], src1.xyz = temp[0], srcp.xyz = (src1 - src0)
>     MIN temp[0].x, srcp.x__, src0.y__
> 11: src0.xyz = temp[0], src0.w = temp[0]
>     MAD temp[0].x, src0.x__, src0.w__, src0.0__
> 12: src0.xyz = temp[0]
>     MAX temp[0].x, src0.x__, src0.0__
> 13: src0.xyz = temp[0]
>     LG2 temp[0].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[0]
>     MAD temp[0].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[0]
>     REPL_ALPHA temp[0].x
>     EX2, src0.w
> 16: src0.xyz = temp[0], src0.w = const[1]
>     MAD temp[0].x, src0.x__, src0.w__, src0.0__
> 17: src0.xyz = temp[0]
>     MAD temp[0].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[2], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=18*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 5, tex_end: 0  (code_addr: 00000140)
>  0: xyz:  c1  t0  t0  bias-> t2.xyz               (03880021)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   c1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w     c1.x     c1.x    op: 05000009
>  1: xyz:  t1  c2  t0  bias-> t0.xy                (01800881)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     c2.y     t1.x     t1.x    op: 05000004
>  2: xyz:  t0  c6  c7  bias-> t0.xy                (01827980)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c6.xyz   c7.xyz    op: 00020200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  c4  t0  bias-> t1.y                 (01040900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xxx   c4.xyz    op: 00010280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t0  t0  t0  bias-> t1.x                 (00840000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  5: xyz:  t1  c8  t0  bias-> t0.xy                (01800a01)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c8.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 1: alu_offset: 6, tex_offset: 0, alu_end: 10, tex_end: 0  (code_addr: 00400286)
>  TEX:
>    TEX t1, t0, texture[0]   (00008040)
>  6: xyz:  t1  c5  t0  bias-> t0.x                 (00800941)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c5.xyz   t1.xxx    op: 00804200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  7: xyz:  t0  c3  t0  bias-> t0.x                 (008008c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:  -t0.xyz   c3.xyz   t0.zzz    op: 8002c220 NOP
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  c2  t0  t0   sub-> t0.x                 (00800022)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz   c2.yyy   c2.xxx    op: 0220410f 
>       w:     c2.x     c2.x     c2.x    op: 00000000
>  9: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 10: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      0.0   t0.xxx    op: 02804a00 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x     t0.x     t0.x    op: 04800000
> 12: xyz:  c0  t0  t0  bias->                      (00000020)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c0.xxx   c0.xxx   c0.xxx    op: 00004081 
>       w:     t0.w     c0.x      0.0    op: 00040009
> 13: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t0.w     t0.x     t0.x    op: 04000009
> 14: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t0.xyz   c1.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x      1.0      0.0    op: 00040880
> 16: xyz:  t2  t0  t0  bias-> o0.xyz               (1c000002)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0..1]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[1]
>  2: MOV TEMP[1].w, TEMP[0].wwww
>  3: MUL TEMP[2].x, CONST[1].xxxx, CONST[0].xxxx
>  4: MUL TEMP[1].xyz, TEMP[0].xyzz, TEMP[2].xxxx
>  5: MOV OUT[0], TEMP[1]
>  6: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].wwww;
>  3: MUL temp[2].x, const[1].xxxx, const[0].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, temp[2].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].___w;
>  3: MUL temp[2].x, const[1].x___, const[0].x___;
>  4: MUL temp[1].xyz, temp[0].xyz_, temp[2].xxx_;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV temp[1].w, temp[0].___w;
>  3: MUL temp[2].x, const[1].x___, const[0].x___;
>  4: MUL temp[1].xyz, temp[0].xyz_, temp[2].xxx_;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[3], input[0].xy__, 2D[0];
>  1: MUL temp[4], temp[3], input[1];
>  2: MOV temp[5].w, temp[4].___w;
>  3: MUL temp[6].x, const[1].x___, const[0].x___;
>  4: MUL temp[5].xyz, temp[4].xyz_, temp[6].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[3], input[0].xy__, 2D[0];
>  1: MUL temp[4], temp[3], input[1];
>  2: MOV temp[5].w, temp[4].___w;
>  3: MUL temp[6].x, const[1].x___, const[0].x___;
>  4: MUL temp[5].xyz, temp[4].xyz_, temp[6].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[3], input[0].xy__, 2D[0];
>  1: MUL temp[4], temp[3], input[1];
>  2: MOV temp[5].w, temp[4].___w;
>  3: MUL temp[6].x, const[1].x___, const[0].x___;
>  4: MUL temp[5].xyz, temp[4].xyz_, temp[6].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[3], input[0].xy__, 2D[0];
>  1: MUL temp[4], temp[3], input[1];
>  2: MOV temp[5].w, temp[4].___w;
>  3: MUL temp[6].x, const[1].x___, const[0].x___;
>  4: MUL temp[5].xyz, temp[4].xyz_, temp[6].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[3], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[1], src1.w = input[1]
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  2: src0.w = temp[4]
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  3: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[6].x, src0.x__, src1.x__, src0.000
>  4: src0.xyz = temp[4], src1.xyz = temp[6]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[3], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  3: src0.xyz = const[1], src0.w = temp[4], src1.xyz = const[0]
>     MAD temp[6].x, src0.x__, src1.x__, src0.000
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[4], src1.xyz = temp[6]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[3], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  3: src0.xyz = const[1], src0.w = temp[4], src1.xyz = const[0]
>     MAD temp[6].x, src0.x__, src1.x__, src0.000
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[4], src1.xyz = temp[6]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = const[1], src0.w = temp[0], src1.xyz = const[0]
>     MAD temp[1].x, src0.x__, src1.x__, src0.0__
>     MAD temp[0].w, src0.w, src0.1, src0.0
>  4: src0.xyz = temp[0], src1.xyz = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
>  5: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=19*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 3, tex_end: 0  (code_addr: 004000c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t0.xyz   t1.xyz      0.0    op: 00050200 
>       w:     t0.w     t1.w      0.0    op: 00040509
>  1: xyz:  c1  c0  t0  bias-> t1.x                 (00840821)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c1.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t0.w      1.0      0.0    op: 00040889
>  2: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], LINEAR
>DCL OUT[0], COLOR
>DCL SAMP[0]
>  0: TEX OUT[0], IN[0], SAMP[0], RECT
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX output[0], input[0], RECT[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX output[0], input[0], RECT[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX output[0], input[0], RECT[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX output[0], input[0], RECT[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX output[0], input[0], RECT[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[0], input[0], const[0];
>  1: TEX temp[1], temp[0], 2D[0];
>  2: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[0], input[0], const[0];
>  1: TEX temp[1], temp[0], 2D[0];
>  2: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[0].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[1], temp[0].xy__, 2D[0];
>  2: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[0].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[1], temp[0].xy__, 2D[0];
>  2: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[2].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[3], temp[2].xy__, 2D[0];
>  2: MOV output[0], temp[3];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[2].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[3], temp[2].xy__, 2D[0];
>  2: MOV output[0], temp[3];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[2].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[3], temp[2].xy__, 2D[0];
>  2: MOV output[0], temp[3];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[2].xy, input[0].xy__, const[0].xy__;
>  1: TEX temp[3], temp[2].xy__, 2D[0];
>  2: MOV output[0], temp[3];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>  1: TEX temp[3], temp[2].xy__, 2D[0];
>  2: src0.xyz = temp[3], src0.w = temp[3]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>  1: BEGIN_TEX;
>  2: TEX temp[3], temp[2].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = temp[3] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>  1: BEGIN_TEX;
>  2: TEX temp[3], temp[2].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = temp[3] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  1: BEGIN_TEX;
>  2: TEX temp[0], temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=20*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00000000)
>  0: xyz:  t0  c0  t0  bias-> t0.xy                (01800800)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 1, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400001)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL IN[2], GENERIC[12], PERSPECTIVE
>DCL IN[3], GENERIC[13], PERSPECTIVE
>DCL IN[4], GENERIC[14], PERSPECTIVE
>DCL IN[5], GENERIC[15], PERSPECTIVE
>DCL IN[6], GENERIC[16], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[13]
>DCL CONST[0..8]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    1.0000,     2.0000,    -1.0000,    10.0000}
>IMM[1] FLT32 {    0.0000,    -1.0000,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[13].xxxx, CONST[13].yyyy
>  2: TEX TEMP[1], IN[1].xyyy, SAMP[3], 2D
>  3: MUL TEMP[1], TEMP[1], IN[3]
>  4: MOV TEMP[2].w, TEMP[1].wwww
>  5: ADD TEMP[3].xyz, IMM[0].xxxx, -TEMP[1].xyzz
>  6: ADD TEMP[4].x, IMM[0].xxxx, -CONST[8].xxxx
>  7: MAD TEMP[2].xyz, TEMP[3].xyzz, TEMP[4].xxxx, TEMP[1].xyzz
>  8: TEX TEMP[1].xyz, IN[1].xyyy, SAMP[2], 2D
>  9: MAD TEMP[1].xyz, TEMP[1].xyzz, IMM[0].yyyy, IMM[0].zzzz
> 10: MUL TEMP[3].xyz, TEMP[1].xxxx, IN[5].xyzz
> 11: MAD TEMP[3].xyz, TEMP[1].yyyy, IN[4].xyzz, TEMP[3].xyzz
> 12: MAD TEMP[1].xyz, TEMP[1].zzzz, IN[6].xyzz, TEMP[3].xyzz
> 13: DP3 TEMP[3].x, TEMP[1].xyzz, TEMP[1].xyzz
> 14: RSQ TEMP[3].x, TEMP[3].xxxx
> 15: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[3].xxxx
> 16: RCP TEMP[3].x, IN[2].zzzz
> 17: MIN TEMP[3].x, TEMP[3].xxxx, IMM[0].wwww
> 18: MUL TEMP[3].x, CONST[6].xxxx, TEMP[3].xxxx
> 19: MAD TEMP[3].xy, TEMP[1].xyyy, TEMP[3].xxxx, TEMP[0].xyyy
> 20: TEX TEMP[3].xyz, TEMP[3].xyyy, SAMP[1], RECT
> 21: MUL TEMP[2].xyz, TEMP[2].xyzz, TEMP[3].xyzz
> 22: DP3 TEMP[3].x, IN[2].xyzz, IN[2].xyzz
> 23: RSQ TEMP[3].x, TEMP[3].xxxx
> 24: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[3].xxxx
> 25: DP3 TEMP[5].x, TEMP[1].xyzz, TEMP[3].xyzz
> 26: MUL TEMP[5].xyz, TEMP[5].xxxx, TEMP[1].xyzz
> 27: MUL TEMP[5].xyz, IMM[0].yyyy, TEMP[5].xyzz
> 28: ADD TEMP[4].xyz, TEMP[3].xyzz, -TEMP[5].xyzz
> 29: MUL TEMP[5], CONST[2], TEMP[4].xxxx
> 30: MAD TEMP[5], CONST[3], TEMP[4].yyyy, TEMP[5]
> 31: MAD TEMP[4], CONST[4], TEMP[4].zzzz, TEMP[5]
> 32: ADD TEMP[4].xyz, TEMP[4], CONST[5]
> 33: TEX TEMP[4].xyz, TEMP[4].xyzz, SAMP[0], CUBE
> 34: ADD TEMP[5].x, IMM[0].xxxx, -CONST[1].xxxx
> 35: DP3 TEMP[3].x, -TEMP[3].xyzz, TEMP[1].xyzz
> 36: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 37: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[3].xxxx
> 38: POW TEMP[3].x, TEMP[3].xxxx, CONST[1].yyyy
> 39: MAD TEMP[3].x, TEMP[5].xxxx, TEMP[3].xxxx, CONST[1].xxxx
> 40: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 41: MUL TEMP[3].xyz, TEMP[4].xyzz, TEMP[3].xxxx
> 42: MUL TEMP[3].xyz, TEMP[3].xyzz, CONST[8].xxxx
> 43: MAD TEMP[2].xyz, TEMP[3].xyzz, CONST[7].xxxx, TEMP[2].xyzz
> 44: DP3 TEMP[1].x, TEMP[1].xyzz, IMM[1].xxyy
> 45: ABS TEMP[1].x, TEMP[1].xxxx
> 46: ADD TEMP[1].x, IMM[0].xxxx, -TEMP[1].xxxx
> 47: POW TEMP[1].x, TEMP[1].xxxx, CONST[0].yyyy
> 48: MUL TEMP[1].x, TEMP[1].xxxx, CONST[0].xxxx
> 49: MUL TEMP[1].xyz, TEMP[4].xyzz, TEMP[1].xxxx
> 50: MUL TEMP[1].xyz, TEMP[1].xyzz, CONST[8].xxxx
> 51: MAD TEMP[2].xyz, TEMP[1].xyzz, CONST[7].xxxx, TEMP[2].xyzz
> 52: MOV OUT[0], TEMP[2]
> 53: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[7], input[1].xyyy, 2D[2];
> 12: MOV temp[1].xyz, temp[7];
> 13: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 14: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 15: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 16: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 17: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 18: RSQ temp[3].x, temp[3].xxxx;
> 19: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 20: RCP temp[3].x, input[2].zzzz;
> 21: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 22: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 23: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 24: MUL temp[8], temp[3].xyyy, const[18];
> 25: TEX temp[9], temp[8], 2D[1];
> 26: MOV temp[3].xyz, temp[9];
> 27: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 28: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 29: RSQ temp[3].x, temp[3].xxxx;
> 30: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 31: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 32: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 33: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 34: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 35: MUL temp[5], const[2], temp[4].xxxx;
> 36: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 37: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 38: ADD temp[4].xyz, temp[4], const[5];
> 39: TEX temp[10], temp[4].xyzz, CUBE[0];
> 40: MOV temp[4].xyz, temp[10];
> 41: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 42: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 45: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 46: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 47: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 48: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 49: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 50: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 51: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 52: ABS temp[1].x, temp[1].xxxx;
> 53: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 54: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 55: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 56: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 57: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 58: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 59: MOV output[0], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[7], input[1].xyyy, 2D[2];
> 12: MOV temp[1].xyz, temp[7];
> 13: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 14: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 15: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 16: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 17: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 18: RSQ temp[3].x, |temp[3].xxxx|;
> 19: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 20: RCP temp[3].x, input[2].zzzz;
> 21: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 22: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 23: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 24: MUL temp[8], temp[3].xyyy, const[18];
> 25: TEX temp[9], temp[8], 2D[1];
> 26: MOV temp[3].xyz, temp[9];
> 27: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 28: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 29: RSQ temp[3].x, |temp[3].xxxx|;
> 30: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 31: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 32: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 33: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 34: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 35: MUL temp[5], const[2], temp[4].xxxx;
> 36: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 37: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 38: ADD temp[4].xyz, temp[4], const[5];
> 39: TEX temp[10], temp[4].xyzz, CUBE[0];
> 40: MOV temp[4].xyz, temp[10];
> 41: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 42: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 45: LG2 temp[11].w, temp[3].xxxx;
> 46: MUL temp[11].w, temp[11].wwww, const[1].yyyy;
> 47: EX2 temp[3].x, temp[11].wwww;
> 48: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 50: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 52: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 53: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 54: MOV temp[1].x, |temp[1].xxxx|;
> 55: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 56: LG2 temp[12].w, temp[1].xxxx;
> 57: MUL temp[12].w, temp[12].wwww, const[0].yyyy;
> 58: EX2 temp[1].x, temp[12].wwww;
> 59: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 60: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 61: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 62: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[6].ww__;
>  2: MAD temp[6].xy, temp[6].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[0].x, temp[6].x___;
>  4: MAD temp[0].y, temp[6]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].___w;
>  8: ADD temp[3].xyz, const[14].xxx_, -temp[1].xyz_;
>  9: ADD temp[4].x, const[14].x___, -const[8].x___;
> 10: MAD temp[2].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 11: TEX temp[7].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[1].xyz, temp[7].xyz_;
> 13: MAD temp[1].xyz, temp[1].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[3].xyz, temp[1].xxx_, input[5].xyz_;
> 15: MAD temp[3].xyz, temp[1].yyy_, input[4].xyz_, temp[3].xyz_;
> 16: MAD temp[1].xyz, temp[1].zzz_, input[6].xyz_, temp[3].xyz_;
> 17: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 18: RSQ temp[3].x, |temp[3].x___|;
> 19: MUL temp[1].xyz, temp[1].xyz_, temp[3].xxx_;
> 20: RCP temp[3].x, input[2].z___;
> 21: MIN temp[3].x, temp[3].x___, const[14].w___;
> 22: MUL temp[3].x, const[6].x___, temp[3].x___;
> 23: MAD temp[3].xy, temp[1].xy__, temp[3].xx__, temp[0].xy__;
> 24: MUL temp[8].xy, temp[3].xy__, const[18].xy__;
> 25: TEX temp[9].xyz, temp[8].xy__, 2D[1];
> 26: MOV temp[3].xyz, temp[9].xyz_;
> 27: MUL temp[2].xyz, temp[2].xyz_, temp[3].xyz_;
> 28: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[3].x, |temp[3].x___|;
> 30: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 31: DP3 temp[5].x, temp[1].xyz_, temp[3].xyz_;
> 32: MUL temp[5].xyz, temp[5].xxx_, temp[1].xyz_;
> 33: MUL temp[5].xyz, const[14].yyy_, temp[5].xyz_;
> 34: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 35: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 36: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 37: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 38: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 39: TEX temp[10].xyz, temp[4].xyz_, CUBE[0];
> 40: MOV temp[4].xyz, temp[10].xyz_;
> 41: ADD temp[5].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[3].x, -temp[3].xyz_, temp[1].xyz_;
> 43: MAX temp[3].x, temp[3].x___, const[15].x___;
> 44: ADD temp[3].x, const[14].x___, -temp[3].x___;
> 45: LG2 temp[11].w, temp[3].___x;
> 46: MUL temp[11].w, temp[11].___w, const[1].___y;
> 47: EX2 temp[3].x, temp[11].w___;
> 48: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 49: MAX temp[3].x, temp[3].x___, const[15].x___;
> 50: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 51: MUL temp[3].xyz, temp[3].xyz_, const[8].xxx_;
> 52: MAD temp[2].xyz, temp[3].xyz_, const[7].xxx_, temp[2].xyz_;
> 53: DP3 temp[1].x, temp[1].xyz_, const[15].xxy_;
> 54: MOV temp[1].x, |temp[1].x___|;
> 55: ADD temp[1].x, const[14].x___, -temp[1].x___;
> 56: LG2 temp[12].w, temp[1].___x;
> 57: MUL temp[12].w, temp[12].___w, const[0].___y;
> 58: EX2 temp[1].x, temp[12].w___;
> 59: MUL temp[1].x, temp[1].x___, const[0].x___;
> 60: MUL temp[1].xyz, temp[4].xyz_, temp[1].xxx_;
> 61: MUL temp[1].xyz, temp[1].xyz_, const[8].xxx_;
> 62: MAD temp[2].xyz, temp[1].xyz_, const[7].xxx_, temp[2].xyz_;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[6].ww__;
>  2: MAD temp[6].xy, temp[6].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[0].x, temp[6].x___;
>  4: MAD temp[0].y, temp[6]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].___w;
>  8: ADD temp[3].xyz, const[14].xxx_, -temp[1].xyz_;
>  9: ADD temp[4].x, const[14].x___, -const[8].x___;
> 10: MAD temp[2].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 11: TEX temp[7].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[1].xyz, temp[7].xyz_;
> 13: MAD temp[1].xyz, temp[1].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[3].xyz, temp[1].xxx_, input[5].xyz_;
> 15: MAD temp[3].xyz, temp[1].yyy_, input[4].xyz_, temp[3].xyz_;
> 16: MAD temp[1].xyz, temp[1].zzz_, input[6].xyz_, temp[3].xyz_;
> 17: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 18: RSQ temp[3].x, |temp[3].x___|;
> 19: MUL temp[1].xyz, temp[1].xyz_, temp[3].xxx_;
> 20: RCP temp[3].x, input[2].z___;
> 21: MIN temp[3].x, temp[3].x___, const[14].w___;
> 22: MUL temp[3].x, const[6].x___, temp[3].x___;
> 23: MAD temp[3].xy, temp[1].xy__, temp[3].xx__, temp[0].xy__;
> 24: MUL temp[8].xy, temp[3].xy__, const[18].xy__;
> 25: TEX temp[9].xyz, temp[8].xy__, 2D[1];
> 26: MOV temp[3].xyz, temp[9].xyz_;
> 27: MUL temp[2].xyz, temp[2].xyz_, temp[3].xyz_;
> 28: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[3].x, |temp[3].x___|;
> 30: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 31: DP3 temp[5].x, temp[1].xyz_, temp[3].xyz_;
> 32: MUL temp[5].xyz, temp[5].xxx_, temp[1].xyz_;
> 33: MUL temp[5].xyz, const[14].yyy_, temp[5].xyz_;
> 34: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 35: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 36: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 37: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 38: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 39: TEX temp[10].xyz, temp[4].xyz_, CUBE[0];
> 40: MOV temp[4].xyz, temp[10].xyz_;
> 41: ADD temp[5].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[3].x, -temp[3].xyz_, temp[1].xyz_;
> 43: MAX temp[3].x, temp[3].x___, const[15].x___;
> 44: ADD temp[3].x, const[14].x___, -temp[3].x___;
> 45: LG2 temp[11].w, temp[3].___x;
> 46: MUL temp[11].w, temp[11].___w, const[1].___y;
> 47: EX2 temp[3].x, temp[11].w___;
> 48: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 49: MAX temp[3].x, temp[3].x___, const[15].x___;
> 50: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 51: MUL temp[3].xyz, temp[3].xyz_, const[8].xxx_;
> 52: MAD temp[2].xyz, temp[3].xyz_, const[7].xxx_, temp[2].xyz_;
> 53: DP3 temp[1].x, temp[1].xyz_, const[15].xxy_;
> 54: MOV temp[1].x, |temp[1].x___|;
> 55: ADD temp[1].x, const[14].x___, -temp[1].x___;
> 56: LG2 temp[12].w, temp[1].___x;
> 57: MUL temp[12].w, temp[12].___w, const[0].___y;
> 58: EX2 temp[1].x, temp[12].w___;
> 59: MUL temp[1].x, temp[1].x___, const[0].x___;
> 60: MUL temp[1].xyz, temp[4].xyz_, temp[1].xxx_;
> 61: MUL temp[1].xyz, temp[1].xyz_, const[8].xxx_;
> 62: MAD temp[2].xyz, temp[1].xyz_, const[7].xxx_, temp[2].xyz_;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, const[14].xxx_, -temp[18].xyz_;
>  9: ADD temp[21].x, const[14].x___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[24].xyz, temp[23].xyz_;
> 13: MAD temp[25].xyz, temp[24].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 15: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 16: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 17: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 18: RSQ temp[30].x, |temp[29].x___|;
> 19: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 20: RCP temp[32].x, input[2].z___;
> 21: MIN temp[33].x, temp[32].x___, const[14].w___;
> 22: MUL temp[34].x, const[6].x___, temp[33].x___;
> 23: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 24: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 25: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 26: MOV temp[38].xyz, temp[37].xyz_;
> 27: MUL temp[39].xyz, temp[22].xyz_, temp[38].xyz_;
> 28: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[41].x, |temp[40].x___|;
> 30: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 31: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 32: MUL temp[44].xyz, temp[43].xxx_, temp[31].xyz_;
> 33: MUL temp[45].xyz, const[14].yyy_, temp[44].xyz_;
> 34: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 35: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 36: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 37: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 38: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 39: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 40: MOV temp[52].xyz, temp[51].xyz_;
> 41: ADD temp[53].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 43: MAX temp[55].x, temp[54].x___, const[15].x___;
> 44: ADD temp[56].x, const[14].x___, -temp[55].x___;
> 45: LG2 temp[57].w, temp[56].___x;
> 46: MUL temp[58].w, temp[57].___w, const[1].___y;
> 47: EX2 temp[59].x, temp[58].w___;
> 48: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 49: MAX temp[61].x, temp[60].x___, const[15].x___;
> 50: MUL temp[62].xyz, temp[52].xyz_, temp[61].xxx_;
> 51: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 52: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 53: DP3 temp[65].x, temp[31].xyz_, const[15].xxy_;
> 54: MOV temp[66].x, |temp[65].x___|;
> 55: ADD temp[67].x, const[14].x___, -temp[66].x___;
> 56: LG2 temp[68].w, temp[67].___x;
> 57: MUL temp[69].w, temp[68].___w, const[0].___y;
> 58: EX2 temp[70].x, temp[69].w___;
> 59: MUL temp[71].x, temp[70].x___, const[0].x___;
> 60: MUL temp[72].xyz, temp[52].xyz_, temp[71].xxx_;
> 61: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 62: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 63: MOV output[0], temp[19];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[14].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[14].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, none.00-1_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>CONST[14] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[15] = {     0.0000    -1.0000     0.0000     0.0000 }
>CONST[19] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[14].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[14].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, const[19].xyz_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>CONST[10] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[14] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[9]._x__, const[9]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[10].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[10].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[13].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, const[14].xyz_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3]
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
>  7: src0.w = temp[18]
>     MAD temp[19].w, src0.w, src0.1, src0.0
>  8: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
>  9: src0.xyz = const[8]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
> 10: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: src0.xyz = temp[23], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
> 13: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 14: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 15: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 16: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 17: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 18: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 19: src0.xyz = input[2]
>     REPL_ALPHA temp[32].x
>     RCP, src0.z
> 20: src0.xyz = temp[32], src0.w = const[10]
>     MIN temp[33].x, src0.x__, src0.w__
> 21: src0.xyz = const[6], src1.xyz = temp[33]
>     MAD temp[34].x, src0.x__, src1.x__, src0.000
> 22: src0.xyz = temp[31], src1.xyz = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src1.xx_, src2.xy_
> 23: src0.xyz = temp[35], src1.xyz = const[13]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: src0.xyz = temp[22], src1.xyz = temp[37]
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 26: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
> 27: src0.xyz = temp[40]
>     REPL_ALPHA temp[41].x
>     RSQ, |src0.x|
> 28: src0.xyz = input[2], src1.xyz = temp[41]
>     MAD temp[42].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[31], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[43], src1.xyz = temp[31]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
> 31: src0.xyz = temp[42], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
> 32: src0.xyz = const[2], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
> 33: src0.xyz = const[3], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
> 34: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 35: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: src0.xyz = const[1]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
> 38: src0.xyz = temp[42], src1.xyz = temp[31]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
> 39: src0.xyz = temp[54]
>     MAX temp[55].x, src0.x__, src0.0__
> 40: src0.xyz = temp[55]
>     MAD temp[56].x, src0.1__, src0.111, -src0.x__
> 41: src0.xyz = temp[56]
>     LG2 temp[57].w, src0.x
> 42: src0.xyz = const[1], src0.w = temp[57]
>     MAD temp[58].w, src0.w, src0.y, src0.0
> 43: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 44: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].x, src0.x__, src1.x__, src2.x__
> 45: src0.xyz = temp[60]
>     MAX temp[61].x, src0.x__, src0.0__
> 46: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 47: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 48: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 49: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 50: src0.xyz = temp[65]
>     MAD temp[67].x, src0.1__, |src0.111|, -|src0.x__|
> 51: src0.xyz = temp[67]
>     LG2 temp[68].w, src0.x
> 52: src0.xyz = const[0], src0.w = temp[68]
>     MAD temp[69].w, src0.w, src0.y, src0.0
> 53: src0.w = temp[69]
>     REPL_ALPHA temp[70].x
>     EX2, src0.w
> 54: src0.xyz = temp[70], src1.xyz = const[0]
>     MAD temp[71].x, src0.x__, src1.x__, src0.000
> 55: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 56: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 57: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 58: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
>     RCP temp[32].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[40]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
>     RSQ temp[41].w, |src1.x|
>  2: src0.xyz = input[2], src0.w = temp[41], src1.xyz = temp[41], src1.w = temp[32], src2.w = const[10]
>     MAD temp[42].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[33].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[0]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
>     RCP temp[13].w, src0.w
>  4: src0.xyz = input[0], src0.w = temp[13], src1.xyz = const[6], src1.w = temp[33]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[34].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[17], input[1].xy__, 2D[3];
> 10: TEX temp[23].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[23], src0.w = temp[18], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[19].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 16: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[31], src0.w = temp[34], src1.xyz = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[42], src1.xyz = temp[31], src2.xyz = temp[65]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
>     MAD temp[67].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[31], src0.w = temp[67], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
>     LG2 temp[68].w, src0.w
> 25: src0.xyz = temp[35], src1.xyz = const[13], src2.xyz = temp[54]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
>     MAX temp[55].w, src2.x, src0.0
> 26: src0.xyz = temp[43], src0.w = temp[68], src1.xyz = temp[31], src2.xyz = const[0]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[69].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[42], src0.w = temp[55], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[56].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[56], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[57].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[69], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[70].w, src0.w
> 30: src0.xyz = temp[70], src0.w = temp[70], src1.xyz = const[0], src1.w = temp[57], src2.xyz = const[1]
>     MAD temp[71].x, src0.w__, src1.x__, src0.000
>     MAD temp[58].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 33: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].w, src0.x, src1.x, src2.x
> 35: src0.xyz = temp[60], src0.w = temp[60]
>     MAX temp[61].x, src0.w__, src0.0__
> 36: BEGIN_TEX;
> 37: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 38: TEX temp[51].xyz, temp[50].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[22], src1.xyz = temp[37] SEM_WAIT
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 42: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
>     RCP temp[32].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[40]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
>     RSQ temp[41].w, |src1.x|
>  2: src0.xyz = input[2], src0.w = temp[41], src1.w = temp[32], src2.w = const[10]
>     MAD temp[42].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[33].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[0]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
>     RCP temp[13].w, src0.w
>  4: src0.xyz = input[0], src0.w = temp[13], src1.xyz = const[6], src1.w = temp[33]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[34].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[17], input[1].xy__, 2D[3];
> 10: TEX temp[23].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[23], src0.w = temp[18], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[19].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 16: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[31], src0.w = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[42], src1.xyz = temp[31], src2.xyz = temp[65]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
>     MAD temp[67].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[31], src0.w = temp[67], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
>     LG2 temp[68].w, src0.w
> 25: src0.xyz = temp[35], src1.xyz = const[13], src2.xyz = temp[54]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
>     MAX temp[55].w, src2.x, src0.0
> 26: src0.xyz = temp[43], src0.w = temp[68], src1.xyz = temp[31], src2.xyz = const[0]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[69].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[42], src0.w = temp[55], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[56].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[56], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[57].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[69], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[70].w, src0.w
> 30: src0.w = temp[70], src1.xyz = const[0], src1.w = temp[57], src2.xyz = const[1]
>     MAD temp[71].x, src0.w__, src1.x__, src0.000
>     MAD temp[58].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 33: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].w, src0.x, src1.x, src2.x
> 35: src0.w = temp[60]
>     MAX temp[61].x, src0.w__, src0.0__
> 36: BEGIN_TEX;
> 37: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 38: TEX temp[51].xyz, temp[50].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[22], src1.xyz = temp[37] SEM_WAIT
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 42: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[1]
>     DP3 temp[0].z, src0.xyz, src0.xyz
>     RCP temp[0].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[0]
>     MAD temp[0].z, src0.__1, src0.__1, -src0.__x
>     RSQ temp[1].w, |src1.z|
>  2: src0.xyz = input[1], src0.w = temp[1], src1.w = temp[0], src2.w = const[10]
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[0].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[6]
>     MAD temp[6].z, src0.__1, src0.__1, -src0.__x
>     RCP temp[1].w, src0.w
>  4: src0.xyz = input[6], src0.w = temp[1], src1.xyz = const[6], src1.w = temp[0]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.00_
>     MAD temp[0].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[6], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[6].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[6], src1.xyz = const[9]
>     MAD temp[7].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[6]
>     MAD temp[7].x, src0.x__, src0.11_, src0.00_
>  8: BEGIN_TEX;
>  9: TEX temp[8], input[0].xy__, 2D[3];
> 10: TEX temp[9].xyz, input[0].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[8], src0.w = temp[8], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[9], src0.w = temp[1], src1.xyz = const[10]
>     MAD temp[8].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[1].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[2]
>     MAD temp[9].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[8], src1.xyz = input[4]
>     MAD temp[4].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[9], src1.xyz = temp[6], src2.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src1.zzz, src2.xyz
> 16: src0.xyz = temp[8], src1.xyz = input[3], src2.xyz = temp[4]
>     MAD temp[3].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[8], src1.xyz = input[5], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[3]
>     DP3 temp[0].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[0]
>     REPL_ALPHA temp[0].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[3], src1.xyz = temp[0]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[3], src0.w = temp[0], src2.xyz = temp[7]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[3], src1.xyz = const[14]
>     DP3 temp[4].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[1], src1.xyz = temp[3], src2.xyz = temp[4]
>     DP3 temp[4].x, -src0.xyz, src1.xyz
>     MAD temp[0].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[3], src0.w = temp[0], src1.xyz = temp[1]
>     DP3 temp[4].y, src0.xyz, src1.xyz
>     LG2 temp[0].w, src0.w
> 25: src0.xyz = temp[0], src1.xyz = const[13], src2.xyz = temp[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>     MAX temp[2].w, src2.x, src0.0
> 26: src0.xyz = temp[4], src0.w = temp[0], src1.xyz = temp[3], src2.xyz = const[0]
>     MAD temp[3].xyz * 2, src0.yyy, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[1], src0.w = temp[2], src1.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[2].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[2], src1.xyz = temp[1]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[2].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[0], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[0].w, src0.w
> 30: src0.w = temp[0], src1.xyz = const[0], src1.w = temp[2], src2.xyz = const[1]
>     MAD temp[4].x, src0.w__, src1.x__, src0.0__
>     MAD temp[0].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[0]
>     REPL_ALPHA temp[3].x
>     EX2, src0.w
> 33: src0.xyz = temp[1], src1.xyz = const[5]
>     MAD temp[1].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[0], src1.xyz = temp[3], src2.xyz = const[1]
>     MAD temp[0].w, src0.z, src1.x, src2.x
> 35: src0.w = temp[0]
>     MAX temp[0].z, src0.__w, src0.__0
> 36: BEGIN_TEX;
> 37: TEX temp[3].xyz, temp[0].xy__, 2D[1];
> 38: TEX temp[5].xyz, temp[1].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[2], src1.xyz = temp[3] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[5], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[5], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 42: src0.xyz = temp[2], src1.xyz = const[8]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[0], src1.xyz = const[8]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[0], src1.xyz = const[7], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[2], src1.xyz = const[7], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=21*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 7, tex_end: 0  (code_addr: 000001c0)
>  0: xyz:  t1  t0  t0  bias-> t0.z                 (02000001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t1.xyz   t1.xxx    op: 00804000 
>       w:     t1.z     t1.x     t1.x    op: 05000002
>  1: xyz:  c1  t0  t0  bias-> t0.z                 (02000021)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:      1.0      1.0  -c1.xxx    op: 00084a95 
>       w:   |t0.z|     c1.x     c1.x    op: 05800045
>  2: xyz:  t1 t32  t0  bias-> t1.xyz               (03840001)
>       w:  t1  t0 c10  bias-> t0.w                 (0082a001)
>     xyz:   t1.xyz   t1.www      0.0    op: 00050600 
>       w:     t0.w    c10.w     t1.x    op: 0100058a
>  3: xyz:  c8  t0  t0  bias-> t6.z                 (02180028)
>       w:  t6  t0  t0  bias-> t1.w                 (00840006)
>     xyz:      1.0      1.0  -c8.xxx    op: 00084a95 
>       w:     t6.w     c8.x     c8.x    op: 05000009
>  4: xyz:  t6  c6  t0  bias-> t6.xy                (01980986)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t6.xyz   t1.www      0.0    op: 00050600 
>       w:     c6.x     t0.w      0.0    op: 00040503
>  5: xyz:  t6 c11 c12  bias-> t6.xy                (019acac6)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz  c11.xyz  c12.xyz    op: 00020200 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  6: xyz:  t6  c9  t0  bias-> t7.y                 (011c0a46)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz   c9.xxx   c9.xyz    op: 00010280 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  7: xyz:  t6  t0  t0  bias-> t7.x                 (009c0006)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz      1.0      0.0    op: 00050a80 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>NODE 1: alu_offset: 8, tex_offset: 0, alu_end: 24, tex_end: 1  (code_addr: 00020608)
>  TEX:
>    TEX t8, t0, texture[3]   (00009a00)
>    TEX t9, t0, texture[2]   (00009240)
>  8: xyz:  t8  t2  t0  bias-> t2.xyz               (03880088)
>       w:  t8  t2  t0  bias-> t1.w                 (00840088)
>     xyz:   t8.xyz   t2.xyz      0.0    op: 00050200 
>       w:     t8.w     t2.w      0.0    op: 00040509
>  9: xyz:  t9 c10  t0  bias-> t8.xyz               (03a00a89)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t9.xyz  c10.yyy     -1.0    op: 000d4300 
>       w:     t1.w      1.0      0.0    op: 00040889
> 10: xyz:  t2  t0  t0  bias-> t9.xyz               (03a40002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t2.xyz    op: 00080a95 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 11: xyz:  t8  t4  t0  bias-> t4.xyz               (03900108)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.xxx   t4.xyz      0.0    op: 00050201 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 12: xyz:  t9  t6  t2  bias-> t2.xyz               (03882189)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t9.xyz   t6.zzz   t2.xyz    op: 00020380 
>       w:     t9.x     t9.x     t9.x    op: 00000000
> 13: xyz:  t8  t3  t4  bias-> t3.xyz               (038c40c8)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.yyy   t3.xyz   t4.xyz    op: 00020202 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 14: xyz:  t8  t5  t3  bias-> t3.xyz               (038c3148)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.zzz   t5.xyz   t3.xyz    op: 00020203 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 15: xyz:  t3  t0  t0  bias-> t0.x                 (00800003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 16: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:   |t0.x|     t0.x     t0.x    op: 05800040
> 17: xyz:  t3  t0  t0  bias-> t3.xyz               (038c0003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t0.xxx      0.0    op: 00050280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 18: xyz:  t3 t32  t7  bias-> t0.xy                (01807003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t0.www   t7.xyz    op: 00020600 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 19: xyz:  t3 c14  t0  bias-> t4.x                 (00900b83)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz  c14.xyz   t3.xxx    op: 00804200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 20: xyz:  t1  t3  t4  bias-> t4.x                 (009040c1)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:  -t1.xyz   t3.xyz   t1.xxx    op: 00804220 
>       w:      1.0    |1.0|  -|t4.x|    op: 0019a891
> 21: xyz:  t3  t1  t0  bias-> t4.y                 (01100043)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t3.xyz   t1.xyz   t3.xxx    op: 00804200 
>       w:     t0.w     t3.x     t3.x    op: 04800009
> 22: xyz:  t0 c13  t4  bias-> t0.xy                (01804b40)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz  c13.xyz      0.0    op: 00050200 
>       w:     t4.x      0.0     t0.x    op: 01800806
> 23: xyz:  t4  t3  c0  bias-> t3.xyz               (038e00c4)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t4.yyy   t3.xyz      0.0    op: 08050202 
>       w:     t0.w     c0.y      0.0    op: 00040389
> 24: xyz:  t1  t3  t0  bias-> t1.xyz               (038400c1)
>       w:  t2  t0  t0  bias-> t2.w                 (00880002)
>     xyz:   t1.xyz      1.0  -t3.xyz    op: 00090a80 
>       w:      1.0      1.0    -t2.w    op: 000a4891
> 25: xyz:  c2  t1  t0  bias-> t3.xyz               (038c0062)
>       w:  t2  t0  t0  bias-> t2.w                 (00880002)
>     xyz:   c2.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t2.w     c2.x     c2.x    op: 04800009
> 26: xyz:  c3  t1  t3  bias-> t3.xyz               (038c3063)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c3.xyz   t1.yyy   t3.xyz    op: 00020300 
>       w:     t0.w     c3.x     c3.x    op: 04000009
> 27: xyz: t32  c0  c1  bias-> t4.x                 (00921800)
>       w:  t0  t2  t0  bias-> t0.w                 (00800080)
>     xyz:   t0.www   c0.xyz      0.0    op: 0005020c 
>       w:     t2.w     c1.y      0.0    op: 0004038a
> 28: xyz:  c4  t1  t3  bias-> t1.xyz               (03843064)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c4.xyz   t1.zzz   t3.xyz    op: 00020380 
>       w:     c4.x     c4.x     c4.x    op: 00000000
> 29: xyz:  t0  t0  t0  bias-> t3.x                 (008c0000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t0.w     t0.x     t0.x    op: 04000009
> 30: xyz:  t1  c5  t0  bias-> t1.xyz               (03840941)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0   c5.xyz    op: 00010a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 31: xyz:  t0  t3  c1  bias->                      (000210c0)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.z     t3.x     c1.x    op: 00018182
> 32: xyz: t32  t0  t0  bias-> t0.z                 (02000000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.www      0.0  t32.xxx    op: 02804a0c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
>NODE 2: alu_offset: 33, tex_offset: 2, alu_end: 7, tex_end: 1  (code_addr: 004221e1)
>  TEX:
>    TEX t3, t0, texture[1]   (000088c0)
>    TEX t5, t1, texture[0]   (00008141)
> 33: xyz:  t2  t3  t0  bias-> t1.xyz               (038400c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 34: xyz:  t5  t4  t0  bias-> t2.xyz               (03880105)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 35: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t0.zzz      0.0    op: 00050380 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 36: xyz:  t2  c8  t0  bias-> t2.xyz               (03880a02)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c8.xxx      0.0    op: 00050280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 37: xyz:  t0  c8  t0  bias-> t0.xyz               (03800a00)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c8.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 38: xyz:  t0  c7  t1  bias-> t0.xyz               (038019c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c7.xxx   t1.xyz    op: 00020280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 39: xyz:  t2  c7  t0  bias-> t0.xyz               (038009c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c7.xxx   t0.xyz    op: 00020280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 40: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t1  t0  t0  bias-> o0.w                 (01000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0..1]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[1]
>  2: MUL TEMP[1].x, TEMP[0].wwww, CONST[1].xxxx
>  3: MOV TEMP[1].w, TEMP[1].xxxx
>  4: MUL TEMP[1].xyz, TEMP[0].xyzz, CONST[0].xxxx
>  5: MOV OUT[0], TEMP[1]
>  6: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].wwww, const[1].xxxx;
>  3: MOV temp[1].w, temp[1].xxxx;
>  4: MUL temp[1].xyz, temp[0].xyzz, const[0].xxxx;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].w___, const[1].x___;
>  3: MOV temp[1].w, temp[1].___x;
>  4: MUL temp[1].xyz, temp[0].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MUL temp[1].x, temp[0].w___, const[1].x___;
>  3: MOV temp[1].w, temp[1].___x;
>  4: MUL temp[1].xyz, temp[0].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[2], input[0].xy__, 2D[0];
>  1: MUL temp[3], temp[2], input[1];
>  2: MUL temp[4].x, temp[3].w___, const[1].x___;
>  3: MOV temp[5].w, temp[4].___x;
>  4: MUL temp[5].xyz, temp[3].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[2], input[0].xy__, 2D[0];
>  1: MUL temp[3], temp[2], input[1];
>  2: MUL temp[4].x, temp[3].w___, const[1].x___;
>  3: MOV temp[5].w, temp[4].___x;
>  4: MUL temp[5].xyz, temp[3].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[2], input[0].xy__, 2D[0];
>  1: MUL temp[3], temp[2], input[1];
>  2: MUL temp[4].x, temp[3].w___, const[1].x___;
>  3: MOV temp[5].w, temp[4].___x;
>  4: MUL temp[5].xyz, temp[3].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[2], input[0].xy__, 2D[0];
>  1: MUL temp[3], temp[2], input[1];
>  2: MUL temp[4].x, temp[3].w___, const[1].x___;
>  3: MOV temp[5].w, temp[4].___x;
>  4: MUL temp[5].xyz, temp[3].xyz_, const[0].xxx_;
>  5: MOV output[0], temp[5];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[2], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[2], src0.w = temp[2], src1.xyz = input[1], src1.w = input[1]
>     MAD temp[3].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[3].w, src0.w, src1.w, src0.0
>  2: src0.xyz = const[1], src0.w = temp[3]
>     MAD temp[4].x, src0.w__, src0.x__, src0.000
>  3: src0.xyz = temp[4]
>     MAD temp[5].w, src0.x, src0.1, src0.0
>  4: src0.xyz = temp[3], src1.xyz = const[0]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[2], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[2], src0.w = temp[2], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[3].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[3].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = const[0], src2.xyz = const[1]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[4].w, src0.w, src2.x, src0.0
>  4: src0.xyz = temp[4], src0.w = temp[4]
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[2], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[2], src0.w = temp[2], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[3].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[3].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = const[0], src2.xyz = const[1]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[4].w, src0.w, src2.x, src0.0
>  4: src0.w = temp[4]
>     MAD temp[5].w, src0.w, src0.1, src0.0
>  5: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = const[0], src2.xyz = const[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[0].w, src0.w, src2.x, src0.0
>  4: src0.w = temp[0]
>     MAD temp[0].w, src0.w, src0.1, src0.0
>  5: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=22*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 3, tex_end: 0  (code_addr: 004000c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t0.xyz   t1.xyz      0.0    op: 00050200 
>       w:     t0.w     t1.w      0.0    op: 00040509
>  1: xyz:  t0  c0  c1  bias-> t0.xyz               (03821800)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xyz   c0.xxx      0.0    op: 00050280 
>       w:     t0.w     c1.x      0.0    op: 00040309
>  2: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.w      1.0      0.0    op: 00040889
>  3: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], COLOR, COLOR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TXP TEMP[0], IN[1].xyyw, SAMP[0], RECT
>  1: MUL TEMP[0], TEMP[0], IN[0]
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TXP temp[0], input[1].xyyw, RECT[0];
>  1: MUL temp[0], temp[0], input[0];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[1], input[1].xyyw, const[0];
>  1: TXP temp[0], temp[1], 2D[0];
>  2: MUL temp[0], temp[0], input[0];
>  3: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[1], input[1].xyyw, const[0];
>  1: TXP temp[0], temp[1], 2D[0];
>  2: MUL temp[0], temp[0], input[0];
>  3: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[1].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[0], temp[1].xy_w, 2D[0];
>  2: MUL temp[0], temp[0], input[0];
>  3: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[1].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[0], temp[1].xy_w, 2D[0];
>  2: MUL temp[0], temp[0], input[0];
>  3: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3], temp[2].xy_w, 2D[0];
>  2: MUL temp[4], temp[3], input[0];
>  3: MOV output[0], temp[4];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3], temp[2].xy_w, 2D[0];
>  2: MUL temp[4], temp[3], input[0];
>  3: MOV output[0], temp[4];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3], temp[2].xy_w, 2D[0];
>  2: MUL temp[4], temp[3], input[0];
>  3: MOV output[0], temp[4];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[2].xyw, input[1].xy_w, const[0].xy_w;
>  1: TXP temp[3], temp[2].xy_w, 2D[0];
>  2: MUL temp[4], temp[3], input[0];
>  3: MOV output[0], temp[4];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: TXP temp[3], temp[2].xy_w, 2D[0];
>  2: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[0], src1.w = input[0]
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[4], src0.w = temp[4]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[3], temp[2].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[4], src0.w = temp[4]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[3], temp[2].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[4].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[4].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[4], src0.w = temp[4]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src0.w = input[1], src1.xyz = const[0], src1.w = const[0]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>     MAD temp[1].w, src0.w, src1.w, src0.0
>  1: BEGIN_TEX;
>  2: TXP temp[1], temp[1].xy_w, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[0], src1.w = input[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=23*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00000000)
>  0: xyz:  t1  c0  t0  bias-> t1.xy                (01840801)
>       w:  t1  c0  t0  bias-> t1.w                 (00840801)
>     xyz:   t1.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t1.w     c0.w      0.0    op: 00040509
>NODE 1: alu_offset: 1, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400041)
>  TEX:
>    TXP t1, t1, texture[0]   (00018041)
>  1: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t1.xyz   t0.xyz      0.0    op: 00050200 
>       w:     t1.w     t0.w      0.0    op: 00040509
>  2: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxx
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].1111;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.1111;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.111, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=24*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      1.0      1.0      0.0    op: 00050a95 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MOV OUT[0], TEMP[0]
>  2: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=25*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL IN[3], GENERIC[14], PERSPECTIVE
>DCL IN[4], GENERIC[15], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {   -0.5000,     0.5000,   255.0000,     0.0039}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[2], 2D
>  1: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[1], 2D
>  2: ADD TEMP[2].xyz, TEMP[1].xyzz, IMM[0].xxxx
>  3: MUL TEMP[3].xyz, TEMP[2].xxxx, IN[3].xyzz
>  4: MAD TEMP[3].xyz, TEMP[2].yyyy, IN[2].xyzz, TEMP[3].xyzz
>  5: MAD TEMP[3].xyz, TEMP[2].zzzz, IN[4].xyzz, TEMP[3].xyzz
>  6: DP3 TEMP[4].x, TEMP[3].xyzz, TEMP[3].xyzz
>  7: RSQ TEMP[4].x, TEMP[4].xxxx
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].yyyy, IMM[0].yyyy
> 10: MUL TEMP[3].x, IN[1].xxxx, IMM[0].zzzz
> 11: FLR TEMP[4].x, TEMP[3].xxxx
> 12: MOV TEMP[5].x, TEMP[4].xxxx
> 13: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 14: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 15: FLR TEMP[4].x, TEMP[3].xxxx
> 16: MOV TEMP[5].y, TEMP[4].xxxx
> 17: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 18: MOV TEMP[5].z, TEMP[3].xxxx
> 19: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].wwww
> 20: MOV TEMP[2].xyz, TEMP[5].xyzx
> 21: TEX TEMP[3].xy, IN[0].xyyy, SAMP[0], 2D
> 22: MOV TEMP[1].w, TEMP[3].xxxx
> 23: MOV TEMP[2].w, TEMP[3].yyyy
> 24: MOV OUT[0], TEMP[0]
> 25: MOV OUT[1], TEMP[1]
> 26: MOV OUT[2], TEMP[2]
> 27: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, temp[4].xxxx;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FLR temp[4].x, temp[3].xxxx;
> 13: MOV temp[5].x, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 16: FLR temp[4].x, temp[3].xxxx;
> 17: MOV temp[5].y, temp[4].xxxx;
> 18: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 19: MOV temp[5].z, temp[3].xxxx;
> 20: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 21: MOV temp[2].xyz, temp[5].xyzx;
> 22: TEX temp[7], input[0].xyyy, 2D[0];
> 23: MOV temp[3].xy, temp[7];
> 24: MOV temp[1].w, temp[3].xxxx;
> 25: MOV temp[2].w, temp[3].yyyy;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, |temp[4].xxxx|;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FRC temp[4].x, temp[3].xxxx;
> 13: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 14: MOV temp[5].x, temp[4].xxxx;
> 15: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 16: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 17: FRC temp[4].x, temp[3].xxxx;
> 18: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 19: MOV temp[5].y, temp[4].xxxx;
> 20: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 21: MOV temp[5].z, temp[3].xxxx;
> 22: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 23: MOV temp[2].xyz, temp[5].xyzx;
> 24: TEX temp[7], input[0].xyyy, 2D[0];
> 25: MOV temp[3].xy, temp[7];
> 26: MOV temp[1].w, temp[3].xxxx;
> 27: MOV temp[2].w, temp[3].yyyy;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[10].xyz, temp[9].xyz_;
>  3: ADD temp[11].xyz, temp[10].xyz_, const[0].xxx_;
>  4: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  5: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  6: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  7: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  8: RSQ temp[16].x, |temp[15].x___|;
>  9: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
> 10: MAD temp[18].xyz, temp[17].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[19].x, input[1].x___, const[0].z___;
> 12: FRC temp[20].x, temp[19].x___;
> 13: ADD temp[21].x, temp[19].x___, -temp[20].x___;
> 14: MOV temp[22].x, temp[21].x___;
> 15: ADD temp[23].x, temp[19].x___, -temp[21].x___;
> 16: MUL temp[24].x, temp[23].x___, const[0].z___;
> 17: FRC temp[25].x, temp[24].x___;
> 18: ADD temp[26].x, temp[24].x___, -temp[25].x___;
> 19: MOV temp[22].y, temp[26]._x__;
> 20: ADD temp[27].x, temp[24].x___, -temp[26].x___;
> 21: MOV temp[28].z, temp[27].__x_;
> 22: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 23: MOV temp[29].xyz, temp[28].xyz_;
> 24: TEX temp[30].xy, input[0].xy__, 2D[0];
> 25: MOV temp[31].xy, temp[30].xy__;
> 26: MOV temp[18].w, temp[31].___x;
> 27: MOV temp[29].w, temp[31].___y;
> 28: MOV output[0], temp[8];
> 29: MOV output[1], temp[18];
> 30: MOV output[2], temp[29];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: src0.xyz = temp[9]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>  3: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  4: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
>  5: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
>  6: src0.xyz = temp[14]
>     DP3 temp[15].x, src0.xyz, src0.xyz
>  7: src0.xyz = temp[15]
>     REPL_ALPHA temp[16].x
>     RSQ, |src0.x|
>  8: src0.xyz = temp[14], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src1.xxx, src0.000
>  9: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 10: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
> 11: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
> 12: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
> 13: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].x, src1.x__, src0.111, -srcp.x__
> 14: src0.xyz = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.x__, src1.z__, src0.000
> 15: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 16: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
> 17: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[27].x, src1.x__, src0.111, -srcp.x__
> 18: src0.xyz = temp[27]
>     MAD temp[28].z, src0.__x, src0.111, src0.000
> 19: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: src0.xyz = temp[30]
>     MAD temp[18].w, src0.x, src0.1, src0.0
> 23: src0.xyz = temp[30]
>     MAD temp[29].w, src0.y, src0.1, src0.0
> 24: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 25: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 26: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.xyz = temp[23], src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.xyz = temp[27], src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[5], input[0].xy__, 2D[2];
>  2: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[0].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[0] SEM_WAIT
>     MAD temp[1].x, src0.x__, src1.z__, src0.0__
>     MAD temp[1].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[6], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[0].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[1]
>     FRC temp[6].x, src0.x__
>  7: src0.xyz = temp[0], src1.xyz = input[3]
>     MAD temp[3].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].x, srcp.x__, src0.11_, src0.00_
>  9: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[0], src1.xyz = input[2], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.z__, src0.0__
> 12: src0.xyz = temp[0], src1.xyz = input[4], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[2]
>     FRC temp[1].x, src0.x__
> 14: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[0], srcp.xyz = (src1 - src0)
>     DP3 temp[1].y, src2.xyz, src2.xyz
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].y, srcp._x_, src0.11_, src0.00_
>     RSQ temp[3].w, |src2.y|
> 16: src0.w = temp[2]
>     MAD temp[1].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[0], src0.w = temp[3]
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[7], src0.w = const[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[1]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=26*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 19, tex_end: 2  (code_addr: 004404c0)
>  TEX:
>    TEX t5, t0, texture[2]   (00009140)
>    TEX t6, t0, texture[1]   (00008980)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t0.x      1.0      0.0    op: 00040886
>  1: xyz:  t6  t0  t0  bias-> t0.xyz               (03800006)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t6.xyz      1.0     -0.5    op: 000d8a80 
>       w:     t0.y      1.0      0.0    op: 00040884
>  2: xyz:  t1  t0  t0  bias-> t6.x                 (00980001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 04804080 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t0  t3  t0  bias-> t3.xyz               (038c00c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t3.xyz      0.0    op: 00050201 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t6  t1  t0   sub-> t7.x                 (009c0046)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  5: xyz:  t6  t1  t0   sub->                      (00000046)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t6.xxx   t6.xxx   t6.xxx    op: 00204081 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  t2  t3  bias-> t1.xyz               (03843080)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.yyy   t2.xyz   t3.xyz    op: 00020202 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.zzz      0.0    op: 0005038c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0  t4  t1  bias-> t0.xyz               (03801100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.zzz   t4.xyz   t1.xyz    op: 00020203 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  9: xyz:  t2  t0  t0  bias-> t1.x                 (00840002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 10: xyz:  t1  t2  t0   sub-> t1.y                 (01040081)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz   t0.xyz   t1.xxx    op: 00a04408 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 11: xyz:  t1  t2  t1   sub-> t7.y                 (011c1081)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:   |t1.y|     t1.x     t1.x    op: 05800047
> 12: xyz:  t0  t0  t0  bias-> t1.z                 (02040000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t0.xyz   t3.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 14: xyz:  t7  t0  t0  bias-> t1.xy                (01840007)
>       w:  c0  t0  t0  bias->                      (00000020)
>     xyz:   t7.xyz   c0.www      0.0    op: 00050600 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      0.5      0.5    op: 00058b00 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t1  t0  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 17: xyz:  t5  t0  t0  bias-> o0.xyz               (1c000005)
>       w:  t5  t0  t0  bias-> o0.w                 (01000005)
>     xyz:   t5.xyz      1.0      0.0    op: 00050a80 
>       w:     t5.w      1.0      0.0    op: 00040889
> 18: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t1  t0  t0  bias-> o1.w                 (03000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
> 19: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t0  t0  t0  bias-> o2.w                 (05000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    0.5000,   255.0000,     0.0039,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: DP3 TEMP[1].x, IN[2].xyzz, IN[2].xyzz
>  2: RSQ TEMP[2].x, TEMP[1].xxxx
>  3: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[2].xxxx
>  4: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].xxxx, IMM[0].xxxx
>  5: MUL TEMP[3].x, IN[1].xxxx, IMM[0].yyyy
>  6: FLR TEMP[4].x, TEMP[3].xxxx
>  7: MOV TEMP[5].x, TEMP[4].xxxx
>  8: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
>  9: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].yyyy
> 10: FLR TEMP[4].x, TEMP[3].xxxx
> 11: MOV TEMP[5].y, TEMP[4].xxxx
> 12: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 13: MOV TEMP[5].z, TEMP[3].xxxx
> 14: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].zzzz
> 15: MOV TEMP[2].xyz, TEMP[5].xyzx
> 16: MOV TEMP[1].w, IMM[0].wwww
> 17: MOV TEMP[2].w, IMM[0].wwww
> 18: MOV OUT[0], TEMP[0]
> 19: MOV OUT[1], TEMP[1]
> 20: MOV OUT[2], TEMP[2]
> 21: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, |temp[1].xxxx|;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FRC temp[4].x, temp[3].xxxx;
>  7: ADD temp[4].x, temp[3].xxxx, -temp[4];
>  8: MOV temp[5].x, temp[4].xxxx;
>  9: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 10: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 11: FRC temp[4].x, temp[3].xxxx;
> 12: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 13: MOV temp[5].y, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MOV temp[5].z, temp[3].xxxx;
> 16: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 17: MOV temp[2].xyz, temp[5].xyzx;
> 18: MOV temp[1].w, const[0].wwww;
> 19: MOV temp[2].w, const[0].wwww;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: ADD temp[13].x, temp[11].x___, -temp[12].x___;
>  8: MOV temp[14].x, temp[13].x___;
>  9: ADD temp[15].x, temp[11].x___, -temp[13].x___;
> 10: MUL temp[16].x, temp[15].x___, const[0].y___;
> 11: FRC temp[17].x, temp[16].x___;
> 12: ADD temp[18].x, temp[16].x___, -temp[17].x___;
> 13: MOV temp[14].y, temp[18]._x__;
> 14: ADD temp[19].x, temp[16].x___, -temp[18].x___;
> 15: MOV temp[20].z, temp[19].__x_;
> 16: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 17: MOV temp[21].xyz, temp[20].xyz_;
> 18: MOV temp[10].w, const[0].___w;
> 19: MOV temp[21].w, const[0].___w;
> 20: MOV output[0], temp[6];
> 21: MOV output[1], temp[10];
> 22: MOV output[2], temp[21];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>  2: src0.xyz = temp[7]
>     REPL_ALPHA temp[8].x
>     RSQ, |src0.x|
>  3: src0.xyz = input[2], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src1.xxx, src0.000
>  4: src0.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.HHH, src0.HHH
>  5: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>  6: src0.xyz = temp[11]
>     FRC temp[12].x, src0.x__
>  7: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  8: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, src1.x__, src0.111, -srcp.x__
>  9: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.x__, src1.y__, src0.000
> 10: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].x, src1.x__, src0.111, -srcp.x__
> 13: src0.xyz = temp[19]
>     MAD temp[20].z, src0.__x, src0.111, src0.000
> 14: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 15: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 16: 
>     MAD temp[10].w, src0.0, src0.1, src0.0
> 17: 
>     MAD temp[21].w, src0.0, src0.1, src0.0
> 18: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 19: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.xyz = temp[15], src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.xyz = temp[19], src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[1].x, src0.x__, src1.y__, src0.0__
>     MAD temp[1].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[1].y, src0.xyz, src0.xyz
>     MAD temp[2].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[1], src1.xyz = temp[1]
>     FRC temp[3].x, src0.x__
>     RSQ temp[3].w, |src1.y|
>  5: src0.xyz = temp[3], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].x, srcp.x__, src0.11_, src0.00_
>  6: src0.xyz = input[2], src0.w = temp[3]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[3], src1.xyz = temp[1], src2.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[1].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[3], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.y__, src0.0__
>  9: src0.xyz = temp[2]
>     FRC temp[3].x, src0.x__
> 10: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[4].y, srcp._x_, src0.11_, src0.00_
> 11: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[4], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[3]
>     MAD temp[2].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[1], src0.w = temp[2]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[2], src0.w = temp[1]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=27*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 15, tex_end: 0  (code_addr: 004003c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.yyy      0.0    op: 00050300 
>       w:      0.0      1.0      0.0    op: 00040890
>  1: xyz:  t2  t0  t0  bias-> t1.y                 (01040002)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:      0.0      1.0      0.0    op: 00040890
>  2: xyz:  t1  t1  t0  bias-> t3.x                 (008c0041)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:   |t1.y|     t1.x     t1.x    op: 05800044
>  3: xyz:  t3  t1  t0   sub-> t4.x                 (00900043)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  4: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t2.xyz   t3.www      0.0    op: 00050600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  5: xyz:  t3  t1  t2   sub-> t1.xyz               (03842043)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t2.xyz      0.5      0.5    op: 00258b08 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www   c0.yyy      0.0    op: 0005030c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t2  t0  t0  bias-> t3.x                 (008c0002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  8: xyz:  t3  t2  t0   sub-> t4.y                 (01100083)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  9: xyz:  t3  t2  t0   sub->                      (00000083)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t3.xxx   t3.xxx   t3.xxx    op: 00204081 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 10: xyz:  t4  c0  t0  bias-> t2.xy                (01880804)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> t2.z                 (02080000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
> 14: xyz:  t1  t0  t0  bias-> o1.xyz               (3c000001)
>       w:  t2  t0  t0  bias-> o1.w                 (03000002)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.w      1.0      0.0    op: 00040889
> 15: xyz:  t2  t0  t0  bias-> o2.xyz               (5c000002)
>       w:  t1  t0  t0  bias-> o2.w                 (05000001)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL IN[3], GENERIC[14], PERSPECTIVE
>DCL IN[4], GENERIC[15], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {   -0.5000,     0.5000,   255.0000,     0.0039}
>IMM[1] FLT32 {    0.0000,     0.0000,     0.0000,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[1], 2D
>  1: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[0], 2D
>  2: ADD TEMP[2].xyz, TEMP[1].xyzz, IMM[0].xxxx
>  3: MUL TEMP[3].xyz, TEMP[2].xxxx, IN[3].xyzz
>  4: MAD TEMP[3].xyz, TEMP[2].yyyy, IN[2].xyzz, TEMP[3].xyzz
>  5: MAD TEMP[3].xyz, TEMP[2].zzzz, IN[4].xyzz, TEMP[3].xyzz
>  6: DP3 TEMP[4].x, TEMP[3].xyzz, TEMP[3].xyzz
>  7: RSQ TEMP[4].x, TEMP[4].xxxx
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].yyyy, IMM[0].yyyy
> 10: MUL TEMP[3].x, IN[1].xxxx, IMM[0].zzzz
> 11: FLR TEMP[4].x, TEMP[3].xxxx
> 12: MOV TEMP[5].x, TEMP[4].xxxx
> 13: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 14: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 15: FLR TEMP[4].x, TEMP[3].xxxx
> 16: MOV TEMP[5].y, TEMP[4].xxxx
> 17: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 18: MOV TEMP[5].z, TEMP[3].xxxx
> 19: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].wwww
> 20: MOV TEMP[2].xyz, TEMP[5].xyzx
> 21: MOV TEMP[1].w, IMM[1].xxxx
> 22: MOV TEMP[2].w, IMM[1].xxxx
> 23: MOV OUT[0], TEMP[0]
> 24: MOV OUT[1], TEMP[1]
> 25: MOV OUT[2], TEMP[2]
> 26: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[0];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: MOV temp[1].w, temp[0].0000;
> 22: MOV temp[2].w, temp[0].0000;
> 23: MOV output[0], temp[0];
> 24: MOV output[1], temp[1];
> 25: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[0];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: MOV temp[1].w, temp[0].0000;
> 22: MOV temp[2].w, temp[0].0000;
> 23: MOV output[0], temp[0];
> 24: MOV output[1], temp[1];
> 25: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[0];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: MOV temp[1].w, temp[0].0000;
> 22: MOV temp[2].w, temp[0].0000;
> 23: MOV output[0], temp[0];
> 24: MOV output[1], temp[1];
> 25: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[0];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: MOV temp[1].w, temp[0].0000;
> 22: MOV temp[2].w, temp[0].0000;
> 23: MOV output[0], temp[0];
> 24: MOV output[1], temp[1];
> 25: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[0];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: MOV temp[1].w, temp[0].0000;
> 22: MOV temp[2].w, temp[0].0000;
> 23: MOV output[0], temp[0];
> 24: MOV output[1], temp[1];
> 25: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[6], input[0].xyyy, 2D[0];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, temp[4].xxxx;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FLR temp[4].x, temp[3].xxxx;
> 13: MOV temp[5].x, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 16: FLR temp[4].x, temp[3].xxxx;
> 17: MOV temp[5].y, temp[4].xxxx;
> 18: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 19: MOV temp[5].z, temp[3].xxxx;
> 20: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 21: MOV temp[2].xyz, temp[5].xyzx;
> 22: MOV temp[1].w, temp[0].0000;
> 23: MOV temp[2].w, temp[0].0000;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: TEX temp[6], input[0].xyyy, 2D[0];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, |temp[4].xxxx|;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FRC temp[4].x, temp[3].xxxx;
> 13: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 14: MOV temp[5].x, temp[4].xxxx;
> 15: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 16: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 17: FRC temp[4].x, temp[3].xxxx;
> 18: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 19: MOV temp[5].y, temp[4].xxxx;
> 20: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 21: MOV temp[5].z, temp[3].xxxx;
> 22: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 23: MOV temp[2].xyz, temp[5].xyzx;
> 24: MOV temp[1].w, temp[0].0000;
> 25: MOV temp[2].w, temp[0].0000;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[0];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: MOV temp[1].w, temp[0].___0;
> 25: MOV temp[2].w, temp[0].___0;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[0];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: MOV temp[1].w, temp[0].___0;
> 25: MOV temp[2].w, temp[0].___0;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: TEX temp[8].xyz, input[0].xy__, 2D[0];
>  2: MOV temp[9].xyz, temp[8].xyz_;
>  3: ADD temp[10].xyz, temp[9].xyz_, const[0].xxx_;
>  4: MUL temp[11].xyz, temp[10].xxx_, input[3].xyz_;
>  5: MAD temp[12].xyz, temp[10].yyy_, input[2].xyz_, temp[11].xyz_;
>  6: MAD temp[13].xyz, temp[10].zzz_, input[4].xyz_, temp[12].xyz_;
>  7: DP3 temp[14].x, temp[13].xyz_, temp[13].xyz_;
>  8: RSQ temp[15].x, |temp[14].x___|;
>  9: MUL temp[16].xyz, temp[13].xyz_, temp[15].xxx_;
> 10: MAD temp[17].xyz, temp[16].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[18].x, input[1].x___, const[0].z___;
> 12: FRC temp[19].x, temp[18].x___;
> 13: ADD temp[20].x, temp[18].x___, -temp[19].x___;
> 14: MOV temp[21].x, temp[20].x___;
> 15: ADD temp[22].x, temp[18].x___, -temp[20].x___;
> 16: MUL temp[23].x, temp[22].x___, const[0].z___;
> 17: FRC temp[24].x, temp[23].x___;
> 18: ADD temp[25].x, temp[23].x___, -temp[24].x___;
> 19: MOV temp[21].y, temp[25]._x__;
> 20: ADD temp[26].x, temp[23].x___, -temp[25].x___;
> 21: MOV temp[27].z, temp[26].__x_;
> 22: MUL temp[27].xy, temp[21].xy__, const[0].ww__;
> 23: MOV temp[28].xyz, temp[27].xyz_;
> 24: MOV temp[17].w, temp[0].___0;
> 25: MOV temp[28].w, temp[0].___0;
> 26: MOV output[0], temp[7];
> 27: MOV output[1], temp[17];
> 28: MOV output[2], temp[28];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: TEX temp[8].xyz, input[0].xy__, 2D[0];
>  2: ADD temp[10].xyz, temp[8].xyz_, none.-H-H-H_;
>  3: MUL temp[11].xyz, temp[10].xxx_, input[3].xyz_;
>  4: MAD temp[12].xyz, temp[10].yyy_, input[2].xyz_, temp[11].xyz_;
>  5: MAD temp[13].xyz, temp[10].zzz_, input[4].xyz_, temp[12].xyz_;
>  6: DP3 temp[14].x, temp[13].xyz_, temp[13].xyz_;
>  7: RSQ temp[15].x, |temp[14].x___|;
>  8: MUL temp[16].xyz, temp[13].xyz_, temp[15].xxx_;
>  9: MAD temp[17].xyz, temp[16].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[18].x, input[1].x___, const[0].z___;
> 11: FRC temp[19].x, temp[18].x___;
> 12: MOV temp[21].x, (temp[18] - temp[19]).x___;
> 13: ADD temp[22].x, temp[18].x___, -(temp[18] - temp[19]).x___;
> 14: MUL temp[23].x, temp[22].x___, const[0].z___;
> 15: FRC temp[24].x, temp[23].x___;
> 16: MOV temp[21].y, (temp[23] - temp[24])._x__;
> 17: ADD temp[26].x, temp[23].x___, -(temp[23] - temp[24]).x___;
> 18: MOV temp[27].z, temp[26].__x_;
> 19: MUL temp[27].xy, temp[21].xy__, const[0].ww__;
> 20: MOV temp[28].xyz, temp[27].xyz_;
> 21: MOV temp[17].w, none.___0;
> 22: MOV temp[28].w, none.___0;
> 23: MOV output[0], temp[7];
> 24: MOV output[1], temp[17];
> 25: MOV output[2], temp[28];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: TEX temp[8].xyz, input[0].xy__, 2D[0];
>  2: ADD temp[10].xyz, temp[8].xyz_, none.-H-H-H_;
>  3: MUL temp[11].xyz, temp[10].xxx_, input[3].xyz_;
>  4: MAD temp[12].xyz, temp[10].yyy_, input[2].xyz_, temp[11].xyz_;
>  5: MAD temp[13].xyz, temp[10].zzz_, input[4].xyz_, temp[12].xyz_;
>  6: DP3 temp[14].x, temp[13].xyz_, temp[13].xyz_;
>  7: RSQ temp[15].x, |temp[14].x___|;
>  8: MUL temp[16].xyz, temp[13].xyz_, temp[15].xxx_;
>  9: MAD temp[17].xyz, temp[16].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[18].x, input[1].x___, const[0].z___;
> 11: FRC temp[19].x, temp[18].x___;
> 12: MOV temp[21].x, (temp[18] - temp[19]).x___;
> 13: ADD temp[22].x, temp[18].x___, -(temp[18] - temp[19]).x___;
> 14: MUL temp[23].x, temp[22].x___, const[0].z___;
> 15: FRC temp[24].x, temp[23].x___;
> 16: MOV temp[21].y, (temp[23] - temp[24])._x__;
> 17: ADD temp[26].x, temp[23].x___, -(temp[23] - temp[24]).x___;
> 18: MOV temp[27].z, temp[26].__x_;
> 19: MUL temp[27].xy, temp[21].xy__, const[0].ww__;
> 20: MOV temp[28].xyz, temp[27].xyz_;
> 21: MOV temp[17].w, none.___0;
> 22: MOV temp[28].w, none.___0;
> 23: MOV output[0], temp[7];
> 24: MOV output[1], temp[17];
> 25: MOV output[2], temp[28];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: TEX temp[8].xyz, input[0].xy__, 2D[0];
>  2: ADD temp[10].xyz, temp[8].xyz_, none.-H-H-H_;
>  3: MUL temp[11].xyz, temp[10].xxx_, input[3].xyz_;
>  4: MAD temp[12].xyz, temp[10].yyy_, input[2].xyz_, temp[11].xyz_;
>  5: MAD temp[13].xyz, temp[10].zzz_, input[4].xyz_, temp[12].xyz_;
>  6: DP3 temp[14].x, temp[13].xyz_, temp[13].xyz_;
>  7: RSQ temp[15].x, |temp[14].x___|;
>  8: MUL temp[16].xyz, temp[13].xyz_, temp[15].xxx_;
>  9: MAD temp[17].xyz, temp[16].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[18].x, input[1].x___, const[0].z___;
> 11: FRC temp[19].x, temp[18].x___;
> 12: MOV temp[21].x, (temp[18] - temp[19]).x___;
> 13: ADD temp[22].x, temp[18].x___, -(temp[18] - temp[19]).x___;
> 14: MUL temp[23].x, temp[22].x___, const[0].z___;
> 15: FRC temp[24].x, temp[23].x___;
> 16: MOV temp[21].y, (temp[23] - temp[24])._x__;
> 17: ADD temp[26].x, temp[23].x___, -(temp[23] - temp[24]).x___;
> 18: MOV temp[27].z, temp[26].__x_;
> 19: MUL temp[27].xy, temp[21].xy__, const[0].ww__;
> 20: MOV temp[28].xyz, temp[27].xyz_;
> 21: MOV temp[17].w, none.___0;
> 22: MOV temp[28].w, none.___0;
> 23: MOV output[0], temp[7];
> 24: MOV output[1], temp[17];
> 25: MOV output[2], temp[28];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: TEX temp[8].xyz, input[0].xy__, 2D[0];
>  2: src0.xyz = temp[8]
>     MAD temp[10].xyz, src0.xyz, src0.111, -src0.HHH
>  3: src0.xyz = temp[10], src1.xyz = input[3]
>     MAD temp[11].xyz, src0.xxx, src1.xyz, src0.000
>  4: src0.xyz = temp[10], src1.xyz = input[2], src2.xyz = temp[11]
>     MAD temp[12].xyz, src0.yyy, src1.xyz, src2.xyz
>  5: src0.xyz = temp[10], src1.xyz = input[4], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.zzz, src1.xyz, src2.xyz
>  6: src0.xyz = temp[13]
>     DP3 temp[14].x, src0.xyz, src0.xyz
>  7: src0.xyz = temp[14]
>     REPL_ALPHA temp[15].x
>     RSQ, |src0.x|
>  8: src0.xyz = temp[13], src1.xyz = temp[15]
>     MAD temp[16].xyz, src0.xyz, src1.xxx, src0.000
>  9: src0.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.HHH, src0.HHH
> 10: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[18].x, src0.x__, src1.z__, src0.000
> 11: src0.xyz = temp[18]
>     FRC temp[19].x, src0.x__
> 12: src0.xyz = temp[19], src1.xyz = temp[18], srcp.xyz = (src1 - src0)
>     MAD temp[21].x, srcp.x__, src0.111, src0.000
> 13: src0.xyz = temp[19], src1.xyz = temp[18], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, src1.x__, src0.111, -srcp.x__
> 14: src0.xyz = temp[22], src1.xyz = const[0]
>     MAD temp[23].x, src0.x__, src1.z__, src0.000
> 15: src0.xyz = temp[23]
>     FRC temp[24].x, src0.x__
> 16: src0.xyz = temp[24], src1.xyz = temp[23], srcp.xyz = (src1 - src0)
>     MAD temp[21].y, srcp._x_, src0.111, src0.000
> 17: src0.xyz = temp[24], src1.xyz = temp[23], srcp.xyz = (src1 - src0)
>     MAD temp[26].x, src1.x__, src0.111, -srcp.x__
> 18: src0.xyz = temp[26]
>     MAD temp[27].z, src0.__x, src0.111, src0.000
> 19: src0.xyz = temp[21], src0.w = const[0]
>     MAD temp[27].xy, src0.xy_, src0.ww_, src0.000
> 20: src0.xyz = temp[27]
>     MAD temp[28].xyz, src0.xyz, src0.111, src0.000
> 21: 
>     MAD temp[17].w, src0.0, src0.1, src0.0
> 22: 
>     MAD temp[28].w, src0.0, src0.1, src0.0
> 23: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 24: src0.xyz = temp[17], src0.w = temp[17]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 25: src0.xyz = temp[28], src0.w = temp[28]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[8].xyz, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[8] SEM_WAIT
>     MAD temp[10].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[28].w, src0.0, src0.1, src0.0
>  4: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[18].x, src0.x__, src1.z__, src0.000
>     MAD temp[17].w, src0.0, src0.1, src0.0
>  5: src0.xyz = temp[10], src1.xyz = input[3]
>     MAD temp[11].xyz, src0.xxx, src1.xyz, src0.000
>  6: src0.xyz = temp[10], src1.xyz = input[2], src2.xyz = temp[11]
>     MAD temp[12].xyz, src0.yyy, src1.xyz, src2.xyz
>  7: src0.xyz = temp[18]
>     FRC temp[19].x, src0.x__
>  8: src0.xyz = temp[10], src1.xyz = input[4], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.zzz, src1.xyz, src2.xyz
>  9: src0.xyz = temp[19], src1.xyz = temp[18], src2.xyz = temp[13], srcp.xyz = (src1 - src0)
>     DP3 temp[14].x, src2.xyz, src2.xyz
>     MAD temp[22].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[19], src1.xyz = temp[18], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     MAD temp[21].x, srcp.x__, src0.111, src0.000
>     RSQ temp[15].w, |src2.x|
> 11: src0.xyz = temp[22], src0.w = temp[22], src1.xyz = const[0]
>     MAD temp[23].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[13], src0.w = temp[15], src1.xyz = temp[15]
>     MAD temp[16].xyz, src0.xyz, src0.www, src0.000
> 13: src0.xyz = temp[23]
>     FRC temp[24].x, src0.x__
> 14: src0.xyz = temp[24], src1.xyz = temp[23], src2.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[17].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[26].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[24], src1.xyz = temp[23], srcp.xyz = (src1 - src0)
>     MAD temp[21].y, srcp._x_, src0.111, src0.000
> 16: src0.xyz = temp[26], src0.w = temp[26]
>     MAD temp[27].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[21], src0.w = const[0]
>     MAD temp[27].xy, src0.xy_, src0.ww_, src0.000
> 18: src0.xyz = temp[27]
>     MAD temp[28].xyz, src0.xyz, src0.111, src0.000
> 19: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[17], src0.w = temp[17]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 21: src0.xyz = temp[28], src0.w = temp[28]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[8].xyz, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[8] SEM_WAIT
>     MAD temp[10].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[28].w, src0.0, src0.1, src0.0
>  4: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[18].x, src0.x__, src1.z__, src0.000
>     MAD temp[17].w, src0.0, src0.1, src0.0
>  5: src0.xyz = temp[10], src1.xyz = input[3]
>     MAD temp[11].xyz, src0.xxx, src1.xyz, src0.000
>  6: src0.xyz = temp[10], src1.xyz = input[2], src2.xyz = temp[11]
>     MAD temp[12].xyz, src0.yyy, src1.xyz, src2.xyz
>  7: src0.xyz = temp[18]
>     FRC temp[19].x, src0.x__
>  8: src0.xyz = temp[10], src1.xyz = input[4], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.zzz, src1.xyz, src2.xyz
>  9: src0.xyz = temp[19], src1.xyz = temp[18], src2.xyz = temp[13], srcp.xyz = (src1 - src0)
>     DP3 temp[14].x, src2.xyz, src2.xyz
>     MAD temp[22].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[19], src1.xyz = temp[18], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     MAD temp[21].x, srcp.x__, src0.111, src0.000
>     RSQ temp[15].w, |src2.x|
> 11: src0.w = temp[22], src1.xyz = const[0]
>     MAD temp[23].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[13], src0.w = temp[15]
>     MAD temp[16].xyz, src0.xyz, src0.www, src0.000
> 13: src0.xyz = temp[23]
>     FRC temp[24].x, src0.x__
> 14: src0.xyz = temp[24], src1.xyz = temp[23], src2.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[17].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[26].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[24], src1.xyz = temp[23], srcp.xyz = (src1 - src0)
>     MAD temp[21].y, srcp._x_, src0.111, src0.000
> 16: src0.w = temp[26]
>     MAD temp[27].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[21], src0.w = const[0]
>     MAD temp[27].xy, src0.xy_, src0.ww_, src0.000
> 18: src0.xyz = temp[27]
>     MAD temp[28].xyz, src0.xyz, src0.111, src0.000
> 19: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[17], src0.w = temp[17]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 21: src0.xyz = temp[28], src0.w = temp[28]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[5], input[0].xy__, 2D[1];
>  2: TEX temp[0].xyz, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[0].w, src0.0, src0.1, src0.0
>  4: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[1].x, src0.x__, src1.z__, src0.0__
>     MAD temp[1].w, src0.0, src0.1, src0.0
>  5: src0.xyz = temp[0], src1.xyz = input[3]
>     MAD temp[3].xyz, src0.xxx, src1.xyz, src0.000
>  6: src0.xyz = temp[0], src1.xyz = input[2], src2.xyz = temp[3]
>     MAD temp[2].xyz, src0.yyy, src1.xyz, src2.xyz
>  7: src0.xyz = temp[1]
>     FRC temp[3].x, src0.x__
>  8: src0.xyz = temp[0], src1.xyz = input[4], src2.xyz = temp[2]
>     MAD temp[0].xyz, src0.zzz, src1.xyz, src2.xyz
>  9: src0.xyz = temp[3], src1.xyz = temp[1], src2.xyz = temp[0], srcp.xyz = (src1 - src0)
>     DP3 temp[1].y, src2.xyz, src2.xyz
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[3], src1.xyz = temp[1], src2.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[1].x, srcp.x__, src0.11_, src0.00_
>     RSQ temp[3].w, |src2.y|
> 11: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.z__, src0.0__
> 12: src0.xyz = temp[0], src0.w = temp[3]
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 13: src0.xyz = temp[2]
>     FRC temp[3].x, src0.x__
> 14: src0.xyz = temp[3], src1.xyz = temp[2], src2.xyz = temp[0], srcp.xyz = (src1 - src0)
>     MAD temp[0].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[1].y, srcp._x_, src0.11_, src0.00_
> 16: src0.w = temp[2]
>     MAD temp[2].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[1], src0.w = const[0]
>     MAD temp[2].xy, src0.xy_, src0.ww_, src0.000
> 18: src0.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 19: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 21: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=28*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 18, tex_end: 1  (code_addr: 00420480)
>  TEX:
>    TEX t5, t0, texture[1]   (00008940)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xyz      1.0     -0.5    op: 000d8a80 
>       w:      0.0      1.0      0.0    op: 00040890
>  1: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:      0.0      1.0      0.0    op: 00040890
>  2: xyz:  t0  t3  t0  bias-> t3.xyz               (038c00c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t3.xyz      0.0    op: 00050201 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  t2  t3  bias-> t2.xyz               (03883080)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.yyy   t2.xyz   t3.xyz    op: 00020202 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t1  t0  t0  bias-> t3.x                 (008c0001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 04804080 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  5: xyz:  t0  t4  t2  bias-> t0.xyz               (03802100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.zzz   t4.xyz   t2.xyz    op: 00020203 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  6: xyz:  t3  t1  t0   sub-> t1.y                 (01040043)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz   t0.xyz   t3.xxx    op: 00a04408 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  7: xyz:  t3  t1  t1   sub-> t1.x                 (00841043)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:   |t1.y|     t3.x     t3.x    op: 05800047
>  8: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.zzz      0.0    op: 0005038c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  9: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t0.xyz   t3.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 10: xyz:  t2  t0  t0  bias-> t3.x                 (008c0002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 11: xyz:  t3  t2  t0   sub-> t0.xyz               (03800083)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz      0.5      0.5    op: 00258b08 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 12: xyz:  t3  t2  t0   sub-> t1.y                 (01040083)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> t2.z                 (02080000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 14: xyz:  t1  t0  t0  bias-> t2.xy                (01880001)
>       w:  c0  t0  t0  bias->                      (00000020)
>     xyz:   t1.xyz   c0.www      0.0    op: 00050600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 15: xyz:  t2  t0  t0  bias-> t1.xyz               (03840002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 16: xyz:  t5  t0  t0  bias-> o0.xyz               (1c000005)
>       w:  t5  t0  t0  bias-> o0.w                 (01000005)
>     xyz:   t5.xyz      1.0      0.0    op: 00050a80 
>       w:     t5.w      1.0      0.0    op: 00040889
> 17: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t1  t0  t0  bias-> o1.w                 (03000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
> 18: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t0  t0  t0  bias-> o2.w                 (05000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[1]
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1]
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=29*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t0.xyz   t1.xyz      0.0    op: 00050200 
>       w:     t0.w     t1.w      0.0    op: 00040509
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], LINEAR
>  0: END
>Fragment Program: before compilation
># Radeon Compiler Program
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>pc=30*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00000000)
>  0: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL SAMP[4]
>DCL SAMP[5]
>DCL CONST[13..14]
>DCL CONST[6..12]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>DCL TEMP[6], LOCAL
>DCL TEMP[7], LOCAL
>DCL TEMP[8], LOCAL
>DCL TEMP[9], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     2.0000}
>IMM[1] FLT32 {   -1.0000,     0.0000,    10.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[14].xxxx, CONST[14].yyyy
>  2: TEX TEMP[1], TEMP[0].xyyy, SAMP[4], RECT
>  3: TEX TEMP[2], TEMP[0].xyyy, SAMP[3], RECT
>  4: RCP TEMP[3].x, IN[1].zzzz
>  5: MUL TEMP[3].xy, IN[1].xyyy, TEMP[3].xxxx
>  6: MOV TEMP[3].z, CONST[10].xxxx
>  7: DP3 TEMP[4].x, TEMP[2].xyzz, IMM[0].xyzz
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: ADD TEMP[4].xyz, CONST[13].xyzz, -TEMP[3].xyzz
> 10: MUL TEMP[4].xyz, TEMP[4].xyzz, CONST[12].xxxx
> 11: DP3 TEMP[5].x, TEMP[4].xyzz, TEMP[4].xyzz
> 12: RSQ TEMP[5].x, TEMP[5].xxxx
> 13: MUL TEMP[5].xyz, TEMP[4].xyzz, TEMP[5].xxxx
> 14: MUL TEMP[6], CONST[6], TEMP[3].xxxx
> 15: MAD TEMP[6], CONST[7], TEMP[3].yyyy, TEMP[6]
> 16: MAD TEMP[6], CONST[8], TEMP[3].zzzz, TEMP[6]
> 17: ADD TEMP[6], TEMP[6], CONST[9]
> 18: MAD TEMP[7].xyz, TEMP[1].xyzz, IMM[0].wwww, IMM[1].xxxx
> 19: DP3 TEMP[8].x, TEMP[7].xyzz, TEMP[7].xyzz
> 20: RSQ TEMP[8].x, TEMP[8].xxxx
> 21: MUL TEMP[7].xyz, TEMP[7].xyzz, TEMP[8].xxxx
> 22: MOV TEMP[3].xyz, -TEMP[3].xyzx
> 23: DP3 TEMP[8].x, TEMP[3].xyzz, TEMP[3].xyzz
> 24: RSQ TEMP[8].x, TEMP[8].xxxx
> 25: MAD TEMP[3].xyz, TEMP[3].xyzz, TEMP[8].xxxx, TEMP[5].xyzz
> 26: TEX TEMP[8].xyz, TEMP[0].xyyy, SAMP[5], RECT
> 27: MUL TEMP[8].xyz, TEMP[8].xyzz, CONST[11].xyzz
> 28: DP3 TEMP[5].x, TEMP[5].xyzz, TEMP[7].xyzz
> 29: MAX TEMP[5].x, TEMP[5].xxxx, IMM[1].yyyy
> 30: MUL TEMP[1].x, CONST[11].wwww, TEMP[1].wwww
> 31: DP3 TEMP[9].x, TEMP[3].xyzz, TEMP[3].xyzz
> 32: RSQ TEMP[9].x, TEMP[9].xxxx
> 33: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[9].xxxx
> 34: DP3_SAT TEMP[3].x, TEMP[3].xyzz, TEMP[7].xyzz
> 35: MUL TEMP[2].x, TEMP[2].wwww, IMM[1].zzzz
> 36: EX2 TEMP[2].x, TEMP[2].xxxx
> 37: ADD TEMP[2].x, TEMP[2].xxxx, IMM[0].xxxx
> 38: POW TEMP[2].x, TEMP[3].xxxx, TEMP[2].xxxx
> 39: MUL TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
> 40: MUL TEMP[1].xyz, TEMP[1].xxxx, CONST[11].xyzz
> 41: MAD TEMP[1].xyz, TEMP[8].xyzz, TEMP[5].xxxx, TEMP[1].xyzz
> 42: TXP TEMP[2].xyz, TEMP[6].xyyw, SAMP[1], 2D
> 43: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[2].xyzz
> 44: DP3 TEMP[2].x, TEMP[4].xyzz, TEMP[4].xyzz
> 45: TEX TEMP[2].x, TEMP[2].xxxx, SAMP[2], 1D
> 46: TXP TEMP[3].x, TEMP[6], SAMP[0], SHADOW2D
> 47: MUL TEMP[2].x, TEMP[2].xxxx, TEMP[3].xxxx
> 48: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[2].xxxx
> 49: MOV OUT[0], TEMP[1]
> 50: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[4];
>  6: TEX temp[2], temp[0].xyyy, RECT[3];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[10].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MUL temp[6], const[6], temp[3].xxxx;
> 18: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 19: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 20: ADD temp[6], temp[6], const[9];
> 21: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 22: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 23: RSQ temp[8].x, temp[8].xxxx;
> 24: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 25: MOV temp[3].xyz, -temp[3].xyzx;
> 26: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 27: RSQ temp[8].x, temp[8].xxxx;
> 28: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 29: TEX temp[8].xyz, temp[0].xyyy, RECT[5];
> 30: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 33: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 34: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[9].x, temp[9].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 44: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: TXP temp[2].xyz, temp[6].xyyw, 2D[1];
> 46: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[2].x, temp[2].xxxx, 1D[2];
> 49: TXP temp[3].x, temp[6], 2DSHADOW[0];
> 50: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 52: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[4];
>  6: TEX temp[2], temp[0].xyyy, RECT[3];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[10].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MUL temp[6], const[6], temp[3].xxxx;
> 18: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 19: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 20: ADD temp[6], temp[6], const[9];
> 21: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 22: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 23: RSQ temp[8].x, temp[8].xxxx;
> 24: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 25: MOV temp[3].xyz, -temp[3].xyzx;
> 26: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 27: RSQ temp[8].x, temp[8].xxxx;
> 28: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 29: TEX temp[8].xyz, temp[0].xyyy, RECT[5];
> 30: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 33: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 34: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[9].x, temp[9].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 44: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: TXP temp[2].xyz, temp[6].xyyw, 2D[1];
> 46: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[2].x, temp[2].xxxx, 1D[2];
> 49: TXP temp[3].x, temp[6], 2DSHADOW[0];
> 50: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 52: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[4];
>  6: TEX temp[2], temp[0].xyyy, RECT[3];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[10].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MUL temp[6], const[6], temp[3].xxxx;
> 18: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 19: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 20: ADD temp[6], temp[6], const[9];
> 21: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 22: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 23: RSQ temp[8].x, temp[8].xxxx;
> 24: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 25: MOV temp[3].xyz, -temp[3].xyzx;
> 26: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 27: RSQ temp[8].x, temp[8].xxxx;
> 28: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 29: TEX temp[8].xyz, temp[0].xyyy, RECT[5];
> 30: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 33: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 34: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[9].x, temp[9].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 44: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: TXP temp[2].xyz, temp[6].xyyw, 2D[1];
> 46: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[2].x, temp[2].xxxx, 1D[2];
> 49: TXP temp[3].x, temp[6], 2DSHADOW[0];
> 50: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 52: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[4];
>  6: TEX temp[2], temp[0].xyyy, RECT[3];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[10].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MUL temp[6], const[6], temp[3].xxxx;
> 18: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 19: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 20: ADD temp[6], temp[6], const[9];
> 21: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 22: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 23: RSQ temp[8].x, temp[8].xxxx;
> 24: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 25: MOV temp[3].xyz, -temp[3].xyzx;
> 26: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 27: RSQ temp[8].x, temp[8].xxxx;
> 28: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 29: TEX temp[8].xyz, temp[0].xyyy, RECT[5];
> 30: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 33: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 34: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[9].x, temp[9].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 44: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: TXP temp[2].xyz, temp[6].xyyw, 2D[1];
> 46: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[2].x, temp[2].xxxx, 1D[2];
> 49: TXP temp[3].x, temp[6], 2DSHADOW[0];
> 50: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 52: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[4];
>  6: TEX temp[2], temp[0].xyyy, RECT[3];
>  7: RCP temp[3].x, input[1].zzzz;
>  8: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
>  9: MOV temp[3].z, const[10].xxxx;
> 10: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 11: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 12: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 13: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 14: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 15: RSQ temp[5].x, temp[5].xxxx;
> 16: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 17: MUL temp[6], const[6], temp[3].xxxx;
> 18: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 19: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 20: ADD temp[6], temp[6], const[9];
> 21: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 22: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 23: RSQ temp[8].x, temp[8].xxxx;
> 24: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 25: MOV temp[3].xyz, -temp[3].xyzx;
> 26: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 27: RSQ temp[8].x, temp[8].xxxx;
> 28: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 29: TEX temp[8].xyz, temp[0].xyyy, RECT[5];
> 30: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 31: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 32: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 33: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 34: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 35: RSQ temp[9].x, temp[9].xxxx;
> 36: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 37: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 38: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 39: EX2 temp[2].x, temp[2].xxxx;
> 40: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 41: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 42: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 43: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 44: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 45: TXP temp[2].xyz, temp[6].xyyw, 2D[1];
> 46: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 47: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 48: TEX temp[2].x, temp[2].xxxx, 1D[2];
> 49: TXP temp[3].x, temp[6], 2DSHADOW[0];
> 50: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 52: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: MUL temp[11], temp[0].xyyy, const[19];
>  6: TEX temp[1], temp[11], 2D[4];
>  7: MUL temp[12], temp[0].xyyy, const[20];
>  8: TEX temp[2], temp[12], 2D[3];
>  9: RCP temp[3].x, input[1].zzzz;
> 10: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
> 11: MOV temp[3].z, const[10].xxxx;
> 12: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 13: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 14: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 15: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 16: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 17: RSQ temp[5].x, temp[5].xxxx;
> 18: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 19: MUL temp[6], const[6], temp[3].xxxx;
> 20: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 21: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 22: ADD temp[6], temp[6], const[9];
> 23: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 24: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 25: RSQ temp[8].x, temp[8].xxxx;
> 26: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 27: MOV temp[3].xyz, -temp[3].xyzx;
> 28: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 29: RSQ temp[8].x, temp[8].xxxx;
> 30: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 31: MUL temp[13], temp[0].xyyy, const[21];
> 32: TEX temp[14], temp[13], 2D[5];
> 33: MOV temp[8].xyz, temp[14];
> 34: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 35: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 36: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 37: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 38: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 39: RSQ temp[9].x, temp[9].xxxx;
> 40: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 41: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 42: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 43: EX2 temp[2].x, temp[2].xxxx;
> 44: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 45: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 46: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 47: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 48: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 49: TXP temp[15], temp[6].xyyw, 2D[1];
> 50: MOV temp[2].xyz, temp[15];
> 51: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 52: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 53: TEX temp[16], temp[2].xxxx, 1D[2];
> 54: MOV temp[2].x, temp[16];
> 55: TXP temp[17], temp[6], 2DSHADOW[0];
> 56: RCP temp[18].w, temp[6].wwww;
> 57: MUL_SAT temp[18].w, temp[6].zzzz, temp[18].wwww;
> 58: ADD temp[18].w, -temp[18].wwww, temp[17].xxxx;
> 59: CMP temp[3].x, temp[18].www1, none.0001, none.1111;
> 60: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 61: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 62: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].wwww;
>  1: MUL temp[10].xyz, input[0], temp[10].wwww;
>  2: MAD temp[10].xyz, temp[10].xyz0, const[17].xyz0, const[18].xyz0;
>  3: MOV temp[0], temp[10];
>  4: MAD temp[0].y, temp[10], const[14].xxxx, const[14].yyyy;
>  5: MUL temp[11], temp[0].xyyy, const[19];
>  6: TEX temp[1], temp[11], 2D[4];
>  7: MUL temp[12], temp[0].xyyy, const[20];
>  8: TEX temp[2], temp[12], 2D[3];
>  9: RCP temp[3].x, input[1].zzzz;
> 10: MUL temp[3].xy, input[1].xyyy, temp[3].xxxx;
> 11: MOV temp[3].z, const[10].xxxx;
> 12: DP3 temp[4].x, temp[2].xyzz, const[15].xyzz;
> 13: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 14: ADD temp[4].xyz, const[13].xyzz, -temp[3].xyzz;
> 15: MUL temp[4].xyz, temp[4].xyzz, const[12].xxxx;
> 16: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 17: RSQ temp[5].x, |temp[5].xxxx|;
> 18: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 19: MUL temp[6], const[6], temp[3].xxxx;
> 20: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 21: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 22: ADD temp[6], temp[6], const[9];
> 23: MAD temp[7].xyz, temp[1].xyzz, const[15].wwww, const[16].xxxx;
> 24: DP3 temp[8].x, temp[7].xyzz, temp[7].xyzz;
> 25: RSQ temp[8].x, |temp[8].xxxx|;
> 26: MUL temp[7].xyz, temp[7].xyzz, temp[8].xxxx;
> 27: MOV temp[3].xyz, -temp[3].xyzx;
> 28: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 29: RSQ temp[8].x, |temp[8].xxxx|;
> 30: MAD temp[3].xyz, temp[3].xyzz, temp[8].xxxx, temp[5].xyzz;
> 31: MUL temp[13], temp[0].xyyy, const[21];
> 32: TEX temp[14], temp[13], 2D[5];
> 33: MOV temp[8].xyz, temp[14];
> 34: MUL temp[8].xyz, temp[8].xyzz, const[11].xyzz;
> 35: DP3 temp[5].x, temp[5].xyzz, temp[7].xyzz;
> 36: MAX temp[5].x, temp[5].xxxx, const[16].yyyy;
> 37: MUL temp[1].x, const[11].wwww, temp[1].wwww;
> 38: DP3 temp[9].x, temp[3].xyzz, temp[3].xyzz;
> 39: RSQ temp[9].x, |temp[9].xxxx|;
> 40: MUL temp[3].xyz, temp[3].xyzz, temp[9].xxxx;
> 41: DP3_SAT temp[3].x, temp[3].xyzz, temp[7].xyzz;
> 42: MUL temp[2].x, temp[2].wwww, const[16].zzzz;
> 43: EX2 temp[2].x, temp[2].xxxx;
> 44: ADD temp[2].x, temp[2].xxxx, const[15].xxxx;
> 45: LG2 temp[19].w, temp[3].xxxx;
> 46: MUL temp[19].w, temp[19].wwww, temp[2].xxxx;
> 47: EX2 temp[2].x, temp[19].wwww;
> 48: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 49: MUL temp[1].xyz, temp[1].xxxx, const[11].xyzz;
> 50: MAD temp[1].xyz, temp[8].xyzz, temp[5].xxxx, temp[1].xyzz;
> 51: TXP temp[15], temp[6].xyyw, 2D[1];
> 52: MOV temp[2].xyz, temp[15];
> 53: MUL temp[1].xyz, temp[1].xyzz, temp[2].xyzz;
> 54: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 55: TEX temp[16], temp[2].xxxx, 1D[2];
> 56: MOV temp[2].x, temp[16];
> 57: TXP temp[17], temp[6], 2DSHADOW[0];
> 58: RCP temp[18].w, temp[6].wwww;
> 59: MUL_SAT temp[18].w, temp[6].zzzz, temp[18].wwww;
> 60: ADD temp[18].w, -temp[18].wwww, temp[17].xxxx;
> 61: CMP temp[3].x, temp[18].www1, none.0001, none.1111;
> 62: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 63: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 64: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].___w;
>  1: MUL temp[10].xy, input[0].xy__, temp[10].ww__;
>  2: MAD temp[10].xy, temp[10].xy__, const[17].xy__, const[18].xy__;
>  3: MOV temp[0].x, temp[10].x___;
>  4: MAD temp[0].y, temp[10]._y__, const[14]._x__, const[14]._y__;
>  5: MUL temp[11].xy, temp[0].xy__, const[19].xy__;
>  6: TEX temp[1], temp[11].xy__, 2D[4];
>  7: MUL temp[12].xy, temp[0].xy__, const[20].xy__;
>  8: TEX temp[2], temp[12].xy__, 2D[3];
>  9: RCP temp[3].x, input[1].z___;
> 10: MUL temp[3].xy, input[1].xy__, temp[3].xx__;
> 11: MOV temp[3].z, const[10].__x_;
> 12: DP3 temp[4].x, temp[2].xyz_, const[15].xyz_;
> 13: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 14: ADD temp[4].xyz, const[13].xyz_, -temp[3].xyz_;
> 15: MUL temp[4].xyz, temp[4].xyz_, const[12].xxx_;
> 16: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 17: RSQ temp[5].x, |temp[5].x___|;
> 18: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 19: MUL temp[6], const[6], temp[3].xxxx;
> 20: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 21: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 22: ADD temp[6], temp[6], const[9];
> 23: MAD temp[7].xyz, temp[1].xyz_, const[15].www_, const[16].xxx_;
> 24: DP3 temp[8].x, temp[7].xyz_, temp[7].xyz_;
> 25: RSQ temp[8].x, |temp[8].x___|;
> 26: MUL temp[7].xyz, temp[7].xyz_, temp[8].xxx_;
> 27: MOV temp[3].xyz, -temp[3].xyz_;
> 28: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 29: RSQ temp[8].x, |temp[8].x___|;
> 30: MAD temp[3].xyz, temp[3].xyz_, temp[8].xxx_, temp[5].xyz_;
> 31: MUL temp[13].xy, temp[0].xy__, const[21].xy__;
> 32: TEX temp[14].xyz, temp[13].xy__, 2D[5];
> 33: MOV temp[8].xyz, temp[14].xyz_;
> 34: MUL temp[8].xyz, temp[8].xyz_, const[11].xyz_;
> 35: DP3 temp[5].x, temp[5].xyz_, temp[7].xyz_;
> 36: MAX temp[5].x, temp[5].x___, const[16].y___;
> 37: MUL temp[1].x, const[11].w___, temp[1].w___;
> 38: DP3 temp[9].x, temp[3].xyz_, temp[3].xyz_;
> 39: RSQ temp[9].x, |temp[9].x___|;
> 40: MUL temp[3].xyz, temp[3].xyz_, temp[9].xxx_;
> 41: DP3_SAT temp[3].x, temp[3].xyz_, temp[7].xyz_;
> 42: MUL temp[2].x, temp[2].w___, const[16].z___;
> 43: EX2 temp[2].x, temp[2].x___;
> 44: ADD temp[2].x, temp[2].x___, const[15].x___;
> 45: LG2 temp[19].w, temp[3].___x;
> 46: MUL temp[19].w, temp[19].___w, temp[2].___x;
> 47: EX2 temp[2].x, temp[19].w___;
> 48: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 49: MUL temp[1].xyz, temp[1].xxx_, const[11].xyz_;
> 50: MAD temp[1].xyz, temp[8].xyz_, temp[5].xxx_, temp[1].xyz_;
> 51: TXP temp[15].xyz, temp[6].xy_w, 2D[1];
> 52: MOV temp[2].xyz, temp[15].xyz_;
> 53: MUL temp[1].xyz, temp[1].xyz_, temp[2].xyz_;
> 54: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 55: TEX temp[16].x, temp[2].x___, 1D[2];
> 56: MOV temp[2].x, temp[16].x___;
> 57: TXP temp[17].x, temp[6].xy_w, 2DSHADOW[0];
> 58: RCP temp[18].w, temp[6].___w;
> 59: MUL_SAT temp[18].w, temp[6].___z, temp[18].___w;
> 60: ADD temp[18].w, -temp[18].___w, temp[17].___x;
> 61: CMP temp[3].x, temp[18].w___, none.0___, none.1___;
> 62: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 63: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 64: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[10].w, input[0].___w;
>  1: MUL temp[10].xy, input[0].xy__, temp[10].ww__;
>  2: MAD temp[10].xy, temp[10].xy__, const[17].xy__, const[18].xy__;
>  3: MOV temp[0].x, temp[10].x___;
>  4: MAD temp[0].y, temp[10]._y__, const[14]._x__, const[14]._y__;
>  5: MUL temp[11].xy, temp[0].xy__, const[19].xy__;
>  6: TEX temp[1], temp[11].xy__, 2D[4];
>  7: MUL temp[12].xy, temp[0].xy__, const[20].xy__;
>  8: TEX temp[2], temp[12].xy__, 2D[3];
>  9: RCP temp[3].x, input[1].z___;
> 10: MUL temp[3].xy, input[1].xy__, temp[3].xx__;
> 11: MOV temp[3].z, const[10].__x_;
> 12: DP3 temp[4].x, temp[2].xyz_, const[15].xyz_;
> 13: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 14: ADD temp[4].xyz, const[13].xyz_, -temp[3].xyz_;
> 15: MUL temp[4].xyz, temp[4].xyz_, const[12].xxx_;
> 16: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 17: RSQ temp[5].x, |temp[5].x___|;
> 18: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 19: MUL temp[6], const[6], temp[3].xxxx;
> 20: MAD temp[6], const[7], temp[3].yyyy, temp[6];
> 21: MAD temp[6], const[8], temp[3].zzzz, temp[6];
> 22: ADD temp[6], temp[6], const[9];
> 23: MAD temp[7].xyz, temp[1].xyz_, const[15].www_, const[16].xxx_;
> 24: DP3 temp[8].x, temp[7].xyz_, temp[7].xyz_;
> 25: RSQ temp[8].x, |temp[8].x___|;
> 26: MUL temp[7].xyz, temp[7].xyz_, temp[8].xxx_;
> 27: MOV temp[3].xyz, -temp[3].xyz_;
> 28: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 29: RSQ temp[8].x, |temp[8].x___|;
> 30: MAD temp[3].xyz, temp[3].xyz_, temp[8].xxx_, temp[5].xyz_;
> 31: MUL temp[13].xy, temp[0].xy__, const[21].xy__;
> 32: TEX temp[14].xyz, temp[13].xy__, 2D[5];
> 33: MOV temp[8].xyz, temp[14].xyz_;
> 34: MUL temp[8].xyz, temp[8].xyz_, const[11].xyz_;
> 35: DP3 temp[5].x, temp[5].xyz_, temp[7].xyz_;
> 36: MAX temp[5].x, temp[5].x___, const[16].y___;
> 37: MUL temp[1].x, const[11].w___, temp[1].w___;
> 38: DP3 temp[9].x, temp[3].xyz_, temp[3].xyz_;
> 39: RSQ temp[9].x, |temp[9].x___|;
> 40: MUL temp[3].xyz, temp[3].xyz_, temp[9].xxx_;
> 41: DP3_SAT temp[3].x, temp[3].xyz_, temp[7].xyz_;
> 42: MUL temp[2].x, temp[2].w___, const[16].z___;
> 43: EX2 temp[2].x, temp[2].x___;
> 44: ADD temp[2].x, temp[2].x___, const[15].x___;
> 45: LG2 temp[19].w, temp[3].___x;
> 46: MUL temp[19].w, temp[19].___w, temp[2].___x;
> 47: EX2 temp[2].x, temp[19].w___;
> 48: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 49: MUL temp[1].xyz, temp[1].xxx_, const[11].xyz_;
> 50: MAD temp[1].xyz, temp[8].xyz_, temp[5].xxx_, temp[1].xyz_;
> 51: TXP temp[15].xyz, temp[6].xy_w, 2D[1];
> 52: MOV temp[2].xyz, temp[15].xyz_;
> 53: MUL temp[1].xyz, temp[1].xyz_, temp[2].xyz_;
> 54: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 55: TEX temp[16].x, temp[2].x___, 1D[2];
> 56: MOV temp[2].x, temp[16].x___;
> 57: TXP temp[17].x, temp[6].xy_w, 2DSHADOW[0];
> 58: RCP temp[18].w, temp[6].___w;
> 59: MUL_SAT temp[18].w, temp[6].___z, temp[18].___w;
> 60: ADD temp[18].w, -temp[18].___w, temp[17].___x;
> 61: CMP temp[3].x, temp[18].w___, none.0___, none.1___;
> 62: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 63: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 64: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[20].w, input[0].___w;
>  1: MUL temp[21].xy, input[0].xy__, temp[20].ww__;
>  2: MAD temp[22].xy, temp[21].xy__, const[17].xy__, const[18].xy__;
>  3: MOV temp[23].x, temp[22].x___;
>  4: MAD temp[23].y, temp[22]._y__, const[14]._x__, const[14]._y__;
>  5: MUL temp[24].xy, temp[23].xy__, const[19].xy__;
>  6: TEX temp[25], temp[24].xy__, 2D[4];
>  7: MUL temp[26].xy, temp[23].xy__, const[20].xy__;
>  8: TEX temp[27], temp[26].xy__, 2D[3];
>  9: RCP temp[28].x, input[1].z___;
> 10: MUL temp[29].xy, input[1].xy__, temp[28].xx__;
> 11: MOV temp[29].z, const[10].__x_;
> 12: DP3 temp[30].x, temp[27].xyz_, const[15].xyz_;
> 13: MUL temp[31].xyz, temp[29].xyz_, temp[30].xxx_;
> 14: ADD temp[32].xyz, const[13].xyz_, -temp[31].xyz_;
> 15: MUL temp[33].xyz, temp[32].xyz_, const[12].xxx_;
> 16: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 17: RSQ temp[35].x, |temp[34].x___|;
> 18: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 19: MUL temp[37], const[6], temp[31].xxxx;
> 20: MAD temp[38], const[7], temp[31].yyyy, temp[37];
> 21: MAD temp[39], const[8], temp[31].zzzz, temp[38];
> 22: ADD temp[40], temp[39], const[9];
> 23: MAD temp[41].xyz, temp[25].xyz_, const[15].www_, const[16].xxx_;
> 24: DP3 temp[42].x, temp[41].xyz_, temp[41].xyz_;
> 25: RSQ temp[43].x, |temp[42].x___|;
> 26: MUL temp[44].xyz, temp[41].xyz_, temp[43].xxx_;
> 27: MOV temp[45].xyz, -temp[31].xyz_;
> 28: DP3 temp[46].x, temp[45].xyz_, temp[45].xyz_;
> 29: RSQ temp[47].x, |temp[46].x___|;
> 30: MAD temp[48].xyz, temp[45].xyz_, temp[47].xxx_, temp[36].xyz_;
> 31: MUL temp[49].xy, temp[23].xy__, const[21].xy__;
> 32: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 33: MOV temp[51].xyz, temp[50].xyz_;
> 34: MUL temp[52].xyz, temp[51].xyz_, const[11].xyz_;
> 35: DP3 temp[53].x, temp[36].xyz_, temp[44].xyz_;
> 36: MAX temp[54].x, temp[53].x___, const[16].y___;
> 37: MUL temp[55].x, const[11].w___, temp[25].w___;
> 38: DP3 temp[56].x, temp[48].xyz_, temp[48].xyz_;
> 39: RSQ temp[57].x, |temp[56].x___|;
> 40: MUL temp[58].xyz, temp[48].xyz_, temp[57].xxx_;
> 41: DP3_SAT temp[59].x, temp[58].xyz_, temp[44].xyz_;
> 42: MUL temp[60].x, temp[27].w___, const[16].z___;
> 43: EX2 temp[61].x, temp[60].x___;
> 44: ADD temp[62].x, temp[61].x___, const[15].x___;
> 45: LG2 temp[63].w, temp[59].___x;
> 46: MUL temp[64].w, temp[63].___w, temp[62].___x;
> 47: EX2 temp[65].x, temp[64].w___;
> 48: MUL temp[66].x, temp[55].x___, temp[65].x___;
> 49: MUL temp[67].xyz, temp[66].xxx_, const[11].xyz_;
> 50: MAD temp[68].xyz, temp[52].xyz_, temp[54].xxx_, temp[67].xyz_;
> 51: TXP temp[69].xyz, temp[40].xy_w, 2D[1];
> 52: MOV temp[70].xyz, temp[69].xyz_;
> 53: MUL temp[71].xyz, temp[68].xyz_, temp[70].xyz_;
> 54: DP3 temp[72].x, temp[33].xyz_, temp[33].xyz_;
> 55: TEX temp[73].x, temp[72].x___, 1D[2];
> 56: MOV temp[74].x, temp[73].x___;
> 57: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 58: RCP temp[76].w, temp[40].___w;
> 59: MUL_SAT temp[77].w, temp[40].___z, temp[76].___w;
> 60: ADD temp[78].w, -temp[77].___w, temp[75].___x;
> 61: CMP temp[79].x, temp[78].w___, none.0___, none.1___;
> 62: MUL temp[80].x, temp[74].x___, temp[79].x___;
> 63: MUL temp[25].xyz, temp[71].xyz_, temp[80].xxx_;
> 64: MOV output[0], temp[25];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[20].w, input[0].___w;
>  1: MUL temp[21].xy, input[0].xy__, temp[20].ww__;
>  2: MAD temp[22].xy, temp[21].xy__, const[17].xy__, const[18].xy__;
>  3: MOV temp[23].x, temp[22].x___;
>  4: MAD temp[23].y, temp[22]._y__, const[14]._x__, const[14]._y__;
>  5: MUL temp[24].xy, temp[23].xy__, const[19].xy__;
>  6: TEX temp[25], temp[24].xy__, 2D[4];
>  7: MUL temp[26].xy, temp[23].xy__, const[20].xy__;
>  8: TEX temp[27], temp[26].xy__, 2D[3];
>  9: RCP temp[28].x, input[1].z___;
> 10: MUL temp[29].xy, input[1].xy__, temp[28].xx__;
> 11: MOV temp[29].z, const[10].__x_;
> 12: DP3 temp[30].x, temp[27].xyz_, const[15].xyz_;
> 13: MUL temp[31].xyz, temp[29].xyz_, temp[30].xxx_;
> 14: MUL temp[33].xyz, (const[13] - temp[31]).xyz_, const[12].xxx_;
> 15: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 16: RSQ temp[35].x, |temp[34].x___|;
> 17: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 18: MUL temp[37], const[6], temp[31].xxxx;
> 19: MAD temp[38], const[7], temp[31].yyyy, temp[37];
> 20: MAD temp[39], const[8], temp[31].zzzz, temp[38];
> 21: ADD temp[40], temp[39], const[9];
> 22: MAD temp[41].xyz, temp[25].xyz_, const[15].www_, none.-1-1-1_;
> 23: DP3 temp[42].x, temp[41].xyz_, temp[41].xyz_;
> 24: RSQ temp[43].x, |temp[42].x___|;
> 25: MUL temp[44].xyz, temp[41].xyz_, temp[43].xxx_;
> 26: DP3 temp[46].x, temp[31].-x-y-z_, temp[31].-x-y-z_;
> 27: RSQ temp[47].x, |temp[46].x___|;
> 28: MAD temp[48].xyz, temp[31].-x-y-z_, temp[47].xxx_, temp[36].xyz_;
> 29: MUL temp[49].xy, temp[23].xy__, const[21].xy__;
> 30: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 31: MUL temp[52].xyz, temp[50].xyz_, const[11].xyz_;
> 32: DP3 temp[53].x, temp[36].xyz_, temp[44].xyz_;
> 33: MAX temp[54].x, temp[53].x___, none.0___;
> 34: MUL temp[55].x, const[11].w___, temp[25].w___;
> 35: DP3 temp[56].x, temp[48].xyz_, temp[48].xyz_;
> 36: RSQ temp[57].x, |temp[56].x___|;
> 37: MUL temp[58].xyz, temp[48].xyz_, temp[57].xxx_;
> 38: DP3_SAT temp[59].x, temp[58].xyz_, temp[44].xyz_;
> 39: MUL temp[60].x, temp[27].w___, const[16].z___;
> 40: EX2 temp[61].x, temp[60].x___;
> 41: ADD temp[62].x, temp[61].x___, none.1___;
> 42: LG2 temp[63].w, temp[59].___x;
> 43: MUL temp[64].w, temp[63].___w, temp[62].___x;
> 44: EX2 temp[65].x, temp[64].w___;
> 45: MUL temp[66].x, temp[55].x___, temp[65].x___;
> 46: MUL temp[67].xyz, temp[66].xxx_, const[11].xyz_;
> 47: MAD temp[68].xyz, temp[52].xyz_, temp[54].xxx_, temp[67].xyz_;
> 48: TXP temp[69].xyz, temp[40].xy_w, 2D[1];
> 49: MUL temp[71].xyz, temp[68].xyz_, temp[69].xyz_;
> 50: DP3 temp[72].x, temp[33].xyz_, temp[33].xyz_;
> 51: TEX temp[73].x, temp[72].x___, 1D[2];
> 52: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 53: RCP temp[76].w, temp[40].___w;
> 54: MUL_SAT temp[77].w, temp[40].___z, temp[76].___w;
> 55: ADD temp[78].w, -temp[77].___w, temp[75].___x;
> 56: CMP temp[79].x, temp[78].w___, none.0___, none.1___;
> 57: MUL temp[80].x, temp[73].x___, temp[79].x___;
> 58: MUL temp[25].xyz, temp[71].xyz_, temp[80].xxx_;
> 59: MOV output[0], temp[25];
>CONST[15] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[16] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[20].w, input[0].___w;
>  1: MUL temp[21].xy, input[0].xy__, temp[20].ww__;
>  2: MAD temp[22].xy, temp[21].xy__, const[17].xy__, const[18].xy__;
>  3: MOV temp[23].x, temp[22].x___;
>  4: MAD temp[23].y, temp[22]._y__, const[14]._x__, const[14]._y__;
>  5: MUL temp[24].xy, temp[23].xy__, const[19].xy__;
>  6: TEX temp[25], temp[24].xy__, 2D[4];
>  7: MUL temp[26].xy, temp[23].xy__, const[20].xy__;
>  8: TEX temp[27], temp[26].xy__, 2D[3];
>  9: RCP temp[28].x, input[1].z___;
> 10: MUL temp[29].xy, input[1].xy__, temp[28].xx__;
> 11: MOV temp[29].z, const[10].__x_;
> 12: DP3 temp[30].x, temp[27].xyz_, const[15].xyz_;
> 13: MUL temp[31].xyz, temp[29].xyz_, temp[30].xxx_;
> 14: MUL temp[33].xyz, (const[13] - temp[31]).xyz_, const[12].xxx_;
> 15: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 16: RSQ temp[35].x, |temp[34].x___|;
> 17: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 18: MUL temp[37], const[6], temp[31].xxxx;
> 19: MAD temp[38], const[7], temp[31].yyyy, temp[37];
> 20: MAD temp[39], const[8], temp[31].zzzz, temp[38];
> 21: ADD temp[40], temp[39], const[9];
> 22: MAD temp[41].xyz, temp[25].xyz_, const[15].www_, none.-1-1-1_;
> 23: DP3 temp[42].x, temp[41].xyz_, temp[41].xyz_;
> 24: RSQ temp[43].x, |temp[42].x___|;
> 25: MUL temp[44].xyz, temp[41].xyz_, temp[43].xxx_;
> 26: DP3 temp[46].x, temp[31].-x-y-z_, temp[31].-x-y-z_;
> 27: RSQ temp[47].x, |temp[46].x___|;
> 28: MAD temp[48].xyz, temp[31].-x-y-z_, temp[47].xxx_, temp[36].xyz_;
> 29: MUL temp[49].xy, temp[23].xy__, const[21].xy__;
> 30: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 31: MUL temp[52].xyz, temp[50].xyz_, const[11].xyz_;
> 32: DP3 temp[53].x, temp[36].xyz_, temp[44].xyz_;
> 33: MAX temp[54].x, temp[53].x___, none.0___;
> 34: MUL temp[55].x, const[11].w___, temp[25].w___;
> 35: DP3 temp[56].x, temp[48].xyz_, temp[48].xyz_;
> 36: RSQ temp[57].x, |temp[56].x___|;
> 37: MUL temp[58].xyz, temp[48].xyz_, temp[57].xxx_;
> 38: DP3_SAT temp[59].x, temp[58].xyz_, temp[44].xyz_;
> 39: MUL temp[60].x, temp[27].w___, const[16].z___;
> 40: EX2 temp[61].x, temp[60].x___;
> 41: ADD temp[62].x, temp[61].x___, none.1___;
> 42: LG2 temp[63].w, temp[59].___x;
> 43: MUL temp[64].w, temp[63].___w, temp[62].___x;
> 44: EX2 temp[65].x, temp[64].w___;
> 45: MUL temp[66].x, temp[55].x___, temp[65].x___;
> 46: MUL temp[67].xyz, temp[66].xxx_, const[11].xyz_;
> 47: MAD temp[68].xyz, temp[52].xyz_, temp[54].xxx_, temp[67].xyz_;
> 48: TXP temp[69].xyz, temp[40].xy_w, 2D[1];
> 49: MUL temp[71].xyz, temp[68].xyz_, temp[69].xyz_;
> 50: DP3 temp[72].x, temp[33].xyz_, temp[33].xyz_;
> 51: TEX temp[73].x, temp[72].x___, 1D[2];
> 52: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 53: RCP temp[76].w, temp[40].___w;
> 54: MUL_SAT temp[77].w, temp[40].___z, temp[76].___w;
> 55: ADD temp[78].w, -temp[77].___w, temp[75].___x;
> 56: CMP temp[79].x, temp[78].w___, none.0___, none.1___;
> 57: MUL temp[80].x, temp[73].x___, temp[79].x___;
> 58: MUL temp[25].xyz, temp[71].xyz_, temp[80].xxx_;
> 59: MOV output[0], temp[25];
>CONST[9] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[10] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[20].w, input[0].___w;
>  1: MUL temp[21].xy, input[0].xy__, temp[20].ww__;
>  2: MAD temp[22].xy, temp[21].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[23].x, temp[22].x___;
>  4: MAD temp[23].y, temp[22]._y__, const[8]._x__, const[8]._y__;
>  5: MUL temp[24].xy, temp[23].xy__, const[13].xy__;
>  6: TEX temp[25], temp[24].xy__, 2D[4];
>  7: MUL temp[26].xy, temp[23].xy__, const[14].xy__;
>  8: TEX temp[27], temp[26].xy__, 2D[3];
>  9: RCP temp[28].x, input[1].z___;
> 10: MUL temp[29].xy, input[1].xy__, temp[28].xx__;
> 11: MOV temp[29].z, const[4].__x_;
> 12: DP3 temp[30].x, temp[27].xyz_, const[9].xyz_;
> 13: MUL temp[31].xyz, temp[29].xyz_, temp[30].xxx_;
> 14: MUL temp[33].xyz, (const[7] - temp[31]).xyz_, const[6].xxx_;
> 15: DP3 temp[34].x, temp[33].xyz_, temp[33].xyz_;
> 16: RSQ temp[35].x, |temp[34].x___|;
> 17: MUL temp[36].xyz, temp[33].xyz_, temp[35].xxx_;
> 18: MUL temp[37], const[0], temp[31].xxxx;
> 19: MAD temp[38], const[1], temp[31].yyyy, temp[37];
> 20: MAD temp[39], const[2], temp[31].zzzz, temp[38];
> 21: ADD temp[40], temp[39], const[3];
> 22: MAD temp[41].xyz, temp[25].xyz_, const[9].www_, none.-1-1-1_;
> 23: DP3 temp[42].x, temp[41].xyz_, temp[41].xyz_;
> 24: RSQ temp[43].x, |temp[42].x___|;
> 25: MUL temp[44].xyz, temp[41].xyz_, temp[43].xxx_;
> 26: DP3 temp[46].x, temp[31].-x-y-z_, temp[31].-x-y-z_;
> 27: RSQ temp[47].x, |temp[46].x___|;
> 28: MAD temp[48].xyz, temp[31].-x-y-z_, temp[47].xxx_, temp[36].xyz_;
> 29: MUL temp[49].xy, temp[23].xy__, const[15].xy__;
> 30: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 31: MUL temp[52].xyz, temp[50].xyz_, const[5].xyz_;
> 32: DP3 temp[53].x, temp[36].xyz_, temp[44].xyz_;
> 33: MAX temp[54].x, temp[53].x___, none.0___;
> 34: MUL temp[55].x, const[5].w___, temp[25].w___;
> 35: DP3 temp[56].x, temp[48].xyz_, temp[48].xyz_;
> 36: RSQ temp[57].x, |temp[56].x___|;
> 37: MUL temp[58].xyz, temp[48].xyz_, temp[57].xxx_;
> 38: DP3_SAT temp[59].x, temp[58].xyz_, temp[44].xyz_;
> 39: MUL temp[60].x, temp[27].w___, const[10].z___;
> 40: EX2 temp[61].x, temp[60].x___;
> 41: ADD temp[62].x, temp[61].x___, none.1___;
> 42: LG2 temp[63].w, temp[59].___x;
> 43: MUL temp[64].w, temp[63].___w, temp[62].___x;
> 44: EX2 temp[65].x, temp[64].w___;
> 45: MUL temp[66].x, temp[55].x___, temp[65].x___;
> 46: MUL temp[67].xyz, temp[66].xxx_, const[5].xyz_;
> 47: MAD temp[68].xyz, temp[52].xyz_, temp[54].xxx_, temp[67].xyz_;
> 48: TXP temp[69].xyz, temp[40].xy_w, 2D[1];
> 49: MUL temp[71].xyz, temp[68].xyz_, temp[69].xyz_;
> 50: DP3 temp[72].x, temp[33].xyz_, temp[33].xyz_;
> 51: TEX temp[73].x, temp[72].x___, 1D[2];
> 52: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 53: RCP temp[76].w, temp[40].___w;
> 54: MUL_SAT temp[77].w, temp[40].___z, temp[76].___w;
> 55: ADD temp[78].w, -temp[77].___w, temp[75].___x;
> 56: CMP temp[79].x, temp[78].w___, none.0___, none.1___;
> 57: MUL temp[80].x, temp[73].x___, temp[79].x___;
> 58: MUL temp[25].xyz, temp[71].xyz_, temp[80].xxx_;
> 59: MOV output[0], temp[25];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[20].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[20]
>     MAD temp[21].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[21], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[22]
>     MAD temp[23].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[22], src1.xyz = const[8]
>     MAD temp[23].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[23], src1.xyz = const[13]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  6: TEX temp[25], temp[24].xy__, 2D[4];
>  7: src0.xyz = temp[23], src1.xyz = const[14]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
>  8: TEX temp[27], temp[26].xy__, 2D[3];
>  9: src0.xyz = input[1]
>     REPL_ALPHA temp[28].x
>     RCP, src0.z
> 10: src0.xyz = input[1], src1.xyz = temp[28]
>     MAD temp[29].xy, src0.xy_, src1.xx_, src0.000
> 11: src0.xyz = const[4]
>     MAD temp[29].z, src0.__x, src0.111, src0.000
> 12: src0.xyz = temp[27], src1.xyz = const[9]
>     DP3 temp[30].x, src0.xyz, src1.xyz
> 13: src0.xyz = temp[29], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 14: src0.xyz = temp[31], src1.xyz = const[7], src2.xyz = const[6], srcp.xyz = (src1 - src0)
>     MAD temp[33].xyz, srcp.xyz, src2.xxx, src0.000
> 15: src0.xyz = temp[33]
>     DP3 temp[34].x, src0.xyz, src0.xyz
> 16: src0.xyz = temp[34]
>     REPL_ALPHA temp[35].x
>     RSQ, |src0.x|
> 17: src0.xyz = temp[33], src1.xyz = temp[35]
>     MAD temp[36].xyz, src0.xyz, src1.xxx, src0.000
> 18: src0.xyz = const[0], src0.w = const[0], src1.xyz = temp[31]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[37].w, src0.w, src1.x, src0.0
> 19: src0.xyz = const[1], src0.w = const[1], src1.xyz = temp[31], src1.w = temp[37], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src2.xyz
>     MAD temp[38].w, src0.w, src1.y, src1.w
> 20: src0.xyz = const[2], src0.w = const[2], src1.xyz = temp[31], src1.w = temp[38], src2.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.zzz, src2.xyz
>     MAD temp[39].w, src0.w, src1.z, src1.w
> 21: src0.xyz = temp[39], src0.w = temp[39], src1.xyz = const[3], src1.w = const[3]
>     MAD temp[40].xyz, src0.xyz, src0.111, src1.xyz
>     MAD temp[40].w, src0.w, src0.1, src1.w
> 22: src0.xyz = temp[25], src0.w = const[9]
>     MAD temp[41].xyz, src0.xyz, src0.www, -src0.111
> 23: src0.xyz = temp[41]
>     DP3 temp[42].x, src0.xyz, src0.xyz
> 24: src0.xyz = temp[42]
>     REPL_ALPHA temp[43].x
>     RSQ, |src0.x|
> 25: src0.xyz = temp[41], src1.xyz = temp[43]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src0.000
> 26: src0.xyz = temp[31]
>     DP3 temp[46].x, -src0.xyz, -src0.xyz
> 27: src0.xyz = temp[46]
>     REPL_ALPHA temp[47].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[31], src1.xyz = temp[47], src2.xyz = temp[36]
>     MAD temp[48].xyz, -src0.xyz, src1.xxx, src2.xyz
> 29: src0.xyz = temp[23], src1.xyz = const[15]
>     MAD temp[49].xy, src0.xy_, src1.xy_, src0.000
> 30: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 31: src0.xyz = temp[50], src1.xyz = const[5]
>     MAD temp[52].xyz, src0.xyz, src1.xyz, src0.000
> 32: src0.xyz = temp[36], src1.xyz = temp[44]
>     DP3 temp[53].x, src0.xyz, src1.xyz
> 33: src0.xyz = temp[53]
>     MAX temp[54].x, src0.x__, src0.0__
> 34: src0.w = const[5], src1.w = temp[25]
>     MAD temp[55].x, src0.w__, src1.w__, src0.000
> 35: src0.xyz = temp[48]
>     DP3 temp[56].x, src0.xyz, src0.xyz
> 36: src0.xyz = temp[56]
>     REPL_ALPHA temp[57].x
>     RSQ, |src0.x|
> 37: src0.xyz = temp[48], src1.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 38: src0.xyz = temp[58], src1.xyz = temp[44]
>     DP3_SAT temp[59].x, src0.xyz, src1.xyz
> 39: src0.xyz = const[10], src0.w = temp[27]
>     MAD temp[60].x, src0.w__, src0.z__, src0.000
> 40: src0.xyz = temp[60]
>     REPL_ALPHA temp[61].x
>     EX2, src0.x
> 41: src0.xyz = temp[61]
>     MAD temp[62].x, src0.x__, src0.111, src0.1__
> 42: src0.xyz = temp[59]
>     LG2 temp[63].w, src0.x
> 43: src0.xyz = temp[62], src0.w = temp[63]
>     MAD temp[64].w, src0.w, src0.x, src0.0
> 44: src0.w = temp[64]
>     REPL_ALPHA temp[65].x
>     EX2, src0.w
> 45: src0.xyz = temp[55], src1.xyz = temp[65]
>     MAD temp[66].x, src0.x__, src1.x__, src0.000
> 46: src0.xyz = temp[66], src1.xyz = const[5]
>     MAD temp[67].xyz, src0.xxx, src1.xyz, src0.000
> 47: src0.xyz = temp[52], src1.xyz = temp[54], src2.xyz = temp[67]
>     MAD temp[68].xyz, src0.xyz, src1.xxx, src2.xyz
> 48: TXP temp[69].xyz, temp[40].xy_w, 2D[1];
> 49: src0.xyz = temp[68], src1.xyz = temp[69]
>     MAD temp[71].xyz, src0.xyz, src1.xyz, src0.000
> 50: src0.xyz = temp[33]
>     DP3 temp[72].x, src0.xyz, src0.xyz
> 51: TEX temp[73].x, temp[72].x___, 1D[2];
> 52: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 53: src0.w = temp[40]
>     RCP temp[76].w, src0.w
> 54: src0.xyz = temp[40], src0.w = temp[76]
>     MAD_SAT temp[77].w, src0.z, src0.w, src0.0
> 55: src0.xyz = temp[75], src0.w = temp[77]
>     MAD temp[78].w, -src0.w, src0.1, src0.x
> 56: src0.w = temp[78]
>     CMP temp[79].x, src0.1__, src0.0__, src0.w__
> 57: src0.xyz = temp[73], src1.xyz = temp[79]
>     MAD temp[80].x, src0.x__, src1.x__, src0.000
> 58: src0.xyz = temp[71], src1.xyz = temp[80]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 59: src0.xyz = temp[25], src0.w = temp[25]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[4], src0.w = input[0]
>     MAD temp[29].z, src0.__x, src0.111, src0.000
>     RCP temp[20].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[20], src1.xyz = input[1]
>     MAD temp[21].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[28].w, src1.z
>  2: src0.xyz = temp[21], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[28], src1.xyz = temp[28]
>     MAD temp[29].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[22], src1.xyz = const[8]
>     MAD temp[23].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[22]
>     MAD temp[23].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[23], src1.xyz = const[15]
>     MAD temp[49].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[23], src1.xyz = const[14]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[23], src1.xyz = const[13]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 11: TEX temp[27], temp[26].xy__, 2D[3];
> 12: TEX temp[25], temp[24].xy__, 2D[4] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[27], src0.w = const[5], src1.xyz = const[9], src1.w = temp[25] SEM_WAIT
>     DP3 temp[30].x, src0.xyz, src1.xyz
>     MAD temp[55].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[50], src0.w = temp[27], src1.xyz = const[5], src2.xyz = const[10]
>     MAD temp[52].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[60].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[25], src0.w = const[9]
>     MAD temp[41].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[29], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[41], src0.w = temp[60]
>     DP3 temp[42].x, src0.xyz, src0.xyz
>     EX2 temp[61].w, src0.w
> 18: src0.xyz = temp[31], src0.w = temp[61], src1.xyz = const[7], src2.xyz = const[6], srcp.xyz = (src1 - src0)
>     MAD temp[33].xyz, srcp.xyz, src2.xxx, src0.000
>     MAD temp[62].w, src0.w, src0.1, src0.1
> 19: src0.xyz = temp[31], src1.xyz = temp[42]
>     DP3 temp[46].x, -src0.xyz, -src0.xyz
>     RSQ temp[43].w, |src1.x|
> 20: src0.xyz = temp[33], src1.xyz = temp[46]
>     DP3 temp[72].x, src0.xyz, src0.xyz
>     RSQ temp[47].w, |src1.x|
> 21: src0.xyz = const[0], src0.w = const[0], src1.xyz = temp[31]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[37].w, src0.w, src1.x, src0.0
> 22: src0.xyz = temp[41], src0.w = temp[43], src1.xyz = temp[43]
>     MAD temp[44].xyz, src0.xyz, src0.www, src0.000
> 23: src0.xyz = const[1], src0.w = const[1], src1.xyz = temp[31], src1.w = temp[37], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src2.xyz
>     MAD temp[38].w, src0.w, src1.y, src1.w
> 24: src0.xyz = temp[33]
>     DP3 temp[34].x, src0.xyz, src0.xyz
> 25: src0.xyz = const[2], src0.w = const[2], src1.xyz = temp[31], src1.w = temp[38], src2.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.zzz, src2.xyz
>     MAD temp[39].w, src0.w, src1.z, src1.w
> 26: src0.xyz = temp[39], src0.w = temp[39], src1.xyz = const[3], src1.w = const[3]
>     MAD temp[40].xyz, src0.xyz, src0.111, src1.xyz
>     MAD temp[40].w, src0.w, src0.1, src1.w
> 27: src0.xyz = temp[34]
>     RSQ temp[35].w, |src0.x|
> 28: src0.xyz = temp[33], src0.w = temp[35], src1.xyz = temp[35], src1.w = temp[40]
>     MAD temp[36].xyz, src0.xyz, src0.www, src0.000
>     RCP temp[76].w, src1.w
> 29: src0.xyz = temp[36], src0.w = temp[76], src1.xyz = temp[44], src2.xyz = temp[40]
>     DP3 temp[53].x, src0.xyz, src1.xyz
>     MAD_SAT temp[77].w, src2.z, src0.w, src0.0
> 30: src0.xyz = temp[31], src0.w = temp[47], src1.xyz = temp[47], src2.xyz = temp[36]
>     MAD temp[48].xyz, -src0.xyz, src0.www, src2.xyz
> 31: src0.xyz = temp[48], src1.xyz = temp[53]
>     DP3 temp[56].x, src0.xyz, src0.xyz
>     MAX temp[54].w, src1.x, src0.0
> 32: src0.xyz = temp[56]
>     REPL_ALPHA temp[57].x
>     RSQ, |src0.x|
> 33: src0.xyz = temp[48], src1.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 34: src0.xyz = temp[58], src1.xyz = temp[44]
>     DP3_SAT temp[59].x, src0.xyz, src1.xyz
> 35: src0.xyz = temp[59]
>     LG2 temp[63].w, src0.x
> 36: src0.xyz = temp[62], src0.w = temp[63], src1.w = temp[62]
>     MAD temp[64].w, src0.w, src1.w, src0.0
> 37: src0.w = temp[64]
>     REPL_ALPHA temp[65].x
>     EX2, src0.w
> 38: src0.xyz = temp[55], src0.w = temp[55], src1.xyz = temp[65]
>     MAD temp[66].x, src0.w__, src1.x__, src0.000
> 39: src0.xyz = temp[66], src1.xyz = const[5]
>     MAD temp[67].xyz, src0.xxx, src1.xyz, src0.000
> 40: src0.xyz = temp[52], src0.w = temp[54], src1.xyz = temp[54], src2.xyz = temp[67]
>     MAD temp[68].xyz, src0.xyz, src0.www, src2.xyz
> 41: BEGIN_TEX;
> 42: TEX temp[73].x, temp[72].x___, 1D[2];
> 43: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 44: TXP temp[69].xyz, temp[40].xy_w, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 45: src0.xyz = temp[68], src0.w = temp[77], src1.xyz = temp[69], src2.xyz = temp[75] SEM_WAIT
>     MAD temp[71].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[78].w, -src0.w, src0.1, src2.x
> 46: src0.w = temp[78]
>     CMP temp[79].x, src0.1__, src0.0__, src0.w__
> 47: src0.xyz = temp[73], src1.xyz = temp[79]
>     MAD temp[80].x, src0.x__, src1.x__, src0.000
> 48: src0.xyz = temp[71], src1.xyz = temp[80]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 49: src0.xyz = temp[25], src0.w = temp[25]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[4], src0.w = input[0]
>     MAD temp[29].z, src0.__x, src0.111, src0.000
>     RCP temp[20].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[20], src1.xyz = input[1]
>     MAD temp[21].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[28].w, src1.z
>  2: src0.xyz = temp[21], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[28]
>     MAD temp[29].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[22], src1.xyz = const[8]
>     MAD temp[23].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[22]
>     MAD temp[23].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[23], src1.xyz = const[15]
>     MAD temp[49].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[23], src1.xyz = const[14]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[23], src1.xyz = const[13]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[50].xyz, temp[49].xy__, 2D[5];
> 11: TEX temp[27], temp[26].xy__, 2D[3];
> 12: TEX temp[25], temp[24].xy__, 2D[4] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[27], src0.w = const[5], src1.xyz = const[9], src1.w = temp[25] SEM_WAIT
>     DP3 temp[30].x, src0.xyz, src1.xyz
>     MAD temp[55].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[50], src0.w = temp[27], src1.xyz = const[5], src2.xyz = const[10]
>     MAD temp[52].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[60].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[25], src0.w = const[9]
>     MAD temp[41].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[29], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[41], src0.w = temp[60]
>     DP3 temp[42].x, src0.xyz, src0.xyz
>     EX2 temp[61].w, src0.w
> 18: src0.xyz = temp[31], src0.w = temp[61], src1.xyz = const[7], src2.xyz = const[6], srcp.xyz = (src1 - src0)
>     MAD temp[33].xyz, srcp.xyz, src2.xxx, src0.000
>     MAD temp[62].w, src0.w, src0.1, src0.1
> 19: src0.xyz = temp[31], src1.xyz = temp[42]
>     DP3 temp[46].x, -src0.xyz, -src0.xyz
>     RSQ temp[43].w, |src1.x|
> 20: src0.xyz = temp[33], src1.xyz = temp[46]
>     DP3 temp[72].x, src0.xyz, src0.xyz
>     RSQ temp[47].w, |src1.x|
> 21: src0.xyz = const[0], src0.w = const[0], src1.xyz = temp[31]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[37].w, src0.w, src1.x, src0.0
> 22: src0.xyz = temp[41], src0.w = temp[43]
>     MAD temp[44].xyz, src0.xyz, src0.www, src0.000
> 23: src0.xyz = const[1], src0.w = const[1], src1.xyz = temp[31], src1.w = temp[37], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src2.xyz
>     MAD temp[38].w, src0.w, src1.y, src1.w
> 24: src0.xyz = temp[33]
>     DP3 temp[34].x, src0.xyz, src0.xyz
> 25: src0.xyz = const[2], src0.w = const[2], src1.xyz = temp[31], src1.w = temp[38], src2.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.zzz, src2.xyz
>     MAD temp[39].w, src0.w, src1.z, src1.w
> 26: src0.xyz = temp[39], src0.w = temp[39], src1.xyz = const[3], src1.w = const[3]
>     MAD temp[40].xyz, src0.xyz, src0.111, src1.xyz
>     MAD temp[40].w, src0.w, src0.1, src1.w
> 27: src0.xyz = temp[34]
>     RSQ temp[35].w, |src0.x|
> 28: src0.xyz = temp[33], src0.w = temp[35], src1.w = temp[40]
>     MAD temp[36].xyz, src0.xyz, src0.www, src0.000
>     RCP temp[76].w, src1.w
> 29: src0.xyz = temp[36], src0.w = temp[76], src1.xyz = temp[44], src2.xyz = temp[40]
>     DP3 temp[53].x, src0.xyz, src1.xyz
>     MAD_SAT temp[77].w, src2.z, src0.w, src0.0
> 30: src0.xyz = temp[31], src0.w = temp[47], src2.xyz = temp[36]
>     MAD temp[48].xyz, -src0.xyz, src0.www, src2.xyz
> 31: src0.xyz = temp[48], src1.xyz = temp[53]
>     DP3 temp[56].x, src0.xyz, src0.xyz
>     MAX temp[54].w, src1.x, src0.0
> 32: src0.xyz = temp[56]
>     REPL_ALPHA temp[57].x
>     RSQ, |src0.x|
> 33: src0.xyz = temp[48], src1.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 34: src0.xyz = temp[58], src1.xyz = temp[44]
>     DP3_SAT temp[59].x, src0.xyz, src1.xyz
> 35: src0.xyz = temp[59]
>     LG2 temp[63].w, src0.x
> 36: src0.w = temp[63], src1.w = temp[62]
>     MAD temp[64].w, src0.w, src1.w, src0.0
> 37: src0.w = temp[64]
>     REPL_ALPHA temp[65].x
>     EX2, src0.w
> 38: src0.w = temp[55], src1.xyz = temp[65]
>     MAD temp[66].x, src0.w__, src1.x__, src0.000
> 39: src0.xyz = temp[66], src1.xyz = const[5]
>     MAD temp[67].xyz, src0.xxx, src1.xyz, src0.000
> 40: src0.xyz = temp[52], src0.w = temp[54], src2.xyz = temp[67]
>     MAD temp[68].xyz, src0.xyz, src0.www, src2.xyz
> 41: BEGIN_TEX;
> 42: TEX temp[73].x, temp[72].x___, 1D[2];
> 43: TXP temp[75].x, temp[40].xy_w, 2DSHADOW[0];
> 44: TXP temp[69].xyz, temp[40].xy_w, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 45: src0.xyz = temp[68], src0.w = temp[77], src1.xyz = temp[69], src2.xyz = temp[75] SEM_WAIT
>     MAD temp[71].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[78].w, -src0.w, src0.1, src2.x
> 46: src0.w = temp[78]
>     CMP temp[79].x, src0.1__, src0.0__, src0.w__
> 47: src0.xyz = temp[73], src1.xyz = temp[79]
>     MAD temp[80].x, src0.x__, src1.x__, src0.000
> 48: src0.xyz = temp[71], src1.xyz = temp[80]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 49: src0.xyz = temp[25], src0.w = temp[25]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[4], src0.w = input[1]
>     MAD temp[2].z, src0.__x, src0.111, src0.000
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0], src1.xyz = input[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.00_
>     RCP temp[0].w, src1.z
>  2: src0.xyz = temp[1], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[0], src0.w = temp[0]
>     MAD temp[2].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[1], src1.xyz = const[8]
>     MAD temp[0].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[1]
>     MAD temp[0].x, src0.x__, src0.11_, src0.00_
>  6: src0.xyz = temp[0], src1.xyz = const[15]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[0], src1.xyz = const[14]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[0], src1.xyz = const[13]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  9: BEGIN_TEX;
> 10: TEX temp[1].xyz, temp[1].xy__, 2D[5];
> 11: TEX temp[3], temp[3].xy__, 2D[3];
> 12: TEX temp[0], temp[0].xy__, 2D[4] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[3], src0.w = const[5], src1.xyz = const[9], src1.w = temp[0] SEM_WAIT
>     DP3 temp[4].x, src0.xyz, src1.xyz
>     MAD temp[2].w, src0.w, src1.w, src0.0
> 14: src0.xyz = temp[1], src0.w = temp[3], src1.xyz = const[5], src2.xyz = const[10]
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, src0.w, src2.z, src0.0
> 15: src0.xyz = temp[0], src0.w = const[9]
>     MAD temp[3].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[2], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[3], src0.w = temp[1]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     EX2 temp[1].w, src0.w
> 18: src0.xyz = temp[2], src0.w = temp[1], src1.xyz = const[7], src2.xyz = const[6], srcp.xyz = (src1 - src0)
>     MAD temp[5].xyz, srcp.xyz, src2.xxx, src0.000
>     MAD temp[1].w, src0.w, src0.1, src0.1
> 19: src0.xyz = temp[2], src1.xyz = temp[4]
>     DP3 temp[4].x, -src0.xyz, -src0.xyz
>     RSQ temp[3].w, |src1.x|
> 20: src0.xyz = temp[5], src1.xyz = temp[4]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     RSQ temp[4].w, |src1.x|
> 21: src0.xyz = const[0], src0.w = const[0], src1.xyz = temp[2]
>     MAD temp[6].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[5].w, src0.w, src1.x, src0.0
> 22: src0.xyz = temp[3], src0.w = temp[3]
>     MAD temp[3].xyz, src0.xyz, src0.www, src0.000
> 23: src0.xyz = const[1], src0.w = const[1], src1.xyz = temp[2], src1.w = temp[5], src2.xyz = temp[6]
>     MAD temp[6].xyz, src0.xyz, src1.yyy, src2.xyz
>     MAD temp[3].w, src0.w, src1.y, src1.w
> 24: src0.xyz = temp[5]
>     DP3 temp[4].y, src0.xyz, src0.xyz
> 25: src0.xyz = const[2], src0.w = const[2], src1.xyz = temp[2], src1.w = temp[3], src2.xyz = temp[6]
>     MAD temp[6].xyz, src0.xyz, src1.zzz, src2.xyz
>     MAD temp[3].w, src0.w, src1.z, src1.w
> 26: src0.xyz = temp[6], src0.w = temp[3], src1.xyz = const[3], src1.w = const[3]
>     MAD temp[6].xyz, src0.xyz, src0.111, src1.xyz
>     MAD temp[6].w, src0.w, src0.1, src1.w
> 27: src0.xyz = temp[4]
>     RSQ temp[3].w, |src0.y|
> 28: src0.xyz = temp[5], src0.w = temp[3], src1.w = temp[6]
>     MAD temp[5].xyz, src0.xyz, src0.www, src0.000
>     RCP temp[3].w, src1.w
> 29: src0.xyz = temp[5], src0.w = temp[3], src1.xyz = temp[3], src2.xyz = temp[6]
>     DP3 temp[4].y, src0.xyz, src1.xyz
>     MAD_SAT temp[3].w, src2.z, src0.w, src0.0
> 30: src0.xyz = temp[2], src0.w = temp[4], src2.xyz = temp[5]
>     MAD temp[2].xyz, -src0.xyz, src0.www, src2.xyz
> 31: src0.xyz = temp[2], src1.xyz = temp[4]
>     DP3 temp[4].y, src0.xyz, src0.xyz
>     MAX temp[4].w, src1.y, src0.0
> 32: src0.xyz = temp[4]
>     REPL_ALPHA temp[4].y
>     RSQ, |src0.y|
> 33: src0.xyz = temp[2], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.yyy, src0.000
> 34: src0.xyz = temp[2], src1.xyz = temp[3]
>     DP3_SAT temp[2].x, src0.xyz, src1.xyz
> 35: src0.xyz = temp[2]
>     LG2 temp[5].w, src0.x
> 36: src0.w = temp[5], src1.w = temp[1]
>     MAD temp[1].w, src0.w, src1.w, src0.0
> 37: src0.w = temp[1]
>     REPL_ALPHA temp[2].x
>     EX2, src0.w
> 38: src0.w = temp[2], src1.xyz = temp[2]
>     MAD temp[2].x, src0.w__, src1.x__, src0.0__
> 39: src0.xyz = temp[2], src1.xyz = const[5]
>     MAD temp[2].xyz, src0.xxx, src1.xyz, src0.000
> 40: src0.xyz = temp[1], src0.w = temp[4], src2.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src0.www, src2.xyz
> 41: BEGIN_TEX;
> 42: TEX temp[2].x, temp[4].x___, 1D[2];
> 43: TXP temp[4].x, temp[6].xy_w, 2DSHADOW[0];
> 44: TXP temp[5].xyz, temp[6].xy_w, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 45: src0.xyz = temp[1], src0.w = temp[3], src1.xyz = temp[5], src2.xyz = temp[4] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, -src0.w, src0.1, src2.x
> 46: src0.w = temp[1]
>     CMP temp[3].x, src0.1__, src0.0__, src0.w__
> 47: src0.xyz = temp[2], src1.xyz = temp[3]
>     MAD temp[2].x, src0.x__, src1.x__, src0.0__
> 48: src0.xyz = temp[1], src1.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src1.xxx, src0.000
> 49: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=31*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 8, tex_end: 0  (code_addr: 00000200)
>  0: xyz:  c4  t0  t0  bias-> t2.z                 (02080024)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   c4.xxx      1.0      0.0    op: 00050a81 
>       w:     t1.w     c4.x     c4.x    op: 05000009
>  1: xyz:  t1  t0  t0  bias-> t1.xy                (01840001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.z     t1.x     t1.x    op: 05000005
>  2: xyz:  t1 c11 c12  bias-> t1.xy                (0186cac1)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz  c11.xyz  c12.xyz    op: 00020200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t0  t0  t0  bias-> t2.xy                (01880000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t1  c8  t0  bias-> t0.y                 (01000a01)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c8.xxx   c8.xyz    op: 00010280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  5: xyz:  t1  t0  t0  bias-> t0.x                 (00800001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  6: xyz:  t0 c15  t0  bias-> t1.xy                (01840bc0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c15.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0 c14  t0  bias-> t3.xy                (018c0b80)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c14.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0 c13  t0  bias-> t0.xy                (01800b40)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c13.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 9, tex_offset: 0, alu_end: 27, tex_end: 2  (code_addr: 000406c9)
>  TEX:
>    TEX t1, t1, texture[5]   (0000a841)
>    TEX t3, t3, texture[3]   (000098c3)
>    TEX t0, t0, texture[4]   (0000a000)
>  9: xyz:  t3  c9  t0  bias-> t4.x                 (00900a43)
>       w:  c5  t0  t0  bias-> t2.w                 (00880025)
>     xyz:   t3.xyz   c9.xyz   t3.xxx    op: 00804200 
>       w:     c5.w     t0.w      0.0    op: 00040509
> 10: xyz:  t1  c5 c10  bias-> t1.xyz               (0386a941)
>       w:  t3  t0  t0  bias-> t1.w                 (00840003)
>     xyz:   t1.xyz   c5.xyz      0.0    op: 00050200 
>       w:     t3.w    c10.z      0.0    op: 00040409
> 11: xyz:  t0  t0  t0  bias-> t3.xyz               (038c0000)
>       w:  c9  t0  t0  bias->                      (00000029)
>     xyz:   t0.xyz   c9.www     -1.0    op: 000d4600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t2  t4  t0  bias-> t2.xyz               (03880102)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t3  t0  t0  bias-> t4.x                 (00900003)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t1.w     t3.x     t3.x    op: 04000009
> 14: xyz:  t2  c7  c6   sub-> t5.xyz               (039669c2)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz: srcp.xyz   c6.xxx      0.0    op: 0025048f 
>       w:     t1.w      1.0      1.0    op: 00044889
> 15: xyz:  t2  t4  t0  bias-> t4.x                 (00900102)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:  -t2.xyz  -t2.xyz   t2.xxx    op: 00805020 
>       w:   |t4.x|     t2.x     t2.x    op: 05800043
> 16: xyz:  t5  t4  t0  bias-> t4.x                 (00900105)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:   |t4.x|     t5.x     t5.x    op: 05800043
> 17: xyz:  c0  t2  t0  bias-> t6.xyz               (039800a0)
>       w:  c0  t0  t0  bias-> t5.w                 (00940020)
>     xyz:   c0.xyz   t2.xxx      0.0    op: 00050280 
>       w:     c0.w     t2.x      0.0    op: 00040189
> 18: xyz:  t3 t32  t0  bias-> t3.xyz               (038c0003)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.xyz   t3.www      0.0    op: 00050600 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 19: xyz:  c1  t2  t6  bias-> t6.xyz               (039860a1)
>       w:  c1  t5  t0  bias-> t3.w                 (008c0161)
>     xyz:   c1.xyz   t2.yyy   t6.xyz    op: 00020300 
>       w:     c1.w     t2.y     t5.w    op: 00028209
> 20: xyz:  t5  t0  t0  bias-> t4.y                 (01100005)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 21: xyz:  c2  t2  t6  bias-> t6.xyz               (039860a2)
>       w:  c2  t3  t0  bias-> t3.w                 (008c00e2)
>     xyz:   c2.xyz   t2.zzz   t6.xyz    op: 00020380 
>       w:     c2.w     t2.z     t3.w    op: 00028289
> 22: xyz:  t6  c3  t0  bias-> t6.xyz               (039808c6)
>       w:  t3  c3  t0  bias-> t6.w                 (009808c3)
>     xyz:   t6.xyz      1.0   c3.xyz    op: 00010a80 
>       w:     t3.w      1.0     c3.w    op: 00028889
> 23: xyz:  t4  t0  t0  bias->                      (00000004)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t4.xxx   t4.xxx   t4.xxx    op: 00004081 
>       w:   |t4.y|     t4.x     t4.x    op: 05800041
> 24: xyz:  t5 t32  t0  bias-> t5.xyz               (03940005)
>       w:  t3  t6  t0  bias-> t3.w                 (008c0183)
>     xyz:   t5.xyz   t3.www      0.0    op: 00050600 
>       w:     t6.w     t5.x     t5.x    op: 0500000a
> 25: xyz:  t5  t3  t6  bias-> t4.y                 (011060c5)
>       w:  t3  t0  t0  bias-> t3.w                 (008c0003)
>     xyz:   t5.xyz   t3.xyz   t5.xxx    op: 00804200 
>       w:     t6.z     t3.w      0.0    op: 40040488
> 26: xyz:  t2 t32  t5  bias-> t2.xyz               (03885002)
>       w:  t4  t0  t0  bias->                      (00000004)
>     xyz:  -t2.xyz   t4.www   t5.xyz    op: 00020620 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 27: xyz:  t2  t4  t0  bias-> t4.y                 (01100102)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:     t4.y      0.0     t2.x    op: 01800804
> 28: xyz:  t4  t0  t0  bias-> t4.y                 (01100004)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xxx   t4.xxx   t4.xxx    op: 05004081 
>       w:   |t4.y|     t4.x     t4.x    op: 05800041
> 29: xyz:  t2  t4  t0  bias-> t2.xyz               (03880102)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t4.yyy      0.0    op: 00050300 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 30: xyz:  t2  t3  t0  bias-> t2.x                 (008800c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz   t2.xxx    op: 40804200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 31: xyz:  t2  t0  t0  bias->                      (00000002)
>       w:  t0  t0  t0  bias-> t5.w                 (00940000)
>     xyz:   t2.xxx   t2.xxx   t2.xxx    op: 00004081 
>       w:     t2.x     t2.x     t2.x    op: 04800000
> 32: xyz: t32  t0  t0  bias->                      (00000000)
>       w:  t5  t1  t0  bias-> t1.w                 (00840045)
>     xyz:  t32.xxx  t32.xxx  t32.xxx    op: 00004081 
>       w:     t5.w     t1.w      0.0    op: 00040509
> 33: xyz:  t0  t0  t0  bias-> t2.x                 (00880000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 34: xyz: t32  t2  t0  bias-> t2.x                 (00880080)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   t2.xyz      0.0    op: 0005020c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
> 35: xyz:  t2  c5  t0  bias-> t2.xyz               (03880942)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xxx   c5.xyz      0.0    op: 00050201 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 36: xyz:  t1 t32  t2  bias-> t1.xyz               (03842001)
>       w:  t4  t0  t0  bias->                      (00000004)
>     xyz:   t1.xyz   t4.www   t2.xyz    op: 00020600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 2: alu_offset: 37, tex_offset: 3, alu_end: 4, tex_end: 2  (code_addr: 00443125)
>  TEX:
>    TEX t2, t4, texture[2]   (00009084)
>    TXP t4, t6, texture[0]   (00018106)
>    TXP t5, t6, texture[1]   (00018946)
> 37: xyz:  t1  t5  t4  bias-> t1.xyz               (03844141)
>       w:  t3  t0  t0  bias-> t1.w                 (00840003)
>     xyz:   t1.xyz   t5.xyz      0.0    op: 00050200 
>       w:    -t3.w      1.0     t4.x    op: 000188a9
> 38: xyz:  t0  t0  t0  bias-> t3.x                 (008c0000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:      1.0      0.0   t1.www    op: 04030a15 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 39: xyz:  t2  t3  t0  bias-> t2.x                 (008800c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 40: xyz:  t1  t2  t0  bias-> t1.xyz               (03840081)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t2.xxx      0.0    op: 00050280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 41: xyz:  t1  t0  t0  bias-> o0.xyz               (1c000001)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[7..8]
>DCL CONST[4..6]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     2.0000}
>IMM[1] FLT32 {   -1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[8].xxxx, CONST[8].yyyy
>  2: RCP TEMP[1].x, IN[1].zzzz
>  3: MUL TEMP[1].xy, IN[1].xyyy, TEMP[1].xxxx
>  4: MOV TEMP[1].z, CONST[4].xxxx
>  5: TEX TEMP[2].xyz, TEMP[0].xyyy, SAMP[1], RECT
>  6: DP3 TEMP[2].x, TEMP[2].xyzz, IMM[0].xyzz
>  7: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[2].xxxx
>  8: ADD TEMP[1].xyz, CONST[7].xyzz, -TEMP[1].xyzz
>  9: MUL TEMP[1].xyz, TEMP[1].xyzz, CONST[6].xxxx
> 10: TEX TEMP[2].xyz, TEMP[0].xyyy, SAMP[3], RECT
> 11: MUL TEMP[2].xyz, TEMP[2].xyzz, CONST[5].xyzz
> 12: DP3 TEMP[3].x, TEMP[1].xyzz, TEMP[1].xyzz
> 13: RSQ TEMP[3].x, TEMP[3].xxxx
> 14: MUL TEMP[3].xyz, TEMP[1].xyzz, TEMP[3].xxxx
> 15: TEX TEMP[4].xyz, TEMP[0].xyyy, SAMP[2], RECT
> 16: MAD TEMP[4].xyz, TEMP[4].xyzz, IMM[0].wwww, IMM[1].xxxx
> 17: DP3 TEMP[3].x, TEMP[3].xyzz, TEMP[4].xyzz
> 18: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].yyyy
> 19: MUL TEMP[2].xyz, TEMP[2].xyzz, TEMP[3].xxxx
> 20: DP3 TEMP[1].x, TEMP[1].xyzz, TEMP[1].xyzz
> 21: TEX TEMP[1].x, TEMP[1].xxxx, SAMP[0], 1D
> 22: MUL TEMP[1].xyz, TEMP[2].xyzz, TEMP[1].xxxx
> 23: MOV OUT[0], TEMP[1]
> 24: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: TEX temp[2].xyz, temp[0].xyyy, RECT[1];
>  9: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 10: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 11: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 13: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 14: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 15: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 16: RSQ temp[3].x, temp[3].xxxx;
> 17: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 18: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 19: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 21: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 22: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 23: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 24: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 25: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 26: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: TEX temp[2].xyz, temp[0].xyyy, RECT[1];
>  9: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 10: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 11: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 13: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 14: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 15: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 16: RSQ temp[3].x, temp[3].xxxx;
> 17: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 18: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 19: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 21: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 22: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 23: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 24: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 25: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 26: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: TEX temp[2].xyz, temp[0].xyyy, RECT[1];
>  9: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 10: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 11: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 13: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 14: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 15: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 16: RSQ temp[3].x, temp[3].xxxx;
> 17: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 18: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 19: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 21: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 22: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 23: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 24: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 25: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 26: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: TEX temp[2].xyz, temp[0].xyyy, RECT[1];
>  9: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 10: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 11: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 13: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 14: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 15: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 16: RSQ temp[3].x, temp[3].xxxx;
> 17: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 18: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 19: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 21: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 22: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 23: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 24: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 25: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 26: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: TEX temp[2].xyz, temp[0].xyyy, RECT[1];
>  9: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 10: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 11: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 13: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 14: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 15: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 16: RSQ temp[3].x, temp[3].xxxx;
> 17: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 18: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 19: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 20: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 21: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 22: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 23: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 24: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 25: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 26: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: MUL temp[6], temp[0].xyyy, const[13];
>  9: TEX temp[7], temp[6], 2D[1];
> 10: MOV temp[2].xyz, temp[7];
> 11: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 13: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 14: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 15: MUL temp[8], temp[0].xyyy, const[14];
> 16: TEX temp[9], temp[8], 2D[3];
> 17: MOV temp[2].xyz, temp[9];
> 18: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 19: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 20: RSQ temp[3].x, temp[3].xxxx;
> 21: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 22: MUL temp[10], temp[0].xyyy, const[15];
> 23: TEX temp[11], temp[10], 2D[2];
> 24: MOV temp[4].xyz, temp[11];
> 25: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 26: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 28: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 29: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 30: TEX temp[12], temp[1].xxxx, 1D[0];
> 31: MOV temp[1].x, temp[12];
> 32: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 33: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[11].xyz0, const[12].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[8].xxxx, const[8].yyyy;
>  5: RCP temp[1].x, input[1].zzzz;
>  6: MUL temp[1].xy, input[1].xyyy, temp[1].xxxx;
>  7: MOV temp[1].z, const[4].xxxx;
>  8: MUL temp[6], temp[0].xyyy, const[13];
>  9: TEX temp[7], temp[6], 2D[1];
> 10: MOV temp[2].xyz, temp[7];
> 11: DP3 temp[2].x, temp[2].xyzz, const[9].xyzz;
> 12: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 13: ADD temp[1].xyz, const[7].xyzz, -temp[1].xyzz;
> 14: MUL temp[1].xyz, temp[1].xyzz, const[6].xxxx;
> 15: MUL temp[8], temp[0].xyyy, const[14];
> 16: TEX temp[9], temp[8], 2D[3];
> 17: MOV temp[2].xyz, temp[9];
> 18: MUL temp[2].xyz, temp[2].xyzz, const[5].xyzz;
> 19: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 20: RSQ temp[3].x, |temp[3].xxxx|;
> 21: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 22: MUL temp[10], temp[0].xyyy, const[15];
> 23: TEX temp[11], temp[10], 2D[2];
> 24: MOV temp[4].xyz, temp[11];
> 25: MAD temp[4].xyz, temp[4].xyzz, const[9].wwww, const[10].xxxx;
> 26: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[10].yyyy;
> 28: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 29: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 30: TEX temp[12], temp[1].xxxx, 1D[0];
> 31: MOV temp[1].x, temp[12];
> 32: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 33: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[5].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[5].xy, temp[5].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[0].x, temp[5].x___;
>  4: MAD temp[0].y, temp[5]._y__, const[8]._x__, const[8]._y__;
>  5: RCP temp[1].x, input[1].z___;
>  6: MUL temp[1].xy, input[1].xy__, temp[1].xx__;
>  7: MOV temp[1].z, const[4].__x_;
>  8: MUL temp[6].xy, temp[0].xy__, const[13].xy__;
>  9: TEX temp[7].xyz, temp[6].xy__, 2D[1];
> 10: MOV temp[2].xyz, temp[7].xyz_;
> 11: DP3 temp[2].x, temp[2].xyz_, const[9].xyz_;
> 12: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 13: ADD temp[1].xyz, const[7].xyz_, -temp[1].xyz_;
> 14: MUL temp[1].xyz, temp[1].xyz_, const[6].xxx_;
> 15: MUL temp[8].xy, temp[0].xy__, const[14].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[3];
> 17: MOV temp[2].xyz, temp[9].xyz_;
> 18: MUL temp[2].xyz, temp[2].xyz_, const[5].xyz_;
> 19: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 20: RSQ temp[3].x, |temp[3].x___|;
> 21: MUL temp[3].xyz, temp[1].xyz_, temp[3].xxx_;
> 22: MUL temp[10].xy, temp[0].xy__, const[15].xy__;
> 23: TEX temp[11].xyz, temp[10].xy__, 2D[2];
> 24: MOV temp[4].xyz, temp[11].xyz_;
> 25: MAD temp[4].xyz, temp[4].xyz_, const[9].www_, const[10].xxx_;
> 26: DP3 temp[3].x, temp[3].xyz_, temp[4].xyz_;
> 27: MAX temp[3].x, temp[3].x___, const[10].y___;
> 28: MUL temp[2].xyz, temp[2].xyz_, temp[3].xxx_;
> 29: DP3 temp[1].x, temp[1].xyz_, temp[1].xyz_;
> 30: TEX temp[12].x, temp[1].x___, 1D[0];
> 31: MOV temp[1].x, temp[12].x___;
> 32: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 33: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[5].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[5].xy, temp[5].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[0].x, temp[5].x___;
>  4: MAD temp[0].y, temp[5]._y__, const[8]._x__, const[8]._y__;
>  5: RCP temp[1].x, input[1].z___;
>  6: MUL temp[1].xy, input[1].xy__, temp[1].xx__;
>  7: MOV temp[1].z, const[4].__x_;
>  8: MUL temp[6].xy, temp[0].xy__, const[13].xy__;
>  9: TEX temp[7].xyz, temp[6].xy__, 2D[1];
> 10: MOV temp[2].xyz, temp[7].xyz_;
> 11: DP3 temp[2].x, temp[2].xyz_, const[9].xyz_;
> 12: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 13: ADD temp[1].xyz, const[7].xyz_, -temp[1].xyz_;
> 14: MUL temp[1].xyz, temp[1].xyz_, const[6].xxx_;
> 15: MUL temp[8].xy, temp[0].xy__, const[14].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[3];
> 17: MOV temp[2].xyz, temp[9].xyz_;
> 18: MUL temp[2].xyz, temp[2].xyz_, const[5].xyz_;
> 19: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 20: RSQ temp[3].x, |temp[3].x___|;
> 21: MUL temp[3].xyz, temp[1].xyz_, temp[3].xxx_;
> 22: MUL temp[10].xy, temp[0].xy__, const[15].xy__;
> 23: TEX temp[11].xyz, temp[10].xy__, 2D[2];
> 24: MOV temp[4].xyz, temp[11].xyz_;
> 25: MAD temp[4].xyz, temp[4].xyz_, const[9].www_, const[10].xxx_;
> 26: DP3 temp[3].x, temp[3].xyz_, temp[4].xyz_;
> 27: MAX temp[3].x, temp[3].x___, const[10].y___;
> 28: MUL temp[2].xyz, temp[2].xyz_, temp[3].xxx_;
> 29: DP3 temp[1].x, temp[1].xyz_, temp[1].xyz_;
> 30: TEX temp[12].x, temp[1].x___, 1D[0];
> 31: MOV temp[1].x, temp[12].x___;
> 32: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 33: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[8]._x__, const[8]._y__;
>  5: RCP temp[17].x, input[1].z___;
>  6: MUL temp[18].xy, input[1].xy__, temp[17].xx__;
>  7: MOV temp[18].z, const[4].__x_;
>  8: MUL temp[19].xy, temp[16].xy__, const[13].xy__;
>  9: TEX temp[20].xyz, temp[19].xy__, 2D[1];
> 10: MOV temp[21].xyz, temp[20].xyz_;
> 11: DP3 temp[22].x, temp[21].xyz_, const[9].xyz_;
> 12: MUL temp[23].xyz, temp[18].xyz_, temp[22].xxx_;
> 13: ADD temp[24].xyz, const[7].xyz_, -temp[23].xyz_;
> 14: MUL temp[25].xyz, temp[24].xyz_, const[6].xxx_;
> 15: MUL temp[26].xy, temp[16].xy__, const[14].xy__;
> 16: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 17: MOV temp[28].xyz, temp[27].xyz_;
> 18: MUL temp[29].xyz, temp[28].xyz_, const[5].xyz_;
> 19: DP3 temp[30].x, temp[25].xyz_, temp[25].xyz_;
> 20: RSQ temp[31].x, |temp[30].x___|;
> 21: MUL temp[32].xyz, temp[25].xyz_, temp[31].xxx_;
> 22: MUL temp[33].xy, temp[16].xy__, const[15].xy__;
> 23: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 24: MOV temp[35].xyz, temp[34].xyz_;
> 25: MAD temp[36].xyz, temp[35].xyz_, const[9].www_, const[10].xxx_;
> 26: DP3 temp[37].x, temp[32].xyz_, temp[36].xyz_;
> 27: MAX temp[38].x, temp[37].x___, const[10].y___;
> 28: MUL temp[39].xyz, temp[29].xyz_, temp[38].xxx_;
> 29: DP3 temp[40].x, temp[25].xyz_, temp[25].xyz_;
> 30: TEX temp[41].x, temp[40].x___, 1D[0];
> 31: MOV temp[42].x, temp[41].x___;
> 32: MUL temp[43].xyz, temp[39].xyz_, temp[42].xxx_;
> 33: MOV output[0], temp[43];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[8]._x__, const[8]._y__;
>  5: RCP temp[17].x, input[1].z___;
>  6: MUL temp[18].xy, input[1].xy__, temp[17].xx__;
>  7: MOV temp[18].z, const[4].__x_;
>  8: MUL temp[19].xy, temp[16].xy__, const[13].xy__;
>  9: TEX temp[20].xyz, temp[19].xy__, 2D[1];
> 10: DP3 temp[22].x, temp[20].xyz_, const[9].xyz_;
> 11: MUL temp[23].xyz, temp[18].xyz_, temp[22].xxx_;
> 12: MUL temp[25].xyz, (const[7] - temp[23]).xyz_, const[6].xxx_;
> 13: MUL temp[26].xy, temp[16].xy__, const[14].xy__;
> 14: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 15: MUL temp[29].xyz, temp[27].xyz_, const[5].xyz_;
> 16: DP3 temp[30].x, temp[25].xyz_, temp[25].xyz_;
> 17: RSQ temp[31].x, |temp[30].x___|;
> 18: MUL temp[32].xyz, temp[25].xyz_, temp[31].xxx_;
> 19: MUL temp[33].xy, temp[16].xy__, const[15].xy__;
> 20: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 21: MAD temp[36].xyz, temp[34].xyz_, const[9].www_, none.-1-1-1_;
> 22: DP3 temp[37].x, temp[32].xyz_, temp[36].xyz_;
> 23: MAX temp[38].x, temp[37].x___, none.0___;
> 24: MUL temp[39].xyz, temp[29].xyz_, temp[38].xxx_;
> 25: DP3 temp[40].x, temp[25].xyz_, temp[25].xyz_;
> 26: TEX temp[41].x, temp[40].x___, 1D[0];
> 27: MUL temp[43].xyz, temp[39].xyz_, temp[41].xxx_;
> 28: MOV output[0], temp[43];
>CONST[9] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[10] = {    -1.0000     0.0000     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[8]._x__, const[8]._y__;
>  5: RCP temp[17].x, input[1].z___;
>  6: MUL temp[18].xy, input[1].xy__, temp[17].xx__;
>  7: MOV temp[18].z, const[4].__x_;
>  8: MUL temp[19].xy, temp[16].xy__, const[13].xy__;
>  9: TEX temp[20].xyz, temp[19].xy__, 2D[1];
> 10: DP3 temp[22].x, temp[20].xyz_, const[9].xyz_;
> 11: MUL temp[23].xyz, temp[18].xyz_, temp[22].xxx_;
> 12: MUL temp[25].xyz, (const[7] - temp[23]).xyz_, const[6].xxx_;
> 13: MUL temp[26].xy, temp[16].xy__, const[14].xy__;
> 14: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 15: MUL temp[29].xyz, temp[27].xyz_, const[5].xyz_;
> 16: DP3 temp[30].x, temp[25].xyz_, temp[25].xyz_;
> 17: RSQ temp[31].x, |temp[30].x___|;
> 18: MUL temp[32].xyz, temp[25].xyz_, temp[31].xxx_;
> 19: MUL temp[33].xy, temp[16].xy__, const[15].xy__;
> 20: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 21: MAD temp[36].xyz, temp[34].xyz_, const[9].www_, none.-1-1-1_;
> 22: DP3 temp[37].x, temp[32].xyz_, temp[36].xyz_;
> 23: MAX temp[38].x, temp[37].x___, none.0___;
> 24: MUL temp[39].xyz, temp[29].xyz_, temp[38].xxx_;
> 25: DP3 temp[40].x, temp[25].xyz_, temp[25].xyz_;
> 26: TEX temp[41].x, temp[40].x___, 1D[0];
> 27: MUL temp[43].xyz, temp[39].xyz_, temp[41].xxx_;
> 28: MOV output[0], temp[43];
>CONST[5] = {     1.0000     0.0039     0.0000     2.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[6].xy__, const[7].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[4]._x__, const[4]._y__;
>  5: RCP temp[17].x, input[1].z___;
>  6: MUL temp[18].xy, input[1].xy__, temp[17].xx__;
>  7: MOV temp[18].z, const[0].__x_;
>  8: MUL temp[19].xy, temp[16].xy__, const[8].xy__;
>  9: TEX temp[20].xyz, temp[19].xy__, 2D[1];
> 10: DP3 temp[22].x, temp[20].xyz_, const[5].xyz_;
> 11: MUL temp[23].xyz, temp[18].xyz_, temp[22].xxx_;
> 12: MUL temp[25].xyz, (const[3] - temp[23]).xyz_, const[2].xxx_;
> 13: MUL temp[26].xy, temp[16].xy__, const[9].xy__;
> 14: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 15: MUL temp[29].xyz, temp[27].xyz_, const[1].xyz_;
> 16: DP3 temp[30].x, temp[25].xyz_, temp[25].xyz_;
> 17: RSQ temp[31].x, |temp[30].x___|;
> 18: MUL temp[32].xyz, temp[25].xyz_, temp[31].xxx_;
> 19: MUL temp[33].xy, temp[16].xy__, const[10].xy__;
> 20: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 21: MAD temp[36].xyz, temp[34].xyz_, const[5].www_, none.-1-1-1_;
> 22: DP3 temp[37].x, temp[32].xyz_, temp[36].xyz_;
> 23: MAX temp[38].x, temp[37].x___, none.0___;
> 24: MUL temp[39].xyz, temp[29].xyz_, temp[38].xxx_;
> 25: DP3 temp[40].x, temp[25].xyz_, temp[25].xyz_;
> 26: TEX temp[41].x, temp[40].x___, 1D[0];
> 27: MUL temp[43].xyz, temp[39].xyz_, temp[41].xxx_;
> 28: MOV output[0], temp[43];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[4]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = input[1]
>     REPL_ALPHA temp[17].x
>     RCP, src0.z
>  6: src0.xyz = input[1], src1.xyz = temp[17]
>     MAD temp[18].xy, src0.xy_, src1.xx_, src0.000
>  7: src0.xyz = const[0]
>     MAD temp[18].z, src0.__x, src0.111, src0.000
>  8: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[19].xy, src0.xy_, src1.xy_, src0.000
>  9: TEX temp[20].xyz, temp[19].xy__, 2D[1];
> 10: src0.xyz = temp[20], src1.xyz = const[5]
>     DP3 temp[22].x, src0.xyz, src1.xyz
> 11: src0.xyz = temp[18], src1.xyz = temp[22]
>     MAD temp[23].xyz, src0.xyz, src1.xxx, src0.000
> 12: src0.xyz = temp[23], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[25].xyz, srcp.xyz, src2.xxx, src0.000
> 13: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
> 14: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 15: src0.xyz = temp[27], src1.xyz = const[1]
>     MAD temp[29].xyz, src0.xyz, src1.xyz, src0.000
> 16: src0.xyz = temp[25]
>     DP3 temp[30].x, src0.xyz, src0.xyz
> 17: src0.xyz = temp[30]
>     REPL_ALPHA temp[31].x
>     RSQ, |src0.x|
> 18: src0.xyz = temp[25], src1.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src0.000
> 19: src0.xyz = temp[16], src1.xyz = const[10]
>     MAD temp[33].xy, src0.xy_, src1.xy_, src0.000
> 20: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 21: src0.xyz = temp[34], src0.w = const[5]
>     MAD temp[36].xyz, src0.xyz, src0.www, -src0.111
> 22: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[37].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[37]
>     MAX temp[38].x, src0.x__, src0.0__
> 24: src0.xyz = temp[29], src1.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.xxx, src0.000
> 25: src0.xyz = temp[25]
>     DP3 temp[40].x, src0.xyz, src0.xyz
> 26: TEX temp[41].x, temp[40].x___, 1D[0];
> 27: src0.xyz = temp[39], src1.xyz = temp[41]
>     MAD temp[43].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[43], src0.w = temp[43]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[0]
>     MAD temp[18].z, src0.__x, src0.111, src0.000
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13], src1.xyz = input[1]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[17].w, src1.z
>  2: src0.xyz = temp[14], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[17], src1.xyz = temp[17]
>     MAD temp[18].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[4]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[16], src1.xyz = const[10]
>     MAD temp[33].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[19].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 11: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 12: TEX temp[20].xyz, temp[19].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[34], src0.w = const[5] SEM_WAIT
>     MAD temp[36].xyz, src0.xyz, src0.www, -src0.111
> 14: src0.xyz = temp[27], src1.xyz = const[1]
>     MAD temp[29].xyz, src0.xyz, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = const[5]
>     DP3 temp[22].x, src0.xyz, src1.xyz
> 16: src0.xyz = temp[18], src1.xyz = temp[22]
>     MAD temp[23].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[23], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[25].xyz, srcp.xyz, src2.xxx, src0.000
> 18: src0.xyz = temp[25]
>     DP3 temp[40].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[25]
>     DP3 temp[30].x, src0.xyz, src0.xyz
> 20: src0.xyz = temp[30]
>     REPL_ALPHA temp[31].x
>     RSQ, |src0.x|
> 21: src0.xyz = temp[25], src1.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src0.000
> 22: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[37].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[37]
>     MAX temp[38].x, src0.x__, src0.0__
> 24: src0.xyz = temp[29], src1.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.xxx, src0.000
> 25: BEGIN_TEX;
> 26: TEX temp[41].x, temp[40].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 27: src0.xyz = temp[39], src1.xyz = temp[41] SEM_WAIT
>     MAD temp[43].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[43], src0.w = temp[43]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[0]
>     MAD temp[18].z, src0.__x, src0.111, src0.000
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13], src1.xyz = input[1]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[17].w, src1.z
>  2: src0.xyz = temp[14], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[1], src0.w = temp[17]
>     MAD temp[18].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[4]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  6: src0.xyz = temp[16], src1.xyz = const[10]
>     MAD temp[33].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[26].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[19].xy, src0.xy_, src1.xy_, src0.000
>  9: BEGIN_TEX;
> 10: TEX temp[34].xyz, temp[33].xy__, 2D[2];
> 11: TEX temp[27].xyz, temp[26].xy__, 2D[3];
> 12: TEX temp[20].xyz, temp[19].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[34], src0.w = const[5] SEM_WAIT
>     MAD temp[36].xyz, src0.xyz, src0.www, -src0.111
> 14: src0.xyz = temp[27], src1.xyz = const[1]
>     MAD temp[29].xyz, src0.xyz, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = const[5]
>     DP3 temp[22].x, src0.xyz, src1.xyz
> 16: src0.xyz = temp[18], src1.xyz = temp[22]
>     MAD temp[23].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[23], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[25].xyz, srcp.xyz, src2.xxx, src0.000
> 18: src0.xyz = temp[25]
>     DP3 temp[40].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[25]
>     DP3 temp[30].x, src0.xyz, src0.xyz
> 20: src0.xyz = temp[30]
>     REPL_ALPHA temp[31].x
>     RSQ, |src0.x|
> 21: src0.xyz = temp[25], src1.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src0.000
> 22: src0.xyz = temp[32], src1.xyz = temp[36]
>     DP3 temp[37].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[37]
>     MAX temp[38].x, src0.x__, src0.0__
> 24: src0.xyz = temp[29], src1.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src1.xxx, src0.000
> 25: BEGIN_TEX;
> 26: TEX temp[41].x, temp[40].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 27: src0.xyz = temp[39], src1.xyz = temp[41] SEM_WAIT
>     MAD temp[43].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[43], src0.w = temp[43]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[0], src0.w = input[1]
>     MAD temp[2].z, src0.__x, src0.111, src0.000
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0], src1.xyz = input[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.00_
>     RCP temp[0].w, src1.z
>  2: src0.xyz = temp[1], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = input[0], src0.w = temp[0]
>     MAD temp[2].xy, src0.xy_, src0.ww_, src0.000
>  4: src0.xyz = temp[1], src1.xyz = const[4]
>     MAD temp[0].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[1]
>     MAD temp[0].x, src0.x__, src0.11_, src0.00_
>  6: src0.xyz = temp[0], src1.xyz = const[10]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[0], src1.xyz = const[9]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[0], src1.xyz = const[8]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  9: BEGIN_TEX;
> 10: TEX temp[1].xyz, temp[1].xy__, 2D[2];
> 11: TEX temp[3].xyz, temp[3].xy__, 2D[3];
> 12: TEX temp[0].xyz, temp[0].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 13: src0.xyz = temp[1], src0.w = const[5] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src0.www, -src0.111
> 14: src0.xyz = temp[3], src1.xyz = const[1]
>     MAD temp[3].xyz, src0.xyz, src1.xyz, src0.000
> 15: src0.xyz = temp[0], src1.xyz = const[5]
>     DP3 temp[0].x, src0.xyz, src1.xyz
> 16: src0.xyz = temp[2], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 17: src0.xyz = temp[0], src1.xyz = const[3], src2.xyz = const[2], srcp.xyz = (src1 - src0)
>     MAD temp[0].xyz, srcp.xyz, src2.xxx, src0.000
> 18: src0.xyz = temp[0]
>     DP3 temp[2].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[0]
>     DP3 temp[2].y, src0.xyz, src0.xyz
> 20: src0.xyz = temp[2]
>     REPL_ALPHA temp[2].y
>     RSQ, |src0.y|
> 21: src0.xyz = temp[0], src1.xyz = temp[2]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[0], src1.xyz = temp[1]
>     DP3 temp[0].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[0]
>     MAX temp[0].x, src0.x__, src0.0__
> 24: src0.xyz = temp[3], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 25: BEGIN_TEX;
> 26: TEX temp[1].x, temp[2].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 27: src0.xyz = temp[0], src1.xyz = temp[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[0], src0.w = temp[43]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>r300compiler error: Too many hardware temporaries used.
>r300 FP: Compiler Error:
>Too many hardware temporaries used.
>Using a dummy shader instead.
>r300: Initial fragment program
>FRAG
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    0.0000,     1.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxy
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=32*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      0.0      1.0      0.0    op: 00050a94 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[6..7]
>DCL CONST[4..5]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>DCL TEMP[6], LOCAL
>DCL TEMP[7], LOCAL
>DCL TEMP[8], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     2.0000}
>IMM[1] FLT32 {   -1.0000,     0.0000,    10.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[7].xxxx, CONST[7].yyyy
>  2: TEX TEMP[1], TEMP[0].xyyy, SAMP[2], RECT
>  3: TEX TEMP[2], TEMP[0].xyyy, SAMP[1], RECT
>  4: DP3 TEMP[3].x, TEMP[2].xyzz, IMM[0].xyzz
>  5: MUL TEMP[3].xyz, IN[1].xyzz, TEMP[3].xxxx
>  6: ADD TEMP[4].xyz, CONST[6].xyzz, -TEMP[3].xyzz
>  7: MUL TEMP[4].xyz, TEMP[4].xyzz, CONST[5].xxxx
>  8: DP3 TEMP[5].x, TEMP[4].xyzz, TEMP[4].xyzz
>  9: RSQ TEMP[5].x, TEMP[5].xxxx
> 10: MUL TEMP[5].xyz, TEMP[4].xyzz, TEMP[5].xxxx
> 11: MAD TEMP[6].xyz, TEMP[1].xyzz, IMM[0].wwww, IMM[1].xxxx
> 12: DP3 TEMP[7].x, TEMP[6].xyzz, TEMP[6].xyzz
> 13: RSQ TEMP[7].x, TEMP[7].xxxx
> 14: MUL TEMP[6].xyz, TEMP[6].xyzz, TEMP[7].xxxx
> 15: MOV TEMP[3].xyz, -TEMP[3].xyzx
> 16: DP3 TEMP[7].x, TEMP[3].xyzz, TEMP[3].xyzz
> 17: RSQ TEMP[7].x, TEMP[7].xxxx
> 18: MAD TEMP[3].xyz, TEMP[3].xyzz, TEMP[7].xxxx, TEMP[5].xyzz
> 19: TEX TEMP[7].xyz, TEMP[0].xyyy, SAMP[3], RECT
> 20: MUL TEMP[7].xyz, TEMP[7].xyzz, CONST[4].xyzz
> 21: DP3 TEMP[5].x, TEMP[5].xyzz, TEMP[6].xyzz
> 22: MAX TEMP[5].x, TEMP[5].xxxx, IMM[1].yyyy
> 23: MUL TEMP[1].x, CONST[4].wwww, TEMP[1].wwww
> 24: DP3 TEMP[8].x, TEMP[3].xyzz, TEMP[3].xyzz
> 25: RSQ TEMP[8].x, TEMP[8].xxxx
> 26: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[8].xxxx
> 27: DP3_SAT TEMP[3].x, TEMP[3].xyzz, TEMP[6].xyzz
> 28: MUL TEMP[2].x, TEMP[2].wwww, IMM[1].zzzz
> 29: EX2 TEMP[2].x, TEMP[2].xxxx
> 30: ADD TEMP[2].x, TEMP[2].xxxx, IMM[0].xxxx
> 31: POW TEMP[2].x, TEMP[3].xxxx, TEMP[2].xxxx
> 32: MUL TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
> 33: MUL TEMP[1].xyz, TEMP[1].xxxx, CONST[4].xyzz
> 34: MAD TEMP[1].xyz, TEMP[7].xyzz, TEMP[5].xxxx, TEMP[1].xyzz
> 35: DP3 TEMP[2].x, TEMP[4].xyzz, TEMP[4].xyzz
> 36: TEX TEMP[2].x, TEMP[2].xxxx, SAMP[0], 1D
> 37: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[2].xxxx
> 38: MOV OUT[0], TEMP[1]
> 39: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
>  8: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
>  9: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 10: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 11: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 12: RSQ temp[5].x, temp[5].xxxx;
> 13: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 14: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 15: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 16: RSQ temp[7].x, temp[7].xxxx;
> 17: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 18: MOV temp[3].xyz, -temp[3].xyzx;
> 19: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 20: RSQ temp[7].x, temp[7].xxxx;
> 21: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 22: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 23: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 24: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 25: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 26: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 27: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 28: RSQ temp[8].x, temp[8].xxxx;
> 29: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 30: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 31: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 32: EX2 temp[2].x, temp[2].xxxx;
> 33: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 34: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 35: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 36: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 37: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 38: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 39: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 40: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 41: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
>  8: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
>  9: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 10: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 11: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 12: RSQ temp[5].x, temp[5].xxxx;
> 13: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 14: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 15: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 16: RSQ temp[7].x, temp[7].xxxx;
> 17: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 18: MOV temp[3].xyz, -temp[3].xyzx;
> 19: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 20: RSQ temp[7].x, temp[7].xxxx;
> 21: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 22: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 23: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 24: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 25: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 26: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 27: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 28: RSQ temp[8].x, temp[8].xxxx;
> 29: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 30: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 31: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 32: EX2 temp[2].x, temp[2].xxxx;
> 33: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 34: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 35: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 36: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 37: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 38: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 39: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 40: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 41: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
>  8: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
>  9: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 10: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 11: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 12: RSQ temp[5].x, temp[5].xxxx;
> 13: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 14: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 15: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 16: RSQ temp[7].x, temp[7].xxxx;
> 17: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 18: MOV temp[3].xyz, -temp[3].xyzx;
> 19: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 20: RSQ temp[7].x, temp[7].xxxx;
> 21: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 22: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 23: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 24: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 25: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 26: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 27: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 28: RSQ temp[8].x, temp[8].xxxx;
> 29: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 30: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 31: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 32: EX2 temp[2].x, temp[2].xxxx;
> 33: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 34: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 35: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 36: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 37: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 38: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 39: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 40: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 41: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
>  8: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
>  9: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 10: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 11: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 12: RSQ temp[5].x, temp[5].xxxx;
> 13: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 14: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 15: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 16: RSQ temp[7].x, temp[7].xxxx;
> 17: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 18: MOV temp[3].xyz, -temp[3].xyzx;
> 19: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 20: RSQ temp[7].x, temp[7].xxxx;
> 21: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 22: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 23: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 24: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 25: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 26: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 27: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 28: RSQ temp[8].x, temp[8].xxxx;
> 29: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 30: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 31: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 32: EX2 temp[2].x, temp[2].xxxx;
> 33: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 34: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 35: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 36: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 37: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 38: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 39: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 40: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 41: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1], temp[0].xyyy, RECT[2];
>  6: TEX temp[2], temp[0].xyyy, RECT[1];
>  7: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
>  8: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
>  9: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 10: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 11: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 12: RSQ temp[5].x, temp[5].xxxx;
> 13: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 14: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 15: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 16: RSQ temp[7].x, temp[7].xxxx;
> 17: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 18: MOV temp[3].xyz, -temp[3].xyzx;
> 19: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 20: RSQ temp[7].x, temp[7].xxxx;
> 21: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 22: TEX temp[7].xyz, temp[0].xyyy, RECT[3];
> 23: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 24: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 25: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 26: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 27: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 28: RSQ temp[8].x, temp[8].xxxx;
> 29: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 30: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 31: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 32: EX2 temp[2].x, temp[2].xxxx;
> 33: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 34: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 35: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 36: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 37: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 38: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 39: TEX temp[2].x, temp[2].xxxx, 1D[0];
> 40: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 41: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: MUL temp[10], temp[0].xyyy, const[12];
>  6: TEX temp[1], temp[10], 2D[2];
>  7: MUL temp[11], temp[0].xyyy, const[13];
>  8: TEX temp[2], temp[11], 2D[1];
>  9: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
> 10: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 11: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 12: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 13: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 14: RSQ temp[5].x, temp[5].xxxx;
> 15: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 16: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 18: RSQ temp[7].x, temp[7].xxxx;
> 19: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 20: MOV temp[3].xyz, -temp[3].xyzx;
> 21: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 22: RSQ temp[7].x, temp[7].xxxx;
> 23: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 24: MUL temp[12], temp[0].xyyy, const[14];
> 25: TEX temp[13], temp[12], 2D[3];
> 26: MOV temp[7].xyz, temp[13];
> 27: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 28: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 29: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 30: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 31: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 32: RSQ temp[8].x, temp[8].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 34: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 35: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 36: EX2 temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 38: POW temp[2].x, temp[3].xxxx, temp[2].xxxx;
> 39: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 40: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 41: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 42: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 43: TEX temp[14], temp[2].xxxx, 1D[0];
> 44: MOV temp[2].x, temp[14];
> 45: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 46: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].wwww;
>  1: MUL temp[9].xyz, input[0], temp[9].wwww;
>  2: MAD temp[9].xyz, temp[9].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[9];
>  4: MAD temp[0].y, temp[9], const[7].xxxx, const[7].yyyy;
>  5: MUL temp[10], temp[0].xyyy, const[12];
>  6: TEX temp[1], temp[10], 2D[2];
>  7: MUL temp[11], temp[0].xyyy, const[13];
>  8: TEX temp[2], temp[11], 2D[1];
>  9: DP3 temp[3].x, temp[2].xyzz, const[8].xyzz;
> 10: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 11: ADD temp[4].xyz, const[6].xyzz, -temp[3].xyzz;
> 12: MUL temp[4].xyz, temp[4].xyzz, const[5].xxxx;
> 13: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 14: RSQ temp[5].x, |temp[5].xxxx|;
> 15: MUL temp[5].xyz, temp[4].xyzz, temp[5].xxxx;
> 16: MAD temp[6].xyz, temp[1].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[7].x, temp[6].xyzz, temp[6].xyzz;
> 18: RSQ temp[7].x, |temp[7].xxxx|;
> 19: MUL temp[6].xyz, temp[6].xyzz, temp[7].xxxx;
> 20: MOV temp[3].xyz, -temp[3].xyzx;
> 21: DP3 temp[7].x, temp[3].xyzz, temp[3].xyzz;
> 22: RSQ temp[7].x, |temp[7].xxxx|;
> 23: MAD temp[3].xyz, temp[3].xyzz, temp[7].xxxx, temp[5].xyzz;
> 24: MUL temp[12], temp[0].xyyy, const[14];
> 25: TEX temp[13], temp[12], 2D[3];
> 26: MOV temp[7].xyz, temp[13];
> 27: MUL temp[7].xyz, temp[7].xyzz, const[4].xyzz;
> 28: DP3 temp[5].x, temp[5].xyzz, temp[6].xyzz;
> 29: MAX temp[5].x, temp[5].xxxx, const[9].yyyy;
> 30: MUL temp[1].x, const[4].wwww, temp[1].wwww;
> 31: DP3 temp[8].x, temp[3].xyzz, temp[3].xyzz;
> 32: RSQ temp[8].x, |temp[8].xxxx|;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[8].xxxx;
> 34: DP3_SAT temp[3].x, temp[3].xyzz, temp[6].xyzz;
> 35: MUL temp[2].x, temp[2].wwww, const[9].zzzz;
> 36: EX2 temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, temp[2].xxxx, const[8].xxxx;
> 38: LG2 temp[15].w, temp[3].xxxx;
> 39: MUL temp[15].w, temp[15].wwww, temp[2].xxxx;
> 40: EX2 temp[2].x, temp[15].wwww;
> 41: MUL temp[1].x, temp[1].xxxx, temp[2].xxxx;
> 42: MUL temp[1].xyz, temp[1].xxxx, const[4].xyzz;
> 43: MAD temp[1].xyz, temp[7].xyzz, temp[5].xxxx, temp[1].xyzz;
> 44: DP3 temp[2].x, temp[4].xyzz, temp[4].xyzz;
> 45: TEX temp[14], temp[2].xxxx, 1D[0];
> 46: MOV temp[2].x, temp[14];
> 47: MUL temp[1].xyz, temp[1].xyzz, temp[2].xxxx;
> 48: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[9].ww__;
>  2: MAD temp[9].xy, temp[9].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[0].x, temp[9].x___;
>  4: MAD temp[0].y, temp[9]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[10].xy, temp[0].xy__, const[12].xy__;
>  6: TEX temp[1], temp[10].xy__, 2D[2];
>  7: MUL temp[11].xy, temp[0].xy__, const[13].xy__;
>  8: TEX temp[2], temp[11].xy__, 2D[1];
>  9: DP3 temp[3].x, temp[2].xyz_, const[8].xyz_;
> 10: MUL temp[3].xyz, input[1].xyz_, temp[3].xxx_;
> 11: ADD temp[4].xyz, const[6].xyz_, -temp[3].xyz_;
> 12: MUL temp[4].xyz, temp[4].xyz_, const[5].xxx_;
> 13: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 14: RSQ temp[5].x, |temp[5].x___|;
> 15: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 16: MAD temp[6].xyz, temp[1].xyz_, const[8].www_, const[9].xxx_;
> 17: DP3 temp[7].x, temp[6].xyz_, temp[6].xyz_;
> 18: RSQ temp[7].x, |temp[7].x___|;
> 19: MUL temp[6].xyz, temp[6].xyz_, temp[7].xxx_;
> 20: MOV temp[3].xyz, -temp[3].xyz_;
> 21: DP3 temp[7].x, temp[3].xyz_, temp[3].xyz_;
> 22: RSQ temp[7].x, |temp[7].x___|;
> 23: MAD temp[3].xyz, temp[3].xyz_, temp[7].xxx_, temp[5].xyz_;
> 24: MUL temp[12].xy, temp[0].xy__, const[14].xy__;
> 25: TEX temp[13].xyz, temp[12].xy__, 2D[3];
> 26: MOV temp[7].xyz, temp[13].xyz_;
> 27: MUL temp[7].xyz, temp[7].xyz_, const[4].xyz_;
> 28: DP3 temp[5].x, temp[5].xyz_, temp[6].xyz_;
> 29: MAX temp[5].x, temp[5].x___, const[9].y___;
> 30: MUL temp[1].x, const[4].w___, temp[1].w___;
> 31: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 32: RSQ temp[8].x, |temp[8].x___|;
> 33: MUL temp[3].xyz, temp[3].xyz_, temp[8].xxx_;
> 34: DP3_SAT temp[3].x, temp[3].xyz_, temp[6].xyz_;
> 35: MUL temp[2].x, temp[2].w___, const[9].z___;
> 36: EX2 temp[2].x, temp[2].x___;
> 37: ADD temp[2].x, temp[2].x___, const[8].x___;
> 38: LG2 temp[15].w, temp[3].___x;
> 39: MUL temp[15].w, temp[15].___w, temp[2].___x;
> 40: EX2 temp[2].x, temp[15].w___;
> 41: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 42: MUL temp[1].xyz, temp[1].xxx_, const[4].xyz_;
> 43: MAD temp[1].xyz, temp[7].xyz_, temp[5].xxx_, temp[1].xyz_;
> 44: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 45: TEX temp[14].x, temp[2].x___, 1D[0];
> 46: MOV temp[2].x, temp[14].x___;
> 47: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 48: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[9].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[9].ww__;
>  2: MAD temp[9].xy, temp[9].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[0].x, temp[9].x___;
>  4: MAD temp[0].y, temp[9]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[10].xy, temp[0].xy__, const[12].xy__;
>  6: TEX temp[1], temp[10].xy__, 2D[2];
>  7: MUL temp[11].xy, temp[0].xy__, const[13].xy__;
>  8: TEX temp[2], temp[11].xy__, 2D[1];
>  9: DP3 temp[3].x, temp[2].xyz_, const[8].xyz_;
> 10: MUL temp[3].xyz, input[1].xyz_, temp[3].xxx_;
> 11: ADD temp[4].xyz, const[6].xyz_, -temp[3].xyz_;
> 12: MUL temp[4].xyz, temp[4].xyz_, const[5].xxx_;
> 13: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 14: RSQ temp[5].x, |temp[5].x___|;
> 15: MUL temp[5].xyz, temp[4].xyz_, temp[5].xxx_;
> 16: MAD temp[6].xyz, temp[1].xyz_, const[8].www_, const[9].xxx_;
> 17: DP3 temp[7].x, temp[6].xyz_, temp[6].xyz_;
> 18: RSQ temp[7].x, |temp[7].x___|;
> 19: MUL temp[6].xyz, temp[6].xyz_, temp[7].xxx_;
> 20: MOV temp[3].xyz, -temp[3].xyz_;
> 21: DP3 temp[7].x, temp[3].xyz_, temp[3].xyz_;
> 22: RSQ temp[7].x, |temp[7].x___|;
> 23: MAD temp[3].xyz, temp[3].xyz_, temp[7].xxx_, temp[5].xyz_;
> 24: MUL temp[12].xy, temp[0].xy__, const[14].xy__;
> 25: TEX temp[13].xyz, temp[12].xy__, 2D[3];
> 26: MOV temp[7].xyz, temp[13].xyz_;
> 27: MUL temp[7].xyz, temp[7].xyz_, const[4].xyz_;
> 28: DP3 temp[5].x, temp[5].xyz_, temp[6].xyz_;
> 29: MAX temp[5].x, temp[5].x___, const[9].y___;
> 30: MUL temp[1].x, const[4].w___, temp[1].w___;
> 31: DP3 temp[8].x, temp[3].xyz_, temp[3].xyz_;
> 32: RSQ temp[8].x, |temp[8].x___|;
> 33: MUL temp[3].xyz, temp[3].xyz_, temp[8].xxx_;
> 34: DP3_SAT temp[3].x, temp[3].xyz_, temp[6].xyz_;
> 35: MUL temp[2].x, temp[2].w___, const[9].z___;
> 36: EX2 temp[2].x, temp[2].x___;
> 37: ADD temp[2].x, temp[2].x___, const[8].x___;
> 38: LG2 temp[15].w, temp[3].___x;
> 39: MUL temp[15].w, temp[15].___w, temp[2].___x;
> 40: EX2 temp[2].x, temp[15].w___;
> 41: MUL temp[1].x, temp[1].x___, temp[2].x___;
> 42: MUL temp[1].xyz, temp[1].xxx_, const[4].xyz_;
> 43: MAD temp[1].xyz, temp[7].xyz_, temp[5].xxx_, temp[1].xyz_;
> 44: DP3 temp[2].x, temp[4].xyz_, temp[4].xyz_;
> 45: TEX temp[14].x, temp[2].x___, 1D[0];
> 46: MOV temp[2].x, temp[14].x___;
> 47: MUL temp[1].xyz, temp[1].xyz_, temp[2].xxx_;
> 48: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[12].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[13].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: DP3 temp[24].x, temp[23].xyz_, const[8].xyz_;
> 10: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 11: ADD temp[26].xyz, const[6].xyz_, -temp[25].xyz_;
> 12: MUL temp[27].xyz, temp[26].xyz_, const[5].xxx_;
> 13: DP3 temp[28].x, temp[27].xyz_, temp[27].xyz_;
> 14: RSQ temp[29].x, |temp[28].x___|;
> 15: MUL temp[30].xyz, temp[27].xyz_, temp[29].xxx_;
> 16: MAD temp[31].xyz, temp[21].xyz_, const[8].www_, const[9].xxx_;
> 17: DP3 temp[32].x, temp[31].xyz_, temp[31].xyz_;
> 18: RSQ temp[33].x, |temp[32].x___|;
> 19: MUL temp[34].xyz, temp[31].xyz_, temp[33].xxx_;
> 20: MOV temp[35].xyz, -temp[25].xyz_;
> 21: DP3 temp[36].x, temp[35].xyz_, temp[35].xyz_;
> 22: RSQ temp[37].x, |temp[36].x___|;
> 23: MAD temp[38].xyz, temp[35].xyz_, temp[37].xxx_, temp[30].xyz_;
> 24: MUL temp[39].xy, temp[19].xy__, const[14].xy__;
> 25: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 26: MOV temp[41].xyz, temp[40].xyz_;
> 27: MUL temp[42].xyz, temp[41].xyz_, const[4].xyz_;
> 28: DP3 temp[43].x, temp[30].xyz_, temp[34].xyz_;
> 29: MAX temp[44].x, temp[43].x___, const[9].y___;
> 30: MUL temp[45].x, const[4].w___, temp[21].w___;
> 31: DP3 temp[46].x, temp[38].xyz_, temp[38].xyz_;
> 32: RSQ temp[47].x, |temp[46].x___|;
> 33: MUL temp[48].xyz, temp[38].xyz_, temp[47].xxx_;
> 34: DP3_SAT temp[49].x, temp[48].xyz_, temp[34].xyz_;
> 35: MUL temp[50].x, temp[23].w___, const[9].z___;
> 36: EX2 temp[51].x, temp[50].x___;
> 37: ADD temp[52].x, temp[51].x___, const[8].x___;
> 38: LG2 temp[53].w, temp[49].___x;
> 39: MUL temp[54].w, temp[53].___w, temp[52].___x;
> 40: EX2 temp[55].x, temp[54].w___;
> 41: MUL temp[56].x, temp[45].x___, temp[55].x___;
> 42: MUL temp[57].xyz, temp[56].xxx_, const[4].xyz_;
> 43: MAD temp[58].xyz, temp[42].xyz_, temp[44].xxx_, temp[57].xyz_;
> 44: DP3 temp[59].x, temp[27].xyz_, temp[27].xyz_;
> 45: TEX temp[60].x, temp[59].x___, 1D[0];
> 46: MOV temp[61].x, temp[60].x___;
> 47: MUL temp[21].xyz, temp[58].xyz_, temp[61].xxx_;
> 48: MOV output[0], temp[21];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[12].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[13].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: DP3 temp[24].x, temp[23].xyz_, const[8].xyz_;
> 10: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 11: MUL temp[27].xyz, (const[6] - temp[25]).xyz_, const[5].xxx_;
> 12: DP3 temp[28].x, temp[27].xyz_, temp[27].xyz_;
> 13: RSQ temp[29].x, |temp[28].x___|;
> 14: MUL temp[30].xyz, temp[27].xyz_, temp[29].xxx_;
> 15: MAD temp[31].xyz, temp[21].xyz_, const[8].www_, none.-1-1-1_;
> 16: DP3 temp[32].x, temp[31].xyz_, temp[31].xyz_;
> 17: RSQ temp[33].x, |temp[32].x___|;
> 18: MUL temp[34].xyz, temp[31].xyz_, temp[33].xxx_;
> 19: DP3 temp[36].x, temp[25].-x-y-z_, temp[25].-x-y-z_;
> 20: RSQ temp[37].x, |temp[36].x___|;
> 21: MAD temp[38].xyz, temp[25].-x-y-z_, temp[37].xxx_, temp[30].xyz_;
> 22: MUL temp[39].xy, temp[19].xy__, const[14].xy__;
> 23: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 24: MUL temp[42].xyz, temp[40].xyz_, const[4].xyz_;
> 25: DP3 temp[43].x, temp[30].xyz_, temp[34].xyz_;
> 26: MAX temp[44].x, temp[43].x___, none.0___;
> 27: MUL temp[45].x, const[4].w___, temp[21].w___;
> 28: DP3 temp[46].x, temp[38].xyz_, temp[38].xyz_;
> 29: RSQ temp[47].x, |temp[46].x___|;
> 30: MUL temp[48].xyz, temp[38].xyz_, temp[47].xxx_;
> 31: DP3_SAT temp[49].x, temp[48].xyz_, temp[34].xyz_;
> 32: MUL temp[50].x, temp[23].w___, const[9].z___;
> 33: EX2 temp[51].x, temp[50].x___;
> 34: ADD temp[52].x, temp[51].x___, none.1___;
> 35: LG2 temp[53].w, temp[49].___x;
> 36: MUL temp[54].w, temp[53].___w, temp[52].___x;
> 37: EX2 temp[55].x, temp[54].w___;
> 38: MUL temp[56].x, temp[45].x___, temp[55].x___;
> 39: MUL temp[57].xyz, temp[56].xxx_, const[4].xyz_;
> 40: MAD temp[58].xyz, temp[42].xyz_, temp[44].xxx_, temp[57].xyz_;
> 41: DP3 temp[59].x, temp[27].xyz_, temp[27].xyz_;
> 42: TEX temp[60].x, temp[59].x___, 1D[0];
> 43: MUL temp[21].xyz, temp[58].xyz_, temp[60].xxx_;
> 44: MOV output[0], temp[21];
>CONST[8] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[9] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[12].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[13].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: DP3 temp[24].x, temp[23].xyz_, const[8].xyz_;
> 10: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 11: MUL temp[27].xyz, (const[6] - temp[25]).xyz_, const[5].xxx_;
> 12: DP3 temp[28].x, temp[27].xyz_, temp[27].xyz_;
> 13: RSQ temp[29].x, |temp[28].x___|;
> 14: MUL temp[30].xyz, temp[27].xyz_, temp[29].xxx_;
> 15: MAD temp[31].xyz, temp[21].xyz_, const[8].www_, none.-1-1-1_;
> 16: DP3 temp[32].x, temp[31].xyz_, temp[31].xyz_;
> 17: RSQ temp[33].x, |temp[32].x___|;
> 18: MUL temp[34].xyz, temp[31].xyz_, temp[33].xxx_;
> 19: DP3 temp[36].x, temp[25].-x-y-z_, temp[25].-x-y-z_;
> 20: RSQ temp[37].x, |temp[36].x___|;
> 21: MAD temp[38].xyz, temp[25].-x-y-z_, temp[37].xxx_, temp[30].xyz_;
> 22: MUL temp[39].xy, temp[19].xy__, const[14].xy__;
> 23: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 24: MUL temp[42].xyz, temp[40].xyz_, const[4].xyz_;
> 25: DP3 temp[43].x, temp[30].xyz_, temp[34].xyz_;
> 26: MAX temp[44].x, temp[43].x___, none.0___;
> 27: MUL temp[45].x, const[4].w___, temp[21].w___;
> 28: DP3 temp[46].x, temp[38].xyz_, temp[38].xyz_;
> 29: RSQ temp[47].x, |temp[46].x___|;
> 30: MUL temp[48].xyz, temp[38].xyz_, temp[47].xxx_;
> 31: DP3_SAT temp[49].x, temp[48].xyz_, temp[34].xyz_;
> 32: MUL temp[50].x, temp[23].w___, const[9].z___;
> 33: EX2 temp[51].x, temp[50].x___;
> 34: ADD temp[52].x, temp[51].x___, none.1___;
> 35: LG2 temp[53].w, temp[49].___x;
> 36: MUL temp[54].w, temp[53].___w, temp[52].___x;
> 37: EX2 temp[55].x, temp[54].w___;
> 38: MUL temp[56].x, temp[45].x___, temp[55].x___;
> 39: MUL temp[57].xyz, temp[56].xxx_, const[4].xyz_;
> 40: MAD temp[58].xyz, temp[42].xyz_, temp[44].xxx_, temp[57].xyz_;
> 41: DP3 temp[59].x, temp[27].xyz_, temp[27].xyz_;
> 42: TEX temp[60].x, temp[59].x___, 1D[0];
> 43: MUL temp[21].xyz, temp[58].xyz_, temp[60].xxx_;
> 44: MOV output[0], temp[21];
>CONST[4] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[5] = {    -1.0000     0.0000    10.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[16].w, input[0].___w;
>  1: MUL temp[17].xy, input[0].xy__, temp[16].ww__;
>  2: MAD temp[18].xy, temp[17].xy__, const[6].xy__, const[7].xy__;
>  3: MOV temp[19].x, temp[18].x___;
>  4: MAD temp[19].y, temp[18]._y__, const[3]._x__, const[3]._y__;
>  5: MUL temp[20].xy, temp[19].xy__, const[8].xy__;
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: MUL temp[22].xy, temp[19].xy__, const[9].xy__;
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: DP3 temp[24].x, temp[23].xyz_, const[4].xyz_;
> 10: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 11: MUL temp[27].xyz, (const[2] - temp[25]).xyz_, const[1].xxx_;
> 12: DP3 temp[28].x, temp[27].xyz_, temp[27].xyz_;
> 13: RSQ temp[29].x, |temp[28].x___|;
> 14: MUL temp[30].xyz, temp[27].xyz_, temp[29].xxx_;
> 15: MAD temp[31].xyz, temp[21].xyz_, const[4].www_, none.-1-1-1_;
> 16: DP3 temp[32].x, temp[31].xyz_, temp[31].xyz_;
> 17: RSQ temp[33].x, |temp[32].x___|;
> 18: MUL temp[34].xyz, temp[31].xyz_, temp[33].xxx_;
> 19: DP3 temp[36].x, temp[25].-x-y-z_, temp[25].-x-y-z_;
> 20: RSQ temp[37].x, |temp[36].x___|;
> 21: MAD temp[38].xyz, temp[25].-x-y-z_, temp[37].xxx_, temp[30].xyz_;
> 22: MUL temp[39].xy, temp[19].xy__, const[10].xy__;
> 23: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 24: MUL temp[42].xyz, temp[40].xyz_, const[0].xyz_;
> 25: DP3 temp[43].x, temp[30].xyz_, temp[34].xyz_;
> 26: MAX temp[44].x, temp[43].x___, none.0___;
> 27: MUL temp[45].x, const[0].w___, temp[21].w___;
> 28: DP3 temp[46].x, temp[38].xyz_, temp[38].xyz_;
> 29: RSQ temp[47].x, |temp[46].x___|;
> 30: MUL temp[48].xyz, temp[38].xyz_, temp[47].xxx_;
> 31: DP3_SAT temp[49].x, temp[48].xyz_, temp[34].xyz_;
> 32: MUL temp[50].x, temp[23].w___, const[5].z___;
> 33: EX2 temp[51].x, temp[50].x___;
> 34: ADD temp[52].x, temp[51].x___, none.1___;
> 35: LG2 temp[53].w, temp[49].___x;
> 36: MUL temp[54].w, temp[53].___w, temp[52].___x;
> 37: EX2 temp[55].x, temp[54].w___;
> 38: MUL temp[56].x, temp[45].x___, temp[55].x___;
> 39: MUL temp[57].xyz, temp[56].xxx_, const[0].xyz_;
> 40: MAD temp[58].xyz, temp[42].xyz_, temp[44].xxx_, temp[57].xyz_;
> 41: DP3 temp[59].x, temp[27].xyz_, temp[27].xyz_;
> 42: TEX temp[60].x, temp[59].x___, 1D[0];
> 43: MUL temp[21].xyz, temp[58].xyz_, temp[60].xxx_;
> 44: MOV output[0], temp[21];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[17], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[18], src1.xyz = const[3]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[19], src1.xyz = const[8]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  6: TEX temp[21], temp[20].xy__, 2D[2];
>  7: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: TEX temp[23], temp[22].xy__, 2D[1];
>  9: src0.xyz = temp[23], src1.xyz = const[4]
>     DP3 temp[24].x, src0.xyz, src1.xyz
> 10: src0.xyz = input[1], src1.xyz = temp[24]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 11: src0.xyz = temp[25], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[27].xyz, srcp.xyz, src2.xxx, src0.000
> 12: src0.xyz = temp[27]
>     DP3 temp[28].x, src0.xyz, src0.xyz
> 13: src0.xyz = temp[28]
>     REPL_ALPHA temp[29].x
>     RSQ, |src0.x|
> 14: src0.xyz = temp[27], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
> 15: src0.xyz = temp[21], src0.w = const[4]
>     MAD temp[31].xyz, src0.xyz, src0.www, -src0.111
> 16: src0.xyz = temp[31]
>     DP3 temp[32].x, src0.xyz, src0.xyz
> 17: src0.xyz = temp[32]
>     REPL_ALPHA temp[33].x
>     RSQ, |src0.x|
> 18: src0.xyz = temp[31], src1.xyz = temp[33]
>     MAD temp[34].xyz, src0.xyz, src1.xxx, src0.000
> 19: src0.xyz = temp[25]
>     DP3 temp[36].x, -src0.xyz, -src0.xyz
> 20: src0.xyz = temp[36]
>     REPL_ALPHA temp[37].x
>     RSQ, |src0.x|
> 21: src0.xyz = temp[25], src1.xyz = temp[37], src2.xyz = temp[30]
>     MAD temp[38].xyz, -src0.xyz, src1.xxx, src2.xyz
> 22: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[39].xy, src0.xy_, src1.xy_, src0.000
> 23: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 24: src0.xyz = temp[40], src1.xyz = const[0]
>     MAD temp[42].xyz, src0.xyz, src1.xyz, src0.000
> 25: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[43].x, src0.xyz, src1.xyz
> 26: src0.xyz = temp[43]
>     MAX temp[44].x, src0.x__, src0.0__
> 27: src0.w = const[0], src1.w = temp[21]
>     MAD temp[45].x, src0.w__, src1.w__, src0.000
> 28: src0.xyz = temp[38]
>     DP3 temp[46].x, src0.xyz, src0.xyz
> 29: src0.xyz = temp[46]
>     REPL_ALPHA temp[47].x
>     RSQ, |src0.x|
> 30: src0.xyz = temp[38], src1.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 31: src0.xyz = temp[48], src1.xyz = temp[34]
>     DP3_SAT temp[49].x, src0.xyz, src1.xyz
> 32: src0.xyz = const[5], src0.w = temp[23]
>     MAD temp[50].x, src0.w__, src0.z__, src0.000
> 33: src0.xyz = temp[50]
>     REPL_ALPHA temp[51].x
>     EX2, src0.x
> 34: src0.xyz = temp[51]
>     MAD temp[52].x, src0.x__, src0.111, src0.1__
> 35: src0.xyz = temp[49]
>     LG2 temp[53].w, src0.x
> 36: src0.xyz = temp[52], src0.w = temp[53]
>     MAD temp[54].w, src0.w, src0.x, src0.0
> 37: src0.w = temp[54]
>     REPL_ALPHA temp[55].x
>     EX2, src0.w
> 38: src0.xyz = temp[45], src1.xyz = temp[55]
>     MAD temp[56].x, src0.x__, src1.x__, src0.000
> 39: src0.xyz = temp[56], src1.xyz = const[0]
>     MAD temp[57].xyz, src0.xxx, src1.xyz, src0.000
> 40: src0.xyz = temp[42], src1.xyz = temp[44], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src2.xyz
> 41: src0.xyz = temp[27]
>     DP3 temp[59].x, src0.xyz, src0.xyz
> 42: TEX temp[60].x, temp[59].x___, 1D[0];
> 43: src0.xyz = temp[58], src1.xyz = temp[60]
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[17], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[18], src1.xyz = const[3]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[39].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[19], src1.xyz = const[8]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 10: TEX temp[23], temp[22].xy__, 2D[1];
> 11: TEX temp[21], temp[20].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[23], src0.w = const[0], src1.xyz = const[4], src1.w = temp[21] SEM_WAIT
>     DP3 temp[24].x, src0.xyz, src1.xyz
>     MAD temp[45].w, src0.w, src1.w, src0.0
> 13: src0.xyz = temp[40], src0.w = temp[23], src1.xyz = const[0], src2.xyz = const[5]
>     MAD temp[42].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[50].w, src0.w, src2.z, src0.0
> 14: src0.xyz = temp[21], src0.w = const[4]
>     MAD temp[31].xyz, src0.xyz, src0.www, -src0.111
> 15: src0.xyz = input[1], src1.xyz = temp[24]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[50], src0.w = temp[50]
>     REPL_ALPHA temp[51].x
>     EX2, src0.w
> 17: src0.xyz = temp[31], src1.xyz = temp[51]
>     DP3 temp[32].x, src0.xyz, src0.xyz
>     MAD temp[52].w, src1.x, src0.1, src0.1
> 18: src0.xyz = temp[25], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[27].xyz, srcp.xyz, src2.xxx, src0.000
> 19: src0.xyz = temp[25]
>     DP3 temp[36].x, -src0.xyz, -src0.xyz
> 20: src0.xyz = temp[27], src1.xyz = temp[36]
>     DP3 temp[28].x, src0.xyz, src0.xyz
>     RSQ temp[37].w, |src1.x|
> 21: src0.xyz = temp[27], src1.xyz = temp[32]
>     DP3 temp[59].x, src0.xyz, src0.xyz
>     RSQ temp[33].w, |src1.x|
> 22: src0.xyz = temp[31], src0.w = temp[33], src1.xyz = temp[33], src2.xyz = temp[28]
>     MAD temp[34].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[29].w, |src2.x|
> 23: src0.xyz = temp[27], src0.w = temp[29], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src0.www, src0.000
> 24: src0.xyz = temp[25], src0.w = temp[37], src1.xyz = temp[37], src2.xyz = temp[30]
>     MAD temp[38].xyz, -src0.xyz, src0.www, src2.xyz
> 25: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[43].x, src0.xyz, src1.xyz
> 26: src0.xyz = temp[38], src1.xyz = temp[43]
>     DP3 temp[46].x, src0.xyz, src0.xyz
>     MAX temp[44].w, src1.x, src0.0
> 27: src0.xyz = temp[46]
>     REPL_ALPHA temp[47].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[38], src1.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[48], src1.xyz = temp[34]
>     DP3_SAT temp[49].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[49]
>     LG2 temp[53].w, src0.x
> 31: src0.xyz = temp[52], src0.w = temp[53], src1.w = temp[52]
>     MAD temp[54].w, src0.w, src1.w, src0.0
> 32: src0.w = temp[54]
>     REPL_ALPHA temp[55].x
>     EX2, src0.w
> 33: src0.xyz = temp[45], src0.w = temp[45], src1.xyz = temp[55]
>     MAD temp[56].x, src0.w__, src1.x__, src0.000
> 34: src0.xyz = temp[56], src1.xyz = const[0]
>     MAD temp[57].xyz, src0.xxx, src1.xyz, src0.000
> 35: src0.xyz = temp[42], src0.w = temp[44], src1.xyz = temp[44], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src0.www, src2.xyz
> 36: BEGIN_TEX;
> 37: TEX temp[60].x, temp[59].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 38: src0.xyz = temp[58], src1.xyz = temp[60] SEM_WAIT
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 39: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[16].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[16]
>     MAD temp[17].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[17], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[18].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[18], src1.xyz = const[3]
>     MAD temp[19].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[18]
>     MAD temp[19].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[19], src1.xyz = const[10]
>     MAD temp[39].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = temp[19], src1.xyz = const[9]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[19], src1.xyz = const[8]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[40].xyz, temp[39].xy__, 2D[3];
> 10: TEX temp[23], temp[22].xy__, 2D[1];
> 11: TEX temp[21], temp[20].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[23], src0.w = const[0], src1.xyz = const[4], src1.w = temp[21] SEM_WAIT
>     DP3 temp[24].x, src0.xyz, src1.xyz
>     MAD temp[45].w, src0.w, src1.w, src0.0
> 13: src0.xyz = temp[40], src0.w = temp[23], src1.xyz = const[0], src2.xyz = const[5]
>     MAD temp[42].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[50].w, src0.w, src2.z, src0.0
> 14: src0.xyz = temp[21], src0.w = const[4]
>     MAD temp[31].xyz, src0.xyz, src0.www, -src0.111
> 15: src0.xyz = input[1], src1.xyz = temp[24]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.w = temp[50]
>     REPL_ALPHA temp[51].x
>     EX2, src0.w
> 17: src0.xyz = temp[31], src1.xyz = temp[51]
>     DP3 temp[32].x, src0.xyz, src0.xyz
>     MAD temp[52].w, src1.x, src0.1, src0.1
> 18: src0.xyz = temp[25], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[27].xyz, srcp.xyz, src2.xxx, src0.000
> 19: src0.xyz = temp[25]
>     DP3 temp[36].x, -src0.xyz, -src0.xyz
> 20: src0.xyz = temp[27], src1.xyz = temp[36]
>     DP3 temp[28].x, src0.xyz, src0.xyz
>     RSQ temp[37].w, |src1.x|
> 21: src0.xyz = temp[27], src1.xyz = temp[32]
>     DP3 temp[59].x, src0.xyz, src0.xyz
>     RSQ temp[33].w, |src1.x|
> 22: src0.xyz = temp[31], src0.w = temp[33], src2.xyz = temp[28]
>     MAD temp[34].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[29].w, |src2.x|
> 23: src0.xyz = temp[27], src0.w = temp[29]
>     MAD temp[30].xyz, src0.xyz, src0.www, src0.000
> 24: src0.xyz = temp[25], src0.w = temp[37], src2.xyz = temp[30]
>     MAD temp[38].xyz, -src0.xyz, src0.www, src2.xyz
> 25: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[43].x, src0.xyz, src1.xyz
> 26: src0.xyz = temp[38], src1.xyz = temp[43]
>     DP3 temp[46].x, src0.xyz, src0.xyz
>     MAX temp[44].w, src1.x, src0.0
> 27: src0.xyz = temp[46]
>     REPL_ALPHA temp[47].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[38], src1.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[48], src1.xyz = temp[34]
>     DP3_SAT temp[49].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[49]
>     LG2 temp[53].w, src0.x
> 31: src0.w = temp[53], src1.w = temp[52]
>     MAD temp[54].w, src0.w, src1.w, src0.0
> 32: src0.w = temp[54]
>     REPL_ALPHA temp[55].x
>     EX2, src0.w
> 33: src0.w = temp[45], src1.xyz = temp[55]
>     MAD temp[56].x, src0.w__, src1.x__, src0.000
> 34: src0.xyz = temp[56], src1.xyz = const[0]
>     MAD temp[57].xyz, src0.xxx, src1.xyz, src0.000
> 35: src0.xyz = temp[42], src0.w = temp[44], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src0.www, src2.xyz
> 36: BEGIN_TEX;
> 37: TEX temp[60].x, temp[59].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 38: src0.xyz = temp[58], src1.xyz = temp[60] SEM_WAIT
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 39: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.w = input[1]
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.00_
>  2: src0.xyz = temp[1], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[1], src1.xyz = const[3]
>     MAD temp[2].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[1]
>     MAD temp[2].x, src0.x__, src0.11_, src0.00_
>  5: src0.xyz = temp[2], src1.xyz = const[10]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  6: src0.xyz = temp[2], src1.xyz = const[9]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[2], src1.xyz = const[8]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.00_
>  8: BEGIN_TEX;
>  9: TEX temp[1].xyz, temp[1].xy__, 2D[3];
> 10: TEX temp[3], temp[3].xy__, 2D[1];
> 11: TEX temp[2], temp[2].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[3], src0.w = const[0], src1.xyz = const[4], src1.w = temp[2] SEM_WAIT
>     DP3 temp[4].x, src0.xyz, src1.xyz
>     MAD temp[0].w, src0.w, src1.w, src0.0
> 13: src0.xyz = temp[1], src0.w = temp[3], src1.xyz = const[0], src2.xyz = const[5]
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, src0.w, src2.z, src0.0
> 14: src0.xyz = temp[2], src0.w = const[4]
>     MAD temp[3].xyz, src0.xyz, src0.www, -src0.111
> 15: src0.xyz = input[0], src1.xyz = temp[4]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.w = temp[1]
>     REPL_ALPHA temp[4].x
>     EX2, src0.w
> 17: src0.xyz = temp[3], src1.xyz = temp[4]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     MAD temp[1].w, src1.x, src0.1, src0.1
> 18: src0.xyz = temp[0], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[5].xyz, srcp.xyz, src2.xxx, src0.000
> 19: src0.xyz = temp[0]
>     DP3 temp[4].y, -src0.xyz, -src0.xyz
> 20: src0.xyz = temp[5], src1.xyz = temp[4]
>     DP3 temp[4].y, src0.xyz, src0.xyz
>     RSQ temp[3].w, |src1.y|
> 21: src0.xyz = temp[5], src1.xyz = temp[4]
>     DP3 temp[4].x, src0.xyz, src0.xyz
>     RSQ temp[4].w, |src1.x|
> 22: src0.xyz = temp[3], src0.w = temp[4], src2.xyz = temp[4]
>     MAD temp[3].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[4].w, |src2.y|
> 23: src0.xyz = temp[5], src0.w = temp[4]
>     MAD temp[5].xyz, src0.xyz, src0.www, src0.000
> 24: src0.xyz = temp[0], src0.w = temp[3], src2.xyz = temp[5]
>     MAD temp[0].xyz, -src0.xyz, src0.www, src2.xyz
> 25: src0.xyz = temp[5], src1.xyz = temp[3]
>     DP3 temp[4].y, src0.xyz, src1.xyz
> 26: src0.xyz = temp[0], src1.xyz = temp[4]
>     DP3 temp[4].y, src0.xyz, src0.xyz
>     MAX temp[3].w, src1.y, src0.0
> 27: src0.xyz = temp[4]
>     REPL_ALPHA temp[4].y
>     RSQ, |src0.y|
> 28: src0.xyz = temp[0], src1.xyz = temp[4]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src0.000
> 29: src0.xyz = temp[0], src1.xyz = temp[3]
>     DP3_SAT temp[0].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[0]
>     LG2 temp[4].w, src0.x
> 31: src0.w = temp[4], src1.w = temp[1]
>     MAD temp[1].w, src0.w, src1.w, src0.0
> 32: src0.w = temp[1]
>     REPL_ALPHA temp[0].x
>     EX2, src0.w
> 33: src0.w = temp[0], src1.xyz = temp[0]
>     MAD temp[0].x, src0.w__, src1.x__, src0.0__
> 34: src0.xyz = temp[0], src1.xyz = const[0]
>     MAD temp[0].xyz, src0.xxx, src1.xyz, src0.000
> 35: src0.xyz = temp[1], src0.w = temp[3], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src2.xyz
> 36: BEGIN_TEX;
> 37: TEX temp[1].x, temp[4].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 38: src0.xyz = temp[0], src1.xyz = temp[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 39: src0.xyz = temp[0], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=33*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 7, tex_end: 0  (code_addr: 000001c0)
>  0: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t1.w     t0.x     t0.x    op: 05000009
>  1: xyz:  t1  t0  t0  bias-> t1.xy                (01840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  2: xyz:  t1  c6  c7  bias-> t1.xy                (01867981)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c6.xyz   c7.xyz    op: 00020200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t1  c3  t0  bias-> t2.y                 (010808c1)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c3.xxx   c3.xyz    op: 00010280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  4: xyz:  t1  t0  t0  bias-> t2.x                 (00880001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  5: xyz:  t2 c10  t0  bias-> t1.xy                (01840a82)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz  c10.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  6: xyz:  t2  c9  t0  bias-> t3.xy                (018c0a42)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c9.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  7: xyz:  t2  c8  t0  bias-> t2.xy                (01880a02)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c8.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>NODE 1: alu_offset: 8, tex_offset: 0, alu_end: 23, tex_end: 2  (code_addr: 000405c8)
>  TEX:
>    TEX t1, t1, texture[3]   (00009841)
>    TEX t3, t3, texture[1]   (000088c3)
>    TEX t2, t2, texture[2]   (00009082)
>  8: xyz:  t3  c4  t0  bias-> t4.x                 (00900903)
>       w:  c0  t2  t0  bias-> t0.w                 (008000a0)
>     xyz:   t3.xyz   c4.xyz   t3.xxx    op: 00804200 
>       w:     c0.w     t2.w      0.0    op: 00040509
>  9: xyz:  t1  c0  c5  bias-> t1.xyz               (03865801)
>       w:  t3  t0  t0  bias-> t1.w                 (00840003)
>     xyz:   t1.xyz   c0.xyz      0.0    op: 00050200 
>       w:     t3.w     c5.z      0.0    op: 00040409
> 10: xyz:  t2  t0  t0  bias-> t3.xyz               (038c0002)
>       w:  c4  t0  t0  bias->                      (00000024)
>     xyz:   t2.xyz   c4.www     -1.0    op: 000d4600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 11: xyz:  t0  t4  t0  bias-> t0.xyz               (03800100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz: t32  t0  t0  bias-> t4.x                 (00900000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:  t32.xxx  t32.xxx  t32.xxx    op: 05004081 
>       w:     t1.w    t32.x    t32.x    op: 04000009
> 13: xyz:  t3  t4  t0  bias-> t4.x                 (00900103)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t4.x      1.0      1.0    op: 00044883
> 14: xyz:  t0  c2  c1   sub-> t5.xyz               (03961880)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz   c1.xxx      0.0    op: 0025048f 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias-> t4.y                 (01100000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:  -t0.xyz  -t0.xyz   t0.xxx    op: 00805020 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t5  t4  t0  bias-> t4.y                 (01100105)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:   |t4.y|     t5.x     t5.x    op: 05800044
> 17: xyz:  t5  t4  t0  bias-> t4.x                 (00900105)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t5.xyz   t5.xyz   t5.xxx    op: 00804000 
>       w:   |t4.x|     t5.x     t5.x    op: 05800043
> 18: xyz:  t3 t32  t4  bias-> t3.xyz               (038c4003)
>       w:  t4  t0  t0  bias-> t4.w                 (00900004)
>     xyz:   t3.xyz   t4.www      0.0    op: 00050600 
>       w:   |t4.y|     t3.x     t3.x    op: 05800047
> 19: xyz:  t5  t0  t0  bias-> t5.xyz               (03940005)
>       w:  t4  t0  t0  bias->                      (00000004)
>     xyz:   t5.xyz   t4.www      0.0    op: 00050600 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 20: xyz:  t0 t32  t5  bias-> t0.xyz               (03805000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:  -t0.xyz   t3.www   t5.xyz    op: 00020620 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 21: xyz:  t5  t3  t0  bias-> t4.y                 (011000c5)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t3.xyz   t5.xxx    op: 00804200 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 22: xyz:  t0  t4  t0  bias-> t4.y                 (01100100)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t0.xyz   t0.xyz   t0.xxx    op: 00804000 
>       w:     t4.y      0.0     t0.x    op: 01800804
> 23: xyz:  t4  t0  t0  bias-> t4.y                 (01100004)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xxx   t4.xxx   t4.xxx    op: 05004081 
>       w:   |t4.y|     t4.x     t4.x    op: 05800041
> 24: xyz:  t0  t4  t0  bias-> t0.xyz               (03800100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t4.yyy      0.0    op: 00050300 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 25: xyz:  t0  t3  t0  bias-> t0.x                 (008000c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t3.xyz   t0.xxx    op: 40804200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 26: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t4.w                 (00900000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x     t0.x     t0.x    op: 04800000
> 27: xyz: t32  t0  t0  bias->                      (00000000)
>       w:  t4  t1  t0  bias-> t1.w                 (00840044)
>     xyz:  t32.xxx  t32.xxx  t32.xxx    op: 00004081 
>       w:     t4.w     t1.w      0.0    op: 00040509
> 28: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 29: xyz: t32  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.www   t0.xyz      0.0    op: 0005020c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
> 30: xyz:  t0  c0  t0  bias-> t0.xyz               (03800800)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   c0.xyz      0.0    op: 00050201 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 31: xyz:  t1 t32  t0  bias-> t0.xyz               (03800001)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t1.xyz   t3.www   t0.xyz    op: 00020600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 2: alu_offset: 32, tex_offset: 3, alu_end: 1, tex_end: 0  (code_addr: 00403060)
>  TEX:
>    TEX t1, t4, texture[0]   (00008044)
> 32: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 33: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t2  t0  t0  bias-> o0.w                 (01000002)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], CONST[0].xxxx
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], const[0].xxxx;
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], const[0].xxxx;
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = const[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[0].w, src0.w, src1.x, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=34*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  c0  t0  bias-> t0.xyz               (03800800)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xyz   c0.xxx      0.0    op: 00050280 
>       w:     t0.w     c0.x      0.0    op: 00040189
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[5]
>DCL CONST[0..3]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[5].xxxx, CONST[5].yyyy
>  2: MOV TEMP[1].xyz, CONST[1].xyzx
>  3: TEX TEMP[2].xyz, TEMP[0].xyyy, SAMP[0], RECT
>  4: DP3 TEMP[2].x, TEMP[2].xyzz, IMM[0].xyzz
>  5: MAD TEMP[2].x, -TEMP[2].xxxx, CONST[3].xxxx, -CONST[2].xxxx
>  6: MIN TEMP[2].x, TEMP[2].xxxx, CONST[2].yyyy
>  7: RCP TEMP[3].x, CONST[2].yyyy
>  8: MUL TEMP[2].x, TEMP[2].xxxx, TEMP[3].xxxx
>  9: MAX TEMP[2].x, TEMP[2].xxxx, IMM[0].wwww
> 10: POW TEMP[2].x, TEMP[2].xxxx, CONST[0].xxxx
> 11: MUL TEMP[2].x, TEMP[2].xxxx, CONST[1].wwww
> 12: MOV TEMP[1].w, TEMP[2].xxxx
> 13: MOV OUT[0], TEMP[1]
> 14: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
>  9: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 10: RCP temp[3].x, const[2].yyyy;
> 11: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 12: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 13: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 14: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 15: MOV temp[1].w, temp[2].xxxx;
> 16: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
>  9: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 10: RCP temp[3].x, const[2].yyyy;
> 11: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 12: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 13: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 14: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 15: MOV temp[1].w, temp[2].xxxx;
> 16: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
>  9: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 10: RCP temp[3].x, const[2].yyyy;
> 11: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 12: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 13: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 14: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 15: MOV temp[1].w, temp[2].xxxx;
> 16: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
>  9: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 10: RCP temp[3].x, const[2].yyyy;
> 11: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 12: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 13: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 14: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 15: MOV temp[1].w, temp[2].xxxx;
> 16: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: TEX temp[2].xyz, temp[0].xyyy, RECT[0];
>  7: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
>  8: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
>  9: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 10: RCP temp[3].x, const[2].yyyy;
> 11: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 12: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 13: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 14: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 15: MOV temp[1].w, temp[2].xxxx;
> 16: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: MUL temp[5], temp[0].xyyy, const[9];
>  7: TEX temp[6], temp[5], 2D[0];
>  8: MOV temp[2].xyz, temp[6];
>  9: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
> 10: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
> 11: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 12: RCP temp[3].x, const[2].yyyy;
> 13: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 14: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 15: POW temp[2].x, temp[2].xxxx, const[0].xxxx;
> 16: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 17: MOV temp[1].w, temp[2].xxxx;
> 18: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].wwww;
>  1: MUL temp[4].xyz, input[0], temp[4].wwww;
>  2: MAD temp[4].xyz, temp[4].xyz0, const[7].xyz0, const[8].xyz0;
>  3: MOV temp[0], temp[4];
>  4: MAD temp[0].y, temp[4], const[5].xxxx, const[5].yyyy;
>  5: MOV temp[1].xyz, const[1].xyzx;
>  6: MUL temp[5], temp[0].xyyy, const[9];
>  7: TEX temp[6], temp[5], 2D[0];
>  8: MOV temp[2].xyz, temp[6];
>  9: DP3 temp[2].x, temp[2].xyzz, const[6].xyzz;
> 10: MAD temp[2].x, -temp[2].xxxx, const[3].xxxx, -const[2].xxxx;
> 11: MIN temp[2].x, temp[2].xxxx, const[2].yyyy;
> 12: RCP temp[3].x, const[2].yyyy;
> 13: MUL temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 14: MAX temp[2].x, temp[2].xxxx, const[6].wwww;
> 15: LG2 temp[7].w, temp[2].xxxx;
> 16: MUL temp[7].w, temp[7].wwww, const[0].xxxx;
> 17: EX2 temp[2].x, temp[7].wwww;
> 18: MUL temp[2].x, temp[2].xxxx, const[1].wwww;
> 19: MOV temp[1].w, temp[2].xxxx;
> 20: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].___w;
>  1: MUL temp[4].xy, input[0].xy__, temp[4].ww__;
>  2: MAD temp[4].xy, temp[4].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[0].x, temp[4].x___;
>  4: MAD temp[0].y, temp[4]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[1].xyz, const[1].xyz_;
>  6: MUL temp[5].xy, temp[0].xy__, const[9].xy__;
>  7: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  8: MOV temp[2].xyz, temp[6].xyz_;
>  9: DP3 temp[2].x, temp[2].xyz_, const[6].xyz_;
> 10: MAD temp[2].x, -temp[2].x___, const[3].x___, -const[2].x___;
> 11: MIN temp[2].x, temp[2].x___, const[2].y___;
> 12: RCP temp[3].x, const[2].y___;
> 13: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 14: MAX temp[2].x, temp[2].x___, const[6].w___;
> 15: LG2 temp[7].w, temp[2].___x;
> 16: MUL temp[7].w, temp[7].___w, const[0].___x;
> 17: EX2 temp[2].x, temp[7].w___;
> 18: MUL temp[2].x, temp[2].x___, const[1].w___;
> 19: MOV temp[1].w, temp[2].___x;
> 20: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[4].w, input[0].___w;
>  1: MUL temp[4].xy, input[0].xy__, temp[4].ww__;
>  2: MAD temp[4].xy, temp[4].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[0].x, temp[4].x___;
>  4: MAD temp[0].y, temp[4]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[1].xyz, const[1].xyz_;
>  6: MUL temp[5].xy, temp[0].xy__, const[9].xy__;
>  7: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  8: MOV temp[2].xyz, temp[6].xyz_;
>  9: DP3 temp[2].x, temp[2].xyz_, const[6].xyz_;
> 10: MAD temp[2].x, -temp[2].x___, const[3].x___, -const[2].x___;
> 11: MIN temp[2].x, temp[2].x___, const[2].y___;
> 12: RCP temp[3].x, const[2].y___;
> 13: MUL temp[2].x, temp[2].x___, temp[3].x___;
> 14: MAX temp[2].x, temp[2].x___, const[6].w___;
> 15: LG2 temp[7].w, temp[2].___x;
> 16: MUL temp[7].w, temp[7].___w, const[0].___x;
> 17: EX2 temp[2].x, temp[7].w___;
> 18: MUL temp[2].x, temp[2].x___, const[1].w___;
> 19: MOV temp[1].w, temp[2].___x;
> 20: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: MOV temp[15].xyz, temp[14].xyz_;
>  9: DP3 temp[16].x, temp[15].xyz_, const[6].xyz_;
> 10: MAD temp[17].x, -temp[16].x___, const[3].x___, -const[2].x___;
> 11: MIN temp[18].x, temp[17].x___, const[2].y___;
> 12: RCP temp[19].x, const[2].y___;
> 13: MUL temp[20].x, temp[18].x___, temp[19].x___;
> 14: MAX temp[21].x, temp[20].x___, const[6].w___;
> 15: LG2 temp[22].w, temp[21].___x;
> 16: MUL temp[23].w, temp[22].___w, const[0].___x;
> 17: EX2 temp[24].x, temp[23].w___;
> 18: MUL temp[25].x, temp[24].x___, const[1].w___;
> 19: MOV temp[12].w, temp[25].___x;
> 20: MOV output[0], temp[12];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[6].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, -const[2].x___;
> 10: MIN temp[18].x, temp[17].x___, const[2].y___;
> 11: RCP temp[19].x, const[2].y___;
> 12: MUL temp[20].x, temp[18].x___, temp[19].x___;
> 13: MAX temp[21].x, temp[20].x___, none.0___;
> 14: LG2 temp[22].w, temp[21].___x;
> 15: MUL temp[23].w, temp[22].___w, const[0].___x;
> 16: EX2 temp[24].x, temp[23].w___;
> 17: MUL temp[25].x, temp[24].x___, const[1].w___;
> 18: MOV temp[12].w, temp[25].___x;
> 19: MOV output[0], temp[12];
>CONST[6] = {     1.0000     0.0039     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[7].xy__, const[8].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[5]._x__, const[5]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[9].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[6].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, -const[2].x___;
> 10: MIN temp[18].x, temp[17].x___, const[2].y___;
> 11: RCP temp[19].x, const[2].y___;
> 12: MUL temp[20].x, temp[18].x___, temp[19].x___;
> 13: MAX temp[21].x, temp[20].x___, none.0___;
> 14: LG2 temp[22].w, temp[21].___x;
> 15: MUL temp[23].w, temp[22].___w, const[0].___x;
> 16: EX2 temp[24].x, temp[23].w___;
> 17: MUL temp[25].x, temp[24].x___, const[1].w___;
> 18: MOV temp[12].w, temp[25].___x;
> 19: MOV output[0], temp[12];
>CONST[5] = {     1.0000     0.0039     0.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[8].w, input[0].___w;
>  1: MUL temp[9].xy, input[0].xy__, temp[8].ww__;
>  2: MAD temp[10].xy, temp[9].xy__, const[6].xy__, const[7].xy__;
>  3: MOV temp[11].x, temp[10].x___;
>  4: MAD temp[11].y, temp[10]._y__, const[4]._x__, const[4]._y__;
>  5: MOV temp[12].xyz, const[1].xyz_;
>  6: MUL temp[13].xy, temp[11].xy__, const[8].xy__;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: DP3 temp[16].x, temp[14].xyz_, const[5].xyz_;
>  9: MAD temp[17].x, -temp[16].x___, const[3].x___, -const[2].x___;
> 10: MIN temp[18].x, temp[17].x___, const[2].y___;
> 11: RCP temp[19].x, const[2].y___;
> 12: MUL temp[20].x, temp[18].x___, temp[19].x___;
> 13: MAX temp[21].x, temp[20].x___, none.0___;
> 14: LG2 temp[22].w, temp[21].___x;
> 15: MUL temp[23].w, temp[22].___w, const[0].___x;
> 16: EX2 temp[24].x, temp[23].w___;
> 17: MUL temp[25].x, temp[24].x___, const[1].w___;
> 18: MOV temp[12].w, temp[25].___x;
> 19: MOV output[0], temp[12];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = const[1]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>  6: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0];
>  8: src0.xyz = temp[14], src1.xyz = const[5]
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = const[2]
>     MAD temp[17].x, -src0.x__, src1.x__, -src2.x__
> 10: src0.xyz = temp[17], src1.xyz = const[2]
>     MIN temp[18].x, src0.x__, src1.y__
> 11: src0.xyz = const[2]
>     REPL_ALPHA temp[19].x
>     RCP, src0.y
> 12: src0.xyz = temp[18], src1.xyz = temp[19]
>     MAD temp[20].x, src0.x__, src1.x__, src0.000
> 13: src0.xyz = temp[20]
>     MAX temp[21].x, src0.x__, src0.0__
> 14: src0.xyz = temp[21]
>     LG2 temp[22].w, src0.x
> 15: src0.xyz = const[0], src0.w = temp[22]
>     MAD temp[23].w, src0.w, src0.x, src0.0
> 16: src0.w = temp[23]
>     REPL_ALPHA temp[24].x
>     EX2, src0.w
> 17: src0.xyz = temp[24], src0.w = const[1]
>     MAD temp[25].x, src0.x__, src0.w__, src0.000
> 18: src0.xyz = temp[25]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 19: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[0]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[19].w, src1.y
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[14], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = const[2]
>     MAD temp[17].x, -src0.x__, src1.x__, -src2.x__
> 10: src0.xyz = temp[17], src1.xyz = const[2]
>     MIN temp[18].x, src0.x__, src1.y__
> 11: src0.xyz = temp[18], src0.w = temp[19], src1.xyz = temp[19]
>     MAD temp[20].x, src0.x__, src0.w__, src0.000
> 12: src0.xyz = temp[20]
>     MAX temp[21].x, src0.x__, src0.0__
> 13: src0.xyz = temp[21]
>     LG2 temp[22].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[22]
>     MAD temp[23].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[23]
>     REPL_ALPHA temp[24].x
>     EX2, src0.w
> 16: src0.xyz = temp[24], src0.w = const[1]
>     MAD temp[25].x, src0.x__, src0.w__, src0.000
> 17: src0.xyz = temp[25]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[0]
>     MAD temp[12].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[8].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[8], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src0.ww_, src0.000
>     RCP temp[19].w, src1.y
>  2: src0.xyz = temp[9], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[10].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[10], src1.xyz = const[4]
>     MAD temp[11].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[10]
>     MAD temp[11].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[11], src1.xyz = const[8]
>     MAD temp[13].xy, src0.xy_, src1.xy_, src0.000
>  6: BEGIN_TEX;
>  7: TEX temp[14].xyz, temp[13].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[14], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[16].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[16], src1.xyz = const[3], src2.xyz = const[2]
>     MAD temp[17].x, -src0.x__, src1.x__, -src2.x__
> 10: src0.xyz = temp[17], src1.xyz = const[2]
>     MIN temp[18].x, src0.x__, src1.y__
> 11: src0.xyz = temp[18], src0.w = temp[19]
>     MAD temp[20].x, src0.x__, src0.w__, src0.000
> 12: src0.xyz = temp[20]
>     MAX temp[21].x, src0.x__, src0.0__
> 13: src0.xyz = temp[21]
>     LG2 temp[22].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[22]
>     MAD temp[23].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[23]
>     REPL_ALPHA temp[24].x
>     EX2, src0.w
> 16: src0.xyz = temp[24], src0.w = const[1]
>     MAD temp[25].x, src0.x__, src0.w__, src0.000
> 17: src0.xyz = temp[25]
>     MAD temp[12].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[12], src0.w = temp[12]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src0.w = input[0]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
>     RCP temp[1].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[1], src1.xyz = const[2]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src0.00_
>     RCP temp[0].w, src1.y
>  2: src0.xyz = temp[0], src1.xyz = const[6], src2.xyz = const[7]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[2].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[0]
>     MAD temp[2].x, src0.x__, src0.11_, src0.00_
>  5: src0.xyz = temp[2], src1.xyz = const[8]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  6: BEGIN_TEX;
>  7: TEX temp[2].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  8: src0.xyz = temp[2], src1.xyz = const[5] SEM_WAIT
>     DP3 temp[0].x, src0.xyz, src1.xyz
>  9: src0.xyz = temp[0], src1.xyz = const[3], src2.xyz = const[2]
>     MAD temp[0].x, -src0.x__, src1.x__, -src2.x__
> 10: src0.xyz = temp[0], src1.xyz = const[2]
>     MIN temp[0].x, src0.x__, src1.y__
> 11: src0.xyz = temp[0], src0.w = temp[0]
>     MAD temp[0].x, src0.x__, src0.w__, src0.0__
> 12: src0.xyz = temp[0]
>     MAX temp[0].x, src0.x__, src0.0__
> 13: src0.xyz = temp[0]
>     LG2 temp[0].w, src0.x
> 14: src0.xyz = const[0], src0.w = temp[0]
>     MAD temp[0].w, src0.w, src0.x, src0.0
> 15: src0.w = temp[0]
>     REPL_ALPHA temp[0].x
>     EX2, src0.w
> 16: src0.xyz = temp[0], src0.w = const[1]
>     MAD temp[0].x, src0.x__, src0.w__, src0.0__
> 17: src0.xyz = temp[0]
>     MAD temp[0].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=35*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 5, tex_end: 0  (code_addr: 00000140)
>  0: xyz:  c1  t0  t0  bias-> t1.xyz               (03840021)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   c1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w     c1.x     c1.x    op: 05000009
>  1: xyz:  t0  c2  t0  bias-> t0.xy                (01800880)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t0.xyz   t1.www      0.0    op: 00050600 
>       w:     c2.y     t0.x     t0.x    op: 05000004
>  2: xyz:  t0  c6  c7  bias-> t0.xy                (01827980)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c6.xyz   c7.xyz    op: 00020200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  c4  t0  bias-> t2.y                 (01080900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xxx   c4.xyz    op: 00010280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t0  t0  t0  bias-> t2.x                 (00880000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  5: xyz:  t2  c8  t0  bias-> t0.xy                (01800a02)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c8.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>NODE 1: alu_offset: 6, tex_offset: 0, alu_end: 10, tex_end: 0  (code_addr: 00400286)
>  TEX:
>    TEX t2, t0, texture[0]   (00008080)
>  6: xyz:  t2  c5  t0  bias-> t0.x                 (00800942)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c5.xyz   t2.xxx    op: 00804200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  7: xyz:  t0  c3  c2  bias-> t0.x                 (008228c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:  -t0.xyz   c3.xyz  -c2.xyz    op: 000a0220 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0  c2  t0  bias-> t0.x                 (00800880)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c2.yyy   t0.xxx    op: 02004300 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  9: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t0.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 10: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      0.0   t0.xxx    op: 02804a00 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x     t0.x     t0.x    op: 04800000
> 12: xyz:  c0  t0  t0  bias->                      (00000020)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c0.xxx   c0.xxx   c0.xxx    op: 00004081 
>       w:     t0.w     c0.x      0.0    op: 00040009
> 13: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t0.w     t0.x     t0.x    op: 04000009
> 14: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t0.xyz   c1.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.x      1.0      0.0    op: 00040880
> 16: xyz:  t1  t0  t0  bias-> o0.xyz               (1c000001)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0..4]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[2]
>  2: ADD TEMP[1].x, -IN[1].zzzz, -CONST[2].xxxx
>  3: RCP TEMP[2].x, CONST[2].yyyy
>  4: MUL_SAT TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
>  5: POW TEMP[1].x, TEMP[1].xxxx, CONST[0].xxxx
>  6: ADD TEMP[2].x, IMM[0].xxxx, -TEMP[1].xxxx
>  7: MAD TEMP[1].x, CONST[1].xxxx, TEMP[1].xxxx, TEMP[2].xxxx
>  8: MUL TEMP[1].x, TEMP[1].xxxx, CONST[4].xxxx
>  9: MUL TEMP[1].x, TEMP[1].xxxx, CONST[3].xxxx
> 10: MUL TEMP[0].xyz, TEMP[0].xyzz, TEMP[1].xxxx
> 11: MOV OUT[0], TEMP[0]
> 12: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 10: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 11: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: LG2 temp[3].w, temp[1].xxxx;
>  6: MUL temp[3].w, temp[3].wwww, const[0].xxxx;
>  7: EX2 temp[1].x, temp[3].wwww;
>  8: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  9: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
> 10: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
> 11: MUL temp[1].x, temp[1].xxxx, const[3].xxxx;
> 12: MUL temp[0].xyz, temp[0].xyzz, temp[1].xxxx;
> 13: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[2].x, const[2].y___;
>  4: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  5: LG2 temp[3].w, temp[1].___x;
>  6: MUL temp[3].w, temp[3].___w, const[0].___x;
>  7: EX2 temp[1].x, temp[3].w___;
>  8: ADD temp[2].x, temp[0].1___, -temp[1].x___;
>  9: MAD temp[1].x, const[1].x___, temp[1].x___, temp[2].x___;
> 10: MUL temp[1].x, temp[1].x___, const[4].x___;
> 11: MUL temp[1].x, temp[1].x___, const[3].x___;
> 12: MUL temp[0].xyz, temp[0].xyz_, temp[1].xxx_;
> 13: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[2].x, const[2].y___;
>  4: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  5: LG2 temp[3].w, temp[1].___x;
>  6: MUL temp[3].w, temp[3].___w, const[0].___x;
>  7: EX2 temp[1].x, temp[3].w___;
>  8: ADD temp[2].x, temp[0].1___, -temp[1].x___;
>  9: MAD temp[1].x, const[1].x___, temp[1].x___, temp[2].x___;
> 10: MUL temp[1].x, temp[1].x___, const[4].x___;
> 11: MUL temp[1].x, temp[1].x___, const[3].x___;
> 12: MUL temp[0].xyz, temp[0].xyz_, temp[1].xxx_;
> 13: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, temp[0].1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[14].x___, const[3].x___;
> 12: MUL temp[5].xyz, temp[5].xyz_, temp[15].xxx_;
> 13: MOV output[0], temp[5];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[14].x___, const[3].x___;
> 12: MUL temp[5].xyz, temp[5].xyz_, temp[15].xxx_;
> 13: MOV output[0], temp[5];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[14].x___, const[3].x___;
> 12: MUL temp[5].xyz, temp[5].xyz_, temp[15].xxx_;
> 13: MOV output[0], temp[5];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[14].x___, const[3].x___;
> 12: MUL temp[5].xyz, temp[5].xyz_, temp[15].xxx_;
> 13: MOV output[0], temp[5];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2]
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>  3: src0.xyz = const[2]
>     REPL_ALPHA temp[7].x
>     RCP, src0.y
>  4: src0.xyz = temp[6], src1.xyz = temp[7]
>     MAD_SAT temp[8].x, src0.x__, src1.x__, src0.000
>  5: src0.xyz = temp[8]
>     LG2 temp[9].w, src0.x
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src1.xyz = const[3]
>     MAD temp[15].x, src0.x__, src1.x__, src0.000
> 12: src0.xyz = temp[5], src1.xyz = temp[15]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
> 13: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[4], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[7].w, src1.y
>  3: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[6], src0.w = temp[7], src1.xyz = temp[7]
>     MAD_SAT temp[8].x, src0.x__, src0.w__, src0.000
>  5: src0.xyz = temp[8]
>     LG2 temp[9].w, src0.x
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src1.xyz = const[3]
>     MAD temp[15].x, src0.x__, src1.x__, src0.000
> 12: src0.xyz = temp[5], src1.xyz = temp[15]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
> 13: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[4], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[7].w, src1.y
>  3: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[6], src0.w = temp[7]
>     MAD_SAT temp[8].x, src0.x__, src0.w__, src0.000
>  5: src0.xyz = temp[8]
>     LG2 temp[9].w, src0.x
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src1.xyz = const[3]
>     MAD temp[15].x, src0.x__, src1.x__, src0.000
> 12: src0.xyz = temp[5], src1.xyz = temp[15]
>     MAD temp[5].xyz, src0.xyz, src1.xxx, src0.000
> 13: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[1].x, -src0.z__, src0.1__, -src1.x__
>     RCP temp[1].w, src1.y
>  3: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[1], src0.w = temp[1]
>     MAD_SAT temp[1].x, src0.x__, src0.w__, src0.0__
>  5: src0.xyz = temp[1]
>     LG2 temp[1].w, src0.x
>  6: src0.xyz = const[0], src0.w = temp[1]
>     MAD temp[1].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[1]
>     REPL_ALPHA temp[1].x
>     EX2, src0.w
>  8: src0.xyz = temp[1]
>     MAD temp[1].y, src0._1_, src0._1_, -src0._x_
>  9: src0.xyz = const[1], src1.xyz = temp[1], src2.xyz = temp[1]
>     MAD temp[1].x, src0.x__, src1.x__, src2.y__
> 10: src0.xyz = temp[1], src1.xyz = const[4]
>     MAD temp[1].x, src0.x__, src1.x__, src0.0__
> 11: src0.xyz = temp[1], src1.xyz = const[3]
>     MAD temp[1].x, src0.x__, src1.x__, src0.0__
> 12: src0.xyz = temp[0], src1.xyz = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 13: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=36*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 11, tex_end: 0  (code_addr: 004002c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c2  t0  bias-> t1.x                 (00840881)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:  -t1.zzz      1.0  -c2.xyz    op: 00090aa3 
>       w:     c2.y     t1.x     t1.x    op: 05000004
>  1: xyz:  t0  t2  t0  bias-> t0.xyz               (03800080)
>       w:  t0  t2  t0  bias-> t0.w                 (00800080)
>     xyz:   t0.xyz   t2.xyz      0.0    op: 00050200 
>       w:     t0.w     t2.w      0.0    op: 00040509
>  2: xyz:  t1  t0  t0  bias-> t1.x                 (00840001)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t1.xyz   t1.www      0.0    op: 40050600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t1  t0  t0  bias->                      (00000001)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xxx   t1.xxx   t1.xxx    op: 00004081 
>       w:     t1.x     t1.x     t1.x    op: 04800000
>  4: xyz:  c0  t0  t0  bias->                      (00000020)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   c0.xxx   c0.xxx   c0.xxx    op: 00004081 
>       w:     t1.w     c0.x      0.0    op: 00040009
>  5: xyz:  t0  t0  t0  bias-> t1.x                 (00840000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
>  6: xyz:  t1  t0  t0  bias-> t1.y                 (01040001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t1.xxx    op: 00084a95 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  7: xyz:  c1  t1  t1  bias-> t1.x                 (00841061)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c1.xyz   t1.xyz   t1.yyy    op: 00028200 
>       w:     c1.x     c1.x     c1.x    op: 00000000
>  8: xyz:  t1  c4  t0  bias-> t1.x                 (00840901)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  9: xyz:  t1  c3  t0  bias-> t1.x                 (008408c1)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c3.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 10: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL IN[2], GENERIC[12], PERSPECTIVE
>DCL IN[3], GENERIC[13], PERSPECTIVE
>DCL IN[4], GENERIC[14], PERSPECTIVE
>DCL IN[5], GENERIC[15], PERSPECTIVE
>DCL IN[6], GENERIC[16], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[16]
>DCL CONST[0..11]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>DCL TEMP[6], LOCAL
>IMM[0] FLT32 {    1.0000,     2.0000,    -1.0000,    10.0000}
>IMM[1] FLT32 {    0.0000,    -1.0000,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[16].xxxx, CONST[16].yyyy
>  2: TEX TEMP[1], IN[1].xyyy, SAMP[3], 2D
>  3: MUL TEMP[1], TEMP[1], IN[3]
>  4: ADD TEMP[2].x, -IN[2].zzzz, -CONST[8].xxxx
>  5: RCP TEMP[3].x, CONST[8].yyyy
>  6: MUL_SAT TEMP[2].x, TEMP[2].xxxx, TEMP[3].xxxx
>  7: POW TEMP[2].x, TEMP[2].xxxx, CONST[6].xxxx
>  8: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[2].xxxx
>  9: MAD TEMP[2].x, CONST[7].xxxx, TEMP[2].xxxx, TEMP[3].xxxx
> 10: MUL TEMP[2].x, TEMP[2].xxxx, CONST[11].xxxx
> 11: ADD TEMP[3].xyz, IMM[0].xxxx, -TEMP[1].xyzz
> 12: ADD TEMP[4].x, IMM[0].xxxx, -TEMP[2].xxxx
> 13: MAD TEMP[1].xyz, TEMP[3].xyzz, TEMP[4].xxxx, TEMP[1].xyzz
> 14: TEX TEMP[3].xyz, IN[1].xyyy, SAMP[2], 2D
> 15: MAD TEMP[3].xyz, TEMP[3].xyzz, IMM[0].yyyy, IMM[0].zzzz
> 16: MUL TEMP[4].xyz, TEMP[3].xxxx, IN[5].xyzz
> 17: MAD TEMP[4].xyz, TEMP[3].yyyy, IN[4].xyzz, TEMP[4].xyzz
> 18: MAD TEMP[4].xyz, TEMP[3].zzzz, IN[6].xyzz, TEMP[4].xyzz
> 19: DP3 TEMP[5].x, TEMP[4].xyzz, TEMP[4].xyzz
> 20: RSQ TEMP[5].x, TEMP[5].xxxx
> 21: MUL TEMP[4].xyz, TEMP[4].xyzz, TEMP[5].xxxx
> 22: RCP TEMP[5].x, IN[2].zzzz
> 23: MIN TEMP[5].x, TEMP[5].xxxx, IMM[0].wwww
> 24: MUL TEMP[5].x, CONST[9].xxxx, TEMP[5].xxxx
> 25: MAD TEMP[3].xy, TEMP[3].xyyy, TEMP[5].xxxx, TEMP[0].xyyy
> 26: TEX TEMP[3].xyz, TEMP[3].xyyy, SAMP[1], RECT
> 27: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[3].xyzz
> 28: DP3 TEMP[3].x, IN[2].xyzz, IN[2].xyzz
> 29: RSQ TEMP[3].x, TEMP[3].xxxx
> 30: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[3].xxxx
> 31: DP3 TEMP[6].x, TEMP[4].xyzz, TEMP[3].xyzz
> 32: MUL TEMP[6].xyz, TEMP[6].xxxx, TEMP[4].xyzz
> 33: MUL TEMP[6].xyz, IMM[0].yyyy, TEMP[6].xyzz
> 34: ADD TEMP[5].xyz, TEMP[3].xyzz, -TEMP[6].xyzz
> 35: MUL TEMP[6], CONST[2], TEMP[5].xxxx
> 36: MAD TEMP[6], CONST[3], TEMP[5].yyyy, TEMP[6]
> 37: MAD TEMP[5], CONST[4], TEMP[5].zzzz, TEMP[6]
> 38: ADD TEMP[5].xyz, TEMP[5], CONST[5]
> 39: TEX TEMP[5].xyz, TEMP[5].xyzz, SAMP[0], CUBE
> 40: ADD TEMP[6].x, IMM[0].xxxx, -CONST[1].xxxx
> 41: DP3 TEMP[3].x, -TEMP[3].xyzz, TEMP[4].xyzz
> 42: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 43: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[3].xxxx
> 44: POW TEMP[3].x, TEMP[3].xxxx, CONST[1].yyyy
> 45: MAD TEMP[3].x, TEMP[6].xxxx, TEMP[3].xxxx, CONST[1].xxxx
> 46: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 47: MUL TEMP[3].xyz, TEMP[5].xyzz, TEMP[3].xxxx
> 48: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[2].xxxx
> 49: MAD TEMP[1].xyz, TEMP[3].xyzz, CONST[10].xxxx, TEMP[1].xyzz
> 50: DP3 TEMP[3].x, TEMP[4].xyzz, IMM[1].xxyy
> 51: ABS TEMP[3].x, TEMP[3].xxxx
> 52: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[3].xxxx
> 53: POW TEMP[3].x, TEMP[3].xxxx, CONST[0].yyyy
> 54: MUL TEMP[3].x, TEMP[3].xxxx, CONST[0].xxxx
> 55: MUL TEMP[3].xyz, TEMP[5].xyzz, TEMP[3].xxxx
> 56: MUL TEMP[2].xyz, TEMP[3].xyzz, TEMP[2].xxxx
> 57: MAD TEMP[1].xyz, TEMP[2].xyzz, CONST[10].xxxx, TEMP[1].xyzz
> 58: MOV OUT[0], TEMP[1]
> 59: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[3].xyz, input[1].xyyy, 2D[2];
> 18: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 19: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 20: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 21: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 22: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 23: RSQ temp[5].x, temp[5].xxxx;
> 24: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 25: RCP temp[5].x, input[2].zzzz;
> 26: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 27: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 28: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 29: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 30: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 31: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 32: RSQ temp[3].x, temp[3].xxxx;
> 33: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 34: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 35: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 36: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 37: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 38: MUL temp[6], const[2], temp[5].xxxx;
> 39: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 40: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 41: ADD temp[5].xyz, temp[5], const[5];
> 42: TEX temp[5].xyz, temp[5].xyzz, CUBE[0];
> 43: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 44: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 45: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 46: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 47: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 48: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 52: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 53: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 54: ABS temp[3].x, temp[3].xxxx;
> 55: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 56: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 57: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[3].xyz, input[1].xyyy, 2D[2];
> 18: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 19: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 20: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 21: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 22: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 23: RSQ temp[5].x, temp[5].xxxx;
> 24: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 25: RCP temp[5].x, input[2].zzzz;
> 26: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 27: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 28: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 29: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 30: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 31: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 32: RSQ temp[3].x, temp[3].xxxx;
> 33: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 34: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 35: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 36: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 37: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 38: MUL temp[6], const[2], temp[5].xxxx;
> 39: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 40: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 41: ADD temp[5].xyz, temp[5], const[5];
> 42: TEX temp[5].xyz, temp[5].xyzz, CUBE[0];
> 43: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 44: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 45: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 46: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 47: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 48: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 52: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 53: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 54: ABS temp[3].x, temp[3].xxxx;
> 55: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 56: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 57: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[3].xyz, input[1].xyyy, 2D[2];
> 18: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 19: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 20: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 21: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 22: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 23: RSQ temp[5].x, temp[5].xxxx;
> 24: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 25: RCP temp[5].x, input[2].zzzz;
> 26: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 27: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 28: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 29: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 30: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 31: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 32: RSQ temp[3].x, temp[3].xxxx;
> 33: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 34: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 35: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 36: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 37: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 38: MUL temp[6], const[2], temp[5].xxxx;
> 39: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 40: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 41: ADD temp[5].xyz, temp[5], const[5];
> 42: TEX temp[5].xyz, temp[5].xyzz, CUBE[0];
> 43: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 44: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 45: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 46: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 47: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 48: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 52: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 53: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 54: ABS temp[3].x, temp[3].xxxx;
> 55: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 56: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 57: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[3].xyz, input[1].xyyy, 2D[2];
> 18: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 19: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 20: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 21: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 22: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 23: RSQ temp[5].x, temp[5].xxxx;
> 24: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 25: RCP temp[5].x, input[2].zzzz;
> 26: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 27: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 28: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 29: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 30: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 31: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 32: RSQ temp[3].x, temp[3].xxxx;
> 33: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 34: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 35: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 36: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 37: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 38: MUL temp[6], const[2], temp[5].xxxx;
> 39: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 40: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 41: ADD temp[5].xyz, temp[5], const[5];
> 42: TEX temp[5].xyz, temp[5].xyzz, CUBE[0];
> 43: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 44: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 45: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 46: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 47: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 48: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 52: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 53: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 54: ABS temp[3].x, temp[3].xxxx;
> 55: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 56: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 57: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[3].xyz, input[1].xyyy, 2D[2];
> 18: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 19: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 20: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 21: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 22: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 23: RSQ temp[5].x, temp[5].xxxx;
> 24: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 25: RCP temp[5].x, input[2].zzzz;
> 26: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 27: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 28: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 29: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 30: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 31: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 32: RSQ temp[3].x, temp[3].xxxx;
> 33: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 34: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 35: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 36: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 37: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 38: MUL temp[6], const[2], temp[5].xxxx;
> 39: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 40: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 41: ADD temp[5].xyz, temp[5], const[5];
> 42: TEX temp[5].xyz, temp[5].xyzz, CUBE[0];
> 43: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 44: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 45: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 46: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 47: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 48: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 52: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 53: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 54: ABS temp[3].x, temp[3].xxxx;
> 55: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 56: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 57: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: POW temp[2].x, temp[2].xxxx, const[6].xxxx;
> 11: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 12: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 13: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 14: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 15: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 16: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 17: TEX temp[8], input[1].xyyy, 2D[2];
> 18: MOV temp[3].xyz, temp[8];
> 19: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 20: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 21: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 22: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 23: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 24: RSQ temp[5].x, temp[5].xxxx;
> 25: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 26: RCP temp[5].x, input[2].zzzz;
> 27: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 28: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 29: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 30: MUL temp[9], temp[3].xyyy, const[21];
> 31: TEX temp[10], temp[9], 2D[1];
> 32: MOV temp[3].xyz, temp[10];
> 33: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 34: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 35: RSQ temp[3].x, temp[3].xxxx;
> 36: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 37: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 38: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 39: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 40: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 41: MUL temp[6], const[2], temp[5].xxxx;
> 42: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 43: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 44: ADD temp[5].xyz, temp[5], const[5];
> 45: TEX temp[11], temp[5].xyzz, CUBE[0];
> 46: MOV temp[5].xyz, temp[11];
> 47: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 48: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 49: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 50: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 51: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 52: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 53: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 54: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 55: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 56: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 57: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 58: ABS temp[3].x, temp[3].xxxx;
> 59: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 60: POW temp[3].x, temp[3].xxxx, const[0].yyyy;
> 61: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 62: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 63: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 64: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 65: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].wwww;
>  1: MUL temp[7].xyz, input[0], temp[7].wwww;
>  2: MAD temp[7].xyz, temp[7].xyz0, const[19].xyz0, const[20].xyz0;
>  3: MOV temp[0], temp[7];
>  4: MAD temp[0].y, temp[7], const[16].xxxx, const[16].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].zzzz, -const[8].xxxx;
>  8: RCP temp[3].x, const[8].yyyy;
>  9: MUL_SAT temp[2].x, temp[2].xxxx, temp[3].xxxx;
> 10: LG2 temp[12].w, temp[2].xxxx;
> 11: MUL temp[12].w, temp[12].wwww, const[6].xxxx;
> 12: EX2 temp[2].x, temp[12].wwww;
> 13: ADD temp[3].x, const[17].xxxx, -temp[2].xxxx;
> 14: MAD temp[2].x, const[7].xxxx, temp[2].xxxx, temp[3].xxxx;
> 15: MUL temp[2].x, temp[2].xxxx, const[11].xxxx;
> 16: ADD temp[3].xyz, const[17].xxxx, -temp[1].xyzz;
> 17: ADD temp[4].x, const[17].xxxx, -temp[2].xxxx;
> 18: MAD temp[1].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 19: TEX temp[8], input[1].xyyy, 2D[2];
> 20: MOV temp[3].xyz, temp[8];
> 21: MAD temp[3].xyz, temp[3].xyzz, const[17].yyyy, const[17].zzzz;
> 22: MUL temp[4].xyz, temp[3].xxxx, input[5].xyzz;
> 23: MAD temp[4].xyz, temp[3].yyyy, input[4].xyzz, temp[4].xyzz;
> 24: MAD temp[4].xyz, temp[3].zzzz, input[6].xyzz, temp[4].xyzz;
> 25: DP3 temp[5].x, temp[4].xyzz, temp[4].xyzz;
> 26: RSQ temp[5].x, |temp[5].xxxx|;
> 27: MUL temp[4].xyz, temp[4].xyzz, temp[5].xxxx;
> 28: RCP temp[5].x, input[2].zzzz;
> 29: MIN temp[5].x, temp[5].xxxx, const[17].wwww;
> 30: MUL temp[5].x, const[9].xxxx, temp[5].xxxx;
> 31: MAD temp[3].xy, temp[3].xyyy, temp[5].xxxx, temp[0].xyyy;
> 32: MUL temp[9], temp[3].xyyy, const[21];
> 33: TEX temp[10], temp[9], 2D[1];
> 34: MOV temp[3].xyz, temp[10];
> 35: MUL temp[1].xyz, temp[1].xyzz, temp[3].xyzz;
> 36: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 37: RSQ temp[3].x, |temp[3].xxxx|;
> 38: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 39: DP3 temp[6].x, temp[4].xyzz, temp[3].xyzz;
> 40: MUL temp[6].xyz, temp[6].xxxx, temp[4].xyzz;
> 41: MUL temp[6].xyz, const[17].yyyy, temp[6].xyzz;
> 42: ADD temp[5].xyz, temp[3].xyzz, -temp[6].xyzz;
> 43: MUL temp[6], const[2], temp[5].xxxx;
> 44: MAD temp[6], const[3], temp[5].yyyy, temp[6];
> 45: MAD temp[5], const[4], temp[5].zzzz, temp[6];
> 46: ADD temp[5].xyz, temp[5], const[5];
> 47: TEX temp[11], temp[5].xyzz, CUBE[0];
> 48: MOV temp[5].xyz, temp[11];
> 49: ADD temp[6].x, const[17].xxxx, -const[1].xxxx;
> 50: DP3 temp[3].x, -temp[3].xyzz, temp[4].xyzz;
> 51: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 52: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 53: LG2 temp[13].w, temp[3].xxxx;
> 54: MUL temp[13].w, temp[13].wwww, const[1].yyyy;
> 55: EX2 temp[3].x, temp[13].wwww;
> 56: MAD temp[3].x, temp[6].xxxx, temp[3].xxxx, const[1].xxxx;
> 57: MAX temp[3].x, temp[3].xxxx, const[18].xxxx;
> 58: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 59: MUL temp[3].xyz, temp[3].xyzz, temp[2].xxxx;
> 60: MAD temp[1].xyz, temp[3].xyzz, const[10].xxxx, temp[1].xyzz;
> 61: DP3 temp[3].x, temp[4].xyzz, const[18].xxyy;
> 62: MOV temp[3].x, |temp[3].xxxx|;
> 63: ADD temp[3].x, const[17].xxxx, -temp[3].xxxx;
> 64: LG2 temp[14].w, temp[3].xxxx;
> 65: MUL temp[14].w, temp[14].wwww, const[0].yyyy;
> 66: EX2 temp[3].x, temp[14].wwww;
> 67: MUL temp[3].x, temp[3].xxxx, const[0].xxxx;
> 68: MUL temp[3].xyz, temp[5].xyzz, temp[3].xxxx;
> 69: MUL temp[2].xyz, temp[3].xyzz, temp[2].xxxx;
> 70: MAD temp[1].xyz, temp[2].xyzz, const[10].xxxx, temp[1].xyzz;
> 71: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].___w;
>  1: MUL temp[7].xy, input[0].xy__, temp[7].ww__;
>  2: MAD temp[7].xy, temp[7].xy__, const[19].xy__, const[20].xy__;
>  3: MOV temp[0].x, temp[7].x___;
>  4: MAD temp[0].y, temp[7]._y__, const[16]._x__, const[16]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[3].x, const[8].y___;
>  9: MUL_SAT temp[2].x, temp[2].x___, temp[3].x___;
> 10: LG2 temp[12].w, temp[2].___x;
> 11: MUL temp[12].w, temp[12].___w, const[6].___x;
> 12: EX2 temp[2].x, temp[12].w___;
> 13: ADD temp[3].x, const[17].x___, -temp[2].x___;
> 14: MAD temp[2].x, const[7].x___, temp[2].x___, temp[3].x___;
> 15: MUL temp[2].x, temp[2].x___, const[11].x___;
> 16: ADD temp[3].xyz, const[17].xxx_, -temp[1].xyz_;
> 17: ADD temp[4].x, const[17].x___, -temp[2].x___;
> 18: MAD temp[1].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 19: TEX temp[8].xyz, input[1].xy__, 2D[2];
> 20: MOV temp[3].xyz, temp[8].xyz_;
> 21: MAD temp[3].xyz, temp[3].xyz_, const[17].yyy_, const[17].zzz_;
> 22: MUL temp[4].xyz, temp[3].xxx_, input[5].xyz_;
> 23: MAD temp[4].xyz, temp[3].yyy_, input[4].xyz_, temp[4].xyz_;
> 24: MAD temp[4].xyz, temp[3].zzz_, input[6].xyz_, temp[4].xyz_;
> 25: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 26: RSQ temp[5].x, |temp[5].x___|;
> 27: MUL temp[4].xyz, temp[4].xyz_, temp[5].xxx_;
> 28: RCP temp[5].x, input[2].z___;
> 29: MIN temp[5].x, temp[5].x___, const[17].w___;
> 30: MUL temp[5].x, const[9].x___, temp[5].x___;
> 31: MAD temp[3].xy, temp[3].xy__, temp[5].xx__, temp[0].xy__;
> 32: MUL temp[9].xy, temp[3].xy__, const[21].xy__;
> 33: TEX temp[10].xyz, temp[9].xy__, 2D[1];
> 34: MOV temp[3].xyz, temp[10].xyz_;
> 35: MUL temp[1].xyz, temp[1].xyz_, temp[3].xyz_;
> 36: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 37: RSQ temp[3].x, |temp[3].x___|;
> 38: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 39: DP3 temp[6].x, temp[4].xyz_, temp[3].xyz_;
> 40: MUL temp[6].xyz, temp[6].xxx_, temp[4].xyz_;
> 41: MUL temp[6].xyz, const[17].yyy_, temp[6].xyz_;
> 42: ADD temp[5].xyz, temp[3].xyz_, -temp[6].xyz_;
> 43: MUL temp[6].xyz, const[2].xyz_, temp[5].xxx_;
> 44: MAD temp[6].xyz, const[3].xyz_, temp[5].yyy_, temp[6].xyz_;
> 45: MAD temp[5].xyz, const[4].xyz_, temp[5].zzz_, temp[6].xyz_;
> 46: ADD temp[5].xyz, temp[5].xyz_, const[5].xyz_;
> 47: TEX temp[11].xyz, temp[5].xyz_, CUBE[0];
> 48: MOV temp[5].xyz, temp[11].xyz_;
> 49: ADD temp[6].x, const[17].x___, -const[1].x___;
> 50: DP3 temp[3].x, -temp[3].xyz_, temp[4].xyz_;
> 51: MAX temp[3].x, temp[3].x___, const[18].x___;
> 52: ADD temp[3].x, const[17].x___, -temp[3].x___;
> 53: LG2 temp[13].w, temp[3].___x;
> 54: MUL temp[13].w, temp[13].___w, const[1].___y;
> 55: EX2 temp[3].x, temp[13].w___;
> 56: MAD temp[3].x, temp[6].x___, temp[3].x___, const[1].x___;
> 57: MAX temp[3].x, temp[3].x___, const[18].x___;
> 58: MUL temp[3].xyz, temp[5].xyz_, temp[3].xxx_;
> 59: MUL temp[3].xyz, temp[3].xyz_, temp[2].xxx_;
> 60: MAD temp[1].xyz, temp[3].xyz_, const[10].xxx_, temp[1].xyz_;
> 61: DP3 temp[3].x, temp[4].xyz_, const[18].xxy_;
> 62: MOV temp[3].x, |temp[3].x___|;
> 63: ADD temp[3].x, const[17].x___, -temp[3].x___;
> 64: LG2 temp[14].w, temp[3].___x;
> 65: MUL temp[14].w, temp[14].___w, const[0].___y;
> 66: EX2 temp[3].x, temp[14].w___;
> 67: MUL temp[3].x, temp[3].x___, const[0].x___;
> 68: MUL temp[3].xyz, temp[5].xyz_, temp[3].xxx_;
> 69: MUL temp[2].xyz, temp[3].xyz_, temp[2].xxx_;
> 70: MAD temp[1].xyz, temp[2].xyz_, const[10].xxx_, temp[1].xyz_;
> 71: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[7].w, input[0].___w;
>  1: MUL temp[7].xy, input[0].xy__, temp[7].ww__;
>  2: MAD temp[7].xy, temp[7].xy__, const[19].xy__, const[20].xy__;
>  3: MOV temp[0].x, temp[7].x___;
>  4: MAD temp[0].y, temp[7]._y__, const[16]._x__, const[16]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: ADD temp[2].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[3].x, const[8].y___;
>  9: MUL_SAT temp[2].x, temp[2].x___, temp[3].x___;
> 10: LG2 temp[12].w, temp[2].___x;
> 11: MUL temp[12].w, temp[12].___w, const[6].___x;
> 12: EX2 temp[2].x, temp[12].w___;
> 13: ADD temp[3].x, const[17].x___, -temp[2].x___;
> 14: MAD temp[2].x, const[7].x___, temp[2].x___, temp[3].x___;
> 15: MUL temp[2].x, temp[2].x___, const[11].x___;
> 16: ADD temp[3].xyz, const[17].xxx_, -temp[1].xyz_;
> 17: ADD temp[4].x, const[17].x___, -temp[2].x___;
> 18: MAD temp[1].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 19: TEX temp[8].xyz, input[1].xy__, 2D[2];
> 20: MOV temp[3].xyz, temp[8].xyz_;
> 21: MAD temp[3].xyz, temp[3].xyz_, const[17].yyy_, const[17].zzz_;
> 22: MUL temp[4].xyz, temp[3].xxx_, input[5].xyz_;
> 23: MAD temp[4].xyz, temp[3].yyy_, input[4].xyz_, temp[4].xyz_;
> 24: MAD temp[4].xyz, temp[3].zzz_, input[6].xyz_, temp[4].xyz_;
> 25: DP3 temp[5].x, temp[4].xyz_, temp[4].xyz_;
> 26: RSQ temp[5].x, |temp[5].x___|;
> 27: MUL temp[4].xyz, temp[4].xyz_, temp[5].xxx_;
> 28: RCP temp[5].x, input[2].z___;
> 29: MIN temp[5].x, temp[5].x___, const[17].w___;
> 30: MUL temp[5].x, const[9].x___, temp[5].x___;
> 31: MAD temp[3].xy, temp[3].xy__, temp[5].xx__, temp[0].xy__;
> 32: MUL temp[9].xy, temp[3].xy__, const[21].xy__;
> 33: TEX temp[10].xyz, temp[9].xy__, 2D[1];
> 34: MOV temp[3].xyz, temp[10].xyz_;
> 35: MUL temp[1].xyz, temp[1].xyz_, temp[3].xyz_;
> 36: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 37: RSQ temp[3].x, |temp[3].x___|;
> 38: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 39: DP3 temp[6].x, temp[4].xyz_, temp[3].xyz_;
> 40: MUL temp[6].xyz, temp[6].xxx_, temp[4].xyz_;
> 41: MUL temp[6].xyz, const[17].yyy_, temp[6].xyz_;
> 42: ADD temp[5].xyz, temp[3].xyz_, -temp[6].xyz_;
> 43: MUL temp[6].xyz, const[2].xyz_, temp[5].xxx_;
> 44: MAD temp[6].xyz, const[3].xyz_, temp[5].yyy_, temp[6].xyz_;
> 45: MAD temp[5].xyz, const[4].xyz_, temp[5].zzz_, temp[6].xyz_;
> 46: ADD temp[5].xyz, temp[5].xyz_, const[5].xyz_;
> 47: TEX temp[11].xyz, temp[5].xyz_, CUBE[0];
> 48: MOV temp[5].xyz, temp[11].xyz_;
> 49: ADD temp[6].x, const[17].x___, -const[1].x___;
> 50: DP3 temp[3].x, -temp[3].xyz_, temp[4].xyz_;
> 51: MAX temp[3].x, temp[3].x___, const[18].x___;
> 52: ADD temp[3].x, const[17].x___, -temp[3].x___;
> 53: LG2 temp[13].w, temp[3].___x;
> 54: MUL temp[13].w, temp[13].___w, const[1].___y;
> 55: EX2 temp[3].x, temp[13].w___;
> 56: MAD temp[3].x, temp[6].x___, temp[3].x___, const[1].x___;
> 57: MAX temp[3].x, temp[3].x___, const[18].x___;
> 58: MUL temp[3].xyz, temp[5].xyz_, temp[3].xxx_;
> 59: MUL temp[3].xyz, temp[3].xyz_, temp[2].xxx_;
> 60: MAD temp[1].xyz, temp[3].xyz_, const[10].xxx_, temp[1].xyz_;
> 61: DP3 temp[3].x, temp[4].xyz_, const[18].xxy_;
> 62: MOV temp[3].x, |temp[3].x___|;
> 63: ADD temp[3].x, const[17].x___, -temp[3].x___;
> 64: LG2 temp[14].w, temp[3].___x;
> 65: MUL temp[14].w, temp[14].___w, const[0].___y;
> 66: EX2 temp[3].x, temp[14].w___;
> 67: MUL temp[3].x, temp[3].x___, const[0].x___;
> 68: MUL temp[3].xyz, temp[5].xyz_, temp[3].xxx_;
> 69: MUL temp[2].xyz, temp[3].xyz_, temp[2].xxx_;
> 70: MAD temp[1].xyz, temp[2].xyz_, const[10].xxx_, temp[1].xyz_;
> 71: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[15].w, input[0].___w;
>  1: MUL temp[16].xy, input[0].xy__, temp[15].ww__;
>  2: MAD temp[17].xy, temp[16].xy__, const[19].xy__, const[20].xy__;
>  3: MOV temp[18].x, temp[17].x___;
>  4: MAD temp[18].y, temp[17]._y__, const[16]._x__, const[16]._y__;
>  5: TEX temp[19], input[1].xy__, 2D[3];
>  6: MUL temp[20], temp[19], input[3];
>  7: ADD temp[21].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[22].x, const[8].y___;
>  9: MUL_SAT temp[23].x, temp[21].x___, temp[22].x___;
> 10: LG2 temp[24].w, temp[23].___x;
> 11: MUL temp[25].w, temp[24].___w, const[6].___x;
> 12: EX2 temp[26].x, temp[25].w___;
> 13: ADD temp[27].x, const[17].x___, -temp[26].x___;
> 14: MAD temp[28].x, const[7].x___, temp[26].x___, temp[27].x___;
> 15: MUL temp[29].x, temp[28].x___, const[11].x___;
> 16: ADD temp[30].xyz, const[17].xxx_, -temp[20].xyz_;
> 17: ADD temp[31].x, const[17].x___, -temp[29].x___;
> 18: MAD temp[32].xyz, temp[30].xyz_, temp[31].xxx_, temp[20].xyz_;
> 19: TEX temp[33].xyz, input[1].xy__, 2D[2];
> 20: MOV temp[34].xyz, temp[33].xyz_;
> 21: MAD temp[35].xyz, temp[34].xyz_, const[17].yyy_, const[17].zzz_;
> 22: MUL temp[36].xyz, temp[35].xxx_, input[5].xyz_;
> 23: MAD temp[37].xyz, temp[35].yyy_, input[4].xyz_, temp[36].xyz_;
> 24: MAD temp[38].xyz, temp[35].zzz_, input[6].xyz_, temp[37].xyz_;
> 25: DP3 temp[39].x, temp[38].xyz_, temp[38].xyz_;
> 26: RSQ temp[40].x, |temp[39].x___|;
> 27: MUL temp[41].xyz, temp[38].xyz_, temp[40].xxx_;
> 28: RCP temp[42].x, input[2].z___;
> 29: MIN temp[43].x, temp[42].x___, const[17].w___;
> 30: MUL temp[44].x, const[9].x___, temp[43].x___;
> 31: MAD temp[45].xy, temp[35].xy__, temp[44].xx__, temp[18].xy__;
> 32: MUL temp[46].xy, temp[45].xy__, const[21].xy__;
> 33: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 34: MOV temp[48].xyz, temp[47].xyz_;
> 35: MUL temp[49].xyz, temp[32].xyz_, temp[48].xyz_;
> 36: DP3 temp[50].x, input[2].xyz_, input[2].xyz_;
> 37: RSQ temp[51].x, |temp[50].x___|;
> 38: MUL temp[52].xyz, input[2].xyz_, temp[51].xxx_;
> 39: DP3 temp[53].x, temp[41].xyz_, temp[52].xyz_;
> 40: MUL temp[54].xyz, temp[53].xxx_, temp[41].xyz_;
> 41: MUL temp[55].xyz, const[17].yyy_, temp[54].xyz_;
> 42: ADD temp[56].xyz, temp[52].xyz_, -temp[55].xyz_;
> 43: MUL temp[57].xyz, const[2].xyz_, temp[56].xxx_;
> 44: MAD temp[58].xyz, const[3].xyz_, temp[56].yyy_, temp[57].xyz_;
> 45: MAD temp[59].xyz, const[4].xyz_, temp[56].zzz_, temp[58].xyz_;
> 46: ADD temp[60].xyz, temp[59].xyz_, const[5].xyz_;
> 47: TEX temp[61].xyz, temp[60].xyz_, CUBE[0];
> 48: MOV temp[62].xyz, temp[61].xyz_;
> 49: ADD temp[63].x, const[17].x___, -const[1].x___;
> 50: DP3 temp[64].x, -temp[52].xyz_, temp[41].xyz_;
> 51: MAX temp[65].x, temp[64].x___, const[18].x___;
> 52: ADD temp[66].x, const[17].x___, -temp[65].x___;
> 53: LG2 temp[67].w, temp[66].___x;
> 54: MUL temp[68].w, temp[67].___w, const[1].___y;
> 55: EX2 temp[69].x, temp[68].w___;
> 56: MAD temp[70].x, temp[63].x___, temp[69].x___, const[1].x___;
> 57: MAX temp[71].x, temp[70].x___, const[18].x___;
> 58: MUL temp[72].xyz, temp[62].xyz_, temp[71].xxx_;
> 59: MUL temp[73].xyz, temp[72].xyz_, temp[29].xxx_;
> 60: MAD temp[74].xyz, temp[73].xyz_, const[10].xxx_, temp[49].xyz_;
> 61: DP3 temp[75].x, temp[41].xyz_, const[18].xxy_;
> 62: MOV temp[76].x, |temp[75].x___|;
> 63: ADD temp[77].x, const[17].x___, -temp[76].x___;
> 64: LG2 temp[78].w, temp[77].___x;
> 65: MUL temp[79].w, temp[78].___w, const[0].___y;
> 66: EX2 temp[80].x, temp[79].w___;
> 67: MUL temp[81].x, temp[80].x___, const[0].x___;
> 68: MUL temp[82].xyz, temp[62].xyz_, temp[81].xxx_;
> 69: MUL temp[83].xyz, temp[82].xyz_, temp[29].xxx_;
> 70: MAD temp[20].xyz, temp[83].xyz_, const[10].xxx_, temp[74].xyz_;
> 71: MOV output[0], temp[20];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[15].w, input[0].___w;
>  1: MUL temp[16].xy, input[0].xy__, temp[15].ww__;
>  2: MAD temp[17].xy, temp[16].xy__, const[19].xy__, const[20].xy__;
>  3: MOV temp[18].x, temp[17].x___;
>  4: MAD temp[18].y, temp[17]._y__, const[16]._x__, const[16]._y__;
>  5: TEX temp[19], input[1].xy__, 2D[3];
>  6: MUL temp[20], temp[19], input[3];
>  7: ADD temp[21].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[22].x, const[8].y___;
>  9: MUL_SAT temp[23].x, temp[21].x___, temp[22].x___;
> 10: LG2 temp[24].w, temp[23].___x;
> 11: MUL temp[25].w, temp[24].___w, const[6].___x;
> 12: EX2 temp[26].x, temp[25].w___;
> 13: ADD temp[27].x, none.1___, -temp[26].x___;
> 14: MAD temp[28].x, const[7].x___, temp[26].x___, temp[27].x___;
> 15: MUL temp[29].x, temp[28].x___, const[11].x___;
> 16: ADD temp[30].xyz, none.111_, -temp[20].xyz_;
> 17: ADD temp[31].x, none.1___, -temp[29].x___;
> 18: MAD temp[32].xyz, temp[30].xyz_, temp[31].xxx_, temp[20].xyz_;
> 19: TEX temp[33].xyz, input[1].xy__, 2D[2];
> 20: MAD temp[35].xyz, temp[33].xyz_, const[17].yyy_, none.-1-1-1_;
> 21: MUL temp[36].xyz, temp[35].xxx_, input[5].xyz_;
> 22: MAD temp[37].xyz, temp[35].yyy_, input[4].xyz_, temp[36].xyz_;
> 23: MAD temp[38].xyz, temp[35].zzz_, input[6].xyz_, temp[37].xyz_;
> 24: DP3 temp[39].x, temp[38].xyz_, temp[38].xyz_;
> 25: RSQ temp[40].x, |temp[39].x___|;
> 26: MUL temp[41].xyz, temp[38].xyz_, temp[40].xxx_;
> 27: RCP temp[42].x, input[2].z___;
> 28: MIN temp[43].x, temp[42].x___, const[17].w___;
> 29: MUL temp[44].x, const[9].x___, temp[43].x___;
> 30: MAD temp[45].xy, temp[35].xy__, temp[44].xx__, temp[18].xy__;
> 31: MUL temp[46].xy, temp[45].xy__, const[21].xy__;
> 32: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 33: MUL temp[49].xyz, temp[32].xyz_, temp[47].xyz_;
> 34: DP3 temp[50].x, input[2].xyz_, input[2].xyz_;
> 35: RSQ temp[51].x, |temp[50].x___|;
> 36: MUL temp[52].xyz, input[2].xyz_, temp[51].xxx_;
> 37: DP3 temp[53].x, temp[41].xyz_, temp[52].xyz_;
> 38: MUL temp[55].xyz * 2, temp[53].xxx_, temp[41].xyz_;
> 39: ADD temp[56].xyz, temp[52].xyz_, -temp[55].xyz_;
> 40: MUL temp[57].xyz, const[2].xyz_, temp[56].xxx_;
> 41: MAD temp[58].xyz, const[3].xyz_, temp[56].yyy_, temp[57].xyz_;
> 42: MAD temp[59].xyz, const[4].xyz_, temp[56].zzz_, temp[58].xyz_;
> 43: ADD temp[60].xyz, temp[59].xyz_, const[5].xyz_;
> 44: TEX temp[61].xyz, temp[60].xyz_, CUBE[0];
> 45: ADD temp[63].x, none.1___, -const[1].x___;
> 46: DP3 temp[64].x, -temp[52].xyz_, temp[41].xyz_;
> 47: MAX temp[65].x, temp[64].x___, none.0___;
> 48: ADD temp[66].x, none.1___, -temp[65].x___;
> 49: LG2 temp[67].w, temp[66].___x;
> 50: MUL temp[68].w, temp[67].___w, const[1].___y;
> 51: EX2 temp[69].x, temp[68].w___;
> 52: MAD temp[70].x, temp[63].x___, temp[69].x___, const[1].x___;
> 53: MAX temp[71].x, temp[70].x___, none.0___;
> 54: MUL temp[72].xyz, temp[61].xyz_, temp[71].xxx_;
> 55: MUL temp[73].xyz, temp[72].xyz_, temp[29].xxx_;
> 56: MAD temp[74].xyz, temp[73].xyz_, const[10].xxx_, temp[49].xyz_;
> 57: DP3 temp[75].x, temp[41].xyz_, none.00-1_;
> 58: ADD temp[77].x, none.1___, -|temp[75].x___|;
> 59: LG2 temp[78].w, temp[77].___x;
> 60: MUL temp[79].w, temp[78].___w, const[0].___y;
> 61: EX2 temp[80].x, temp[79].w___;
> 62: MUL temp[81].x, temp[80].x___, const[0].x___;
> 63: MUL temp[82].xyz, temp[61].xyz_, temp[81].xxx_;
> 64: MUL temp[83].xyz, temp[82].xyz_, temp[29].xxx_;
> 65: MAD temp[20].xyz, temp[83].xyz_, const[10].xxx_, temp[74].xyz_;
> 66: MOV output[0], temp[20];
>CONST[17] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[18] = {     0.0000    -1.0000     0.0000     0.0000 }
>CONST[22] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[15].w, input[0].___w;
>  1: MUL temp[16].xy, input[0].xy__, temp[15].ww__;
>  2: MAD temp[17].xy, temp[16].xy__, const[19].xy__, const[20].xy__;
>  3: MOV temp[18].x, temp[17].x___;
>  4: MAD temp[18].y, temp[17]._y__, const[16]._x__, const[16]._y__;
>  5: TEX temp[19], input[1].xy__, 2D[3];
>  6: MUL temp[20], temp[19], input[3];
>  7: ADD temp[21].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[22].x, const[8].y___;
>  9: MUL_SAT temp[23].x, temp[21].x___, temp[22].x___;
> 10: LG2 temp[24].w, temp[23].___x;
> 11: MUL temp[25].w, temp[24].___w, const[6].___x;
> 12: EX2 temp[26].x, temp[25].w___;
> 13: ADD temp[27].x, none.1___, -temp[26].x___;
> 14: MAD temp[28].x, const[7].x___, temp[26].x___, temp[27].x___;
> 15: MUL temp[29].x, temp[28].x___, const[11].x___;
> 16: ADD temp[30].xyz, none.111_, -temp[20].xyz_;
> 17: ADD temp[31].x, none.1___, -temp[29].x___;
> 18: MAD temp[32].xyz, temp[30].xyz_, temp[31].xxx_, temp[20].xyz_;
> 19: TEX temp[33].xyz, input[1].xy__, 2D[2];
> 20: MAD temp[35].xyz, temp[33].xyz_, const[17].yyy_, none.-1-1-1_;
> 21: MUL temp[36].xyz, temp[35].xxx_, input[5].xyz_;
> 22: MAD temp[37].xyz, temp[35].yyy_, input[4].xyz_, temp[36].xyz_;
> 23: MAD temp[38].xyz, temp[35].zzz_, input[6].xyz_, temp[37].xyz_;
> 24: DP3 temp[39].x, temp[38].xyz_, temp[38].xyz_;
> 25: RSQ temp[40].x, |temp[39].x___|;
> 26: MUL temp[41].xyz, temp[38].xyz_, temp[40].xxx_;
> 27: RCP temp[42].x, input[2].z___;
> 28: MIN temp[43].x, temp[42].x___, const[17].w___;
> 29: MUL temp[44].x, const[9].x___, temp[43].x___;
> 30: MAD temp[45].xy, temp[35].xy__, temp[44].xx__, temp[18].xy__;
> 31: MUL temp[46].xy, temp[45].xy__, const[21].xy__;
> 32: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 33: MUL temp[49].xyz, temp[32].xyz_, temp[47].xyz_;
> 34: DP3 temp[50].x, input[2].xyz_, input[2].xyz_;
> 35: RSQ temp[51].x, |temp[50].x___|;
> 36: MUL temp[52].xyz, input[2].xyz_, temp[51].xxx_;
> 37: DP3 temp[53].x, temp[41].xyz_, temp[52].xyz_;
> 38: MUL temp[55].xyz * 2, temp[53].xxx_, temp[41].xyz_;
> 39: ADD temp[56].xyz, temp[52].xyz_, -temp[55].xyz_;
> 40: MUL temp[57].xyz, const[2].xyz_, temp[56].xxx_;
> 41: MAD temp[58].xyz, const[3].xyz_, temp[56].yyy_, temp[57].xyz_;
> 42: MAD temp[59].xyz, const[4].xyz_, temp[56].zzz_, temp[58].xyz_;
> 43: ADD temp[60].xyz, temp[59].xyz_, const[5].xyz_;
> 44: TEX temp[61].xyz, temp[60].xyz_, CUBE[0];
> 45: ADD temp[63].x, none.1___, -const[1].x___;
> 46: DP3 temp[64].x, -temp[52].xyz_, temp[41].xyz_;
> 47: MAX temp[65].x, temp[64].x___, none.0___;
> 48: ADD temp[66].x, none.1___, -temp[65].x___;
> 49: LG2 temp[67].w, temp[66].___x;
> 50: MUL temp[68].w, temp[67].___w, const[1].___y;
> 51: EX2 temp[69].x, temp[68].w___;
> 52: MAD temp[70].x, temp[63].x___, temp[69].x___, const[1].x___;
> 53: MAX temp[71].x, temp[70].x___, none.0___;
> 54: MUL temp[72].xyz, temp[61].xyz_, temp[71].xxx_;
> 55: MUL temp[73].xyz, temp[72].xyz_, temp[29].xxx_;
> 56: MAD temp[74].xyz, temp[73].xyz_, const[10].xxx_, temp[49].xyz_;
> 57: DP3 temp[75].x, temp[41].xyz_, const[22].xyz_;
> 58: ADD temp[77].x, none.1___, -|temp[75].x___|;
> 59: LG2 temp[78].w, temp[77].___x;
> 60: MUL temp[79].w, temp[78].___w, const[0].___y;
> 61: EX2 temp[80].x, temp[79].w___;
> 62: MUL temp[81].x, temp[80].x___, const[0].x___;
> 63: MUL temp[82].xyz, temp[61].xyz_, temp[81].xxx_;
> 64: MUL temp[83].xyz, temp[82].xyz_, temp[29].xxx_;
> 65: MAD temp[20].xyz, temp[83].xyz_, const[10].xxx_, temp[74].xyz_;
> 66: MOV output[0], temp[20];
>CONST[13] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[17] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[15].w, input[0].___w;
>  1: MUL temp[16].xy, input[0].xy__, temp[15].ww__;
>  2: MAD temp[17].xy, temp[16].xy__, const[14].xy__, const[15].xy__;
>  3: MOV temp[18].x, temp[17].x___;
>  4: MAD temp[18].y, temp[17]._y__, const[12]._x__, const[12]._y__;
>  5: TEX temp[19], input[1].xy__, 2D[3];
>  6: MUL temp[20], temp[19], input[3];
>  7: ADD temp[21].x, -input[2].z___, -const[8].x___;
>  8: RCP temp[22].x, const[8].y___;
>  9: MUL_SAT temp[23].x, temp[21].x___, temp[22].x___;
> 10: LG2 temp[24].w, temp[23].___x;
> 11: MUL temp[25].w, temp[24].___w, const[6].___x;
> 12: EX2 temp[26].x, temp[25].w___;
> 13: ADD temp[27].x, none.1___, -temp[26].x___;
> 14: MAD temp[28].x, const[7].x___, temp[26].x___, temp[27].x___;
> 15: MUL temp[29].x, temp[28].x___, const[11].x___;
> 16: ADD temp[30].xyz, none.111_, -temp[20].xyz_;
> 17: ADD temp[31].x, none.1___, -temp[29].x___;
> 18: MAD temp[32].xyz, temp[30].xyz_, temp[31].xxx_, temp[20].xyz_;
> 19: TEX temp[33].xyz, input[1].xy__, 2D[2];
> 20: MAD temp[35].xyz, temp[33].xyz_, const[13].yyy_, none.-1-1-1_;
> 21: MUL temp[36].xyz, temp[35].xxx_, input[5].xyz_;
> 22: MAD temp[37].xyz, temp[35].yyy_, input[4].xyz_, temp[36].xyz_;
> 23: MAD temp[38].xyz, temp[35].zzz_, input[6].xyz_, temp[37].xyz_;
> 24: DP3 temp[39].x, temp[38].xyz_, temp[38].xyz_;
> 25: RSQ temp[40].x, |temp[39].x___|;
> 26: MUL temp[41].xyz, temp[38].xyz_, temp[40].xxx_;
> 27: RCP temp[42].x, input[2].z___;
> 28: MIN temp[43].x, temp[42].x___, const[13].w___;
> 29: MUL temp[44].x, const[9].x___, temp[43].x___;
> 30: MAD temp[45].xy, temp[35].xy__, temp[44].xx__, temp[18].xy__;
> 31: MUL temp[46].xy, temp[45].xy__, const[16].xy__;
> 32: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 33: MUL temp[49].xyz, temp[32].xyz_, temp[47].xyz_;
> 34: DP3 temp[50].x, input[2].xyz_, input[2].xyz_;
> 35: RSQ temp[51].x, |temp[50].x___|;
> 36: MUL temp[52].xyz, input[2].xyz_, temp[51].xxx_;
> 37: DP3 temp[53].x, temp[41].xyz_, temp[52].xyz_;
> 38: MUL temp[55].xyz * 2, temp[53].xxx_, temp[41].xyz_;
> 39: ADD temp[56].xyz, temp[52].xyz_, -temp[55].xyz_;
> 40: MUL temp[57].xyz, const[2].xyz_, temp[56].xxx_;
> 41: MAD temp[58].xyz, const[3].xyz_, temp[56].yyy_, temp[57].xyz_;
> 42: MAD temp[59].xyz, const[4].xyz_, temp[56].zzz_, temp[58].xyz_;
> 43: ADD temp[60].xyz, temp[59].xyz_, const[5].xyz_;
> 44: TEX temp[61].xyz, temp[60].xyz_, CUBE[0];
> 45: ADD temp[63].x, none.1___, -const[1].x___;
> 46: DP3 temp[64].x, -temp[52].xyz_, temp[41].xyz_;
> 47: MAX temp[65].x, temp[64].x___, none.0___;
> 48: ADD temp[66].x, none.1___, -temp[65].x___;
> 49: LG2 temp[67].w, temp[66].___x;
> 50: MUL temp[68].w, temp[67].___w, const[1].___y;
> 51: EX2 temp[69].x, temp[68].w___;
> 52: MAD temp[70].x, temp[63].x___, temp[69].x___, const[1].x___;
> 53: MAX temp[71].x, temp[70].x___, none.0___;
> 54: MUL temp[72].xyz, temp[61].xyz_, temp[71].xxx_;
> 55: MUL temp[73].xyz, temp[72].xyz_, temp[29].xxx_;
> 56: MAD temp[74].xyz, temp[73].xyz_, const[10].xxx_, temp[49].xyz_;
> 57: DP3 temp[75].x, temp[41].xyz_, const[17].xyz_;
> 58: ADD temp[77].x, none.1___, -|temp[75].x___|;
> 59: LG2 temp[78].w, temp[77].___x;
> 60: MUL temp[79].w, temp[78].___w, const[0].___y;
> 61: EX2 temp[80].x, temp[79].w___;
> 62: MUL temp[81].x, temp[80].x___, const[0].x___;
> 63: MUL temp[82].xyz, temp[61].xyz_, temp[81].xxx_;
> 64: MUL temp[83].xyz, temp[82].xyz_, temp[29].xxx_;
> 65: MAD temp[20].xyz, temp[83].xyz_, const[10].xxx_, temp[74].xyz_;
> 66: MOV output[0], temp[20];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[15].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[15]
>     MAD temp[16].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[16], src1.xyz = const[14], src2.xyz = const[15]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[17]
>     MAD temp[18].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[17], src1.xyz = const[12]
>     MAD temp[18].y, src0._y_, src1._x_, src1._y_
>  5: TEX temp[19], input[1].xy__, 2D[3];
>  6: src0.xyz = temp[19], src0.w = temp[19], src1.xyz = input[3], src1.w = input[3]
>     MAD temp[20].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[20].w, src0.w, src1.w, src0.0
>  7: src0.xyz = input[2], src1.xyz = const[8]
>     MAD temp[21].x, -src0.z__, src0.111, -src1.x__
>  8: src0.xyz = const[8]
>     REPL_ALPHA temp[22].x
>     RCP, src0.y
>  9: src0.xyz = temp[21], src1.xyz = temp[22]
>     MAD_SAT temp[23].x, src0.x__, src1.x__, src0.000
> 10: src0.xyz = temp[23]
>     LG2 temp[24].w, src0.x
> 11: src0.xyz = const[6], src0.w = temp[24]
>     MAD temp[25].w, src0.w, src0.x, src0.0
> 12: src0.w = temp[25]
>     REPL_ALPHA temp[26].x
>     EX2, src0.w
> 13: src0.xyz = temp[26]
>     MAD temp[27].x, src0.1__, src0.111, -src0.x__
> 14: src0.xyz = const[7], src1.xyz = temp[26], src2.xyz = temp[27]
>     MAD temp[28].x, src0.x__, src1.x__, src2.x__
> 15: src0.xyz = temp[28], src1.xyz = const[11]
>     MAD temp[29].x, src0.x__, src1.x__, src0.000
> 16: src0.xyz = temp[20]
>     MAD temp[30].xyz, src0.111, src0.111, -src0.xyz
> 17: src0.xyz = temp[29]
>     MAD temp[31].x, src0.1__, src0.111, -src0.x__
> 18: src0.xyz = temp[30], src1.xyz = temp[31], src2.xyz = temp[20]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src2.xyz
> 19: TEX temp[33].xyz, input[1].xy__, 2D[2];
> 20: src0.xyz = temp[33], src1.xyz = const[13]
>     MAD temp[35].xyz, src0.xyz, src1.yyy, -src0.111
> 21: src0.xyz = temp[35], src1.xyz = input[5]
>     MAD temp[36].xyz, src0.xxx, src1.xyz, src0.000
> 22: src0.xyz = temp[35], src1.xyz = input[4], src2.xyz = temp[36]
>     MAD temp[37].xyz, src0.yyy, src1.xyz, src2.xyz
> 23: src0.xyz = temp[35], src1.xyz = input[6], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.zzz, src1.xyz, src2.xyz
> 24: src0.xyz = temp[38]
>     DP3 temp[39].x, src0.xyz, src0.xyz
> 25: src0.xyz = temp[39]
>     REPL_ALPHA temp[40].x
>     RSQ, |src0.x|
> 26: src0.xyz = temp[38], src1.xyz = temp[40]
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = input[2]
>     REPL_ALPHA temp[42].x
>     RCP, src0.z
> 28: src0.xyz = temp[42], src0.w = const[13]
>     MIN temp[43].x, src0.x__, src0.w__
> 29: src0.xyz = const[9], src1.xyz = temp[43]
>     MAD temp[44].x, src0.x__, src1.x__, src0.000
> 30: src0.xyz = temp[35], src1.xyz = temp[44], src2.xyz = temp[18]
>     MAD temp[45].xy, src0.xy_, src1.xx_, src2.xy_
> 31: src0.xyz = temp[45], src1.xyz = const[16]
>     MAD temp[46].xy, src0.xy_, src1.xy_, src0.000
> 32: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 33: src0.xyz = temp[32], src1.xyz = temp[47]
>     MAD temp[49].xyz, src0.xyz, src1.xyz, src0.000
> 34: src0.xyz = input[2]
>     DP3 temp[50].x, src0.xyz, src0.xyz
> 35: src0.xyz = temp[50]
>     REPL_ALPHA temp[51].x
>     RSQ, |src0.x|
> 36: src0.xyz = input[2], src1.xyz = temp[51]
>     MAD temp[52].xyz, src0.xyz, src1.xxx, src0.000
> 37: src0.xyz = temp[41], src1.xyz = temp[52]
>     DP3 temp[53].x, src0.xyz, src1.xyz
> 38: src0.xyz = temp[53], src1.xyz = temp[41]
>     MAD temp[55].xyz * 2, src0.xxx, src1.xyz, src0.000
> 39: src0.xyz = temp[52], src1.xyz = temp[55]
>     MAD temp[56].xyz, src0.xyz, src0.111, -src1.xyz
> 40: src0.xyz = const[2], src1.xyz = temp[56]
>     MAD temp[57].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = const[3], src1.xyz = temp[56], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.yyy, src2.xyz
> 42: src0.xyz = const[4], src1.xyz = temp[56], src2.xyz = temp[58]
>     MAD temp[59].xyz, src0.xyz, src1.zzz, src2.xyz
> 43: src0.xyz = temp[59], src1.xyz = const[5]
>     MAD temp[60].xyz, src0.xyz, src0.111, src1.xyz
> 44: TEX temp[61].xyz, temp[60].xyz_, CUBE[0];
> 45: src0.xyz = const[1]
>     MAD temp[63].x, src0.1__, src0.111, -src0.x__
> 46: src0.xyz = temp[52], src1.xyz = temp[41]
>     DP3 temp[64].x, -src0.xyz, src1.xyz
> 47: src0.xyz = temp[64]
>     MAX temp[65].x, src0.x__, src0.0__
> 48: src0.xyz = temp[65]
>     MAD temp[66].x, src0.1__, src0.111, -src0.x__
> 49: src0.xyz = temp[66]
>     LG2 temp[67].w, src0.x
> 50: src0.xyz = const[1], src0.w = temp[67]
>     MAD temp[68].w, src0.w, src0.y, src0.0
> 51: src0.w = temp[68]
>     REPL_ALPHA temp[69].x
>     EX2, src0.w
> 52: src0.xyz = temp[63], src1.xyz = temp[69], src2.xyz = const[1]
>     MAD temp[70].x, src0.x__, src1.x__, src2.x__
> 53: src0.xyz = temp[70]
>     MAX temp[71].x, src0.x__, src0.0__
> 54: src0.xyz = temp[61], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 55: src0.xyz = temp[72], src1.xyz = temp[29]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 56: src0.xyz = temp[73], src1.xyz = const[10], src2.xyz = temp[49]
>     MAD temp[74].xyz, src0.xyz, src1.xxx, src2.xyz
> 57: src0.xyz = temp[41], src1.xyz = const[17]
>     DP3 temp[75].x, src0.xyz, src1.xyz
> 58: src0.xyz = temp[75]
>     MAD temp[77].x, src0.1__, |src0.111|, -|src0.x__|
> 59: src0.xyz = temp[77]
>     LG2 temp[78].w, src0.x
> 60: src0.xyz = const[0], src0.w = temp[78]
>     MAD temp[79].w, src0.w, src0.y, src0.0
> 61: src0.w = temp[79]
>     REPL_ALPHA temp[80].x
>     EX2, src0.w
> 62: src0.xyz = temp[80], src1.xyz = const[0]
>     MAD temp[81].x, src0.x__, src1.x__, src0.000
> 63: src0.xyz = temp[61], src1.xyz = temp[81]
>     MAD temp[82].xyz, src0.xyz, src1.xxx, src0.000
> 64: src0.xyz = temp[82], src1.xyz = temp[29]
>     MAD temp[83].xyz, src0.xyz, src1.xxx, src0.000
> 65: src0.xyz = temp[83], src1.xyz = const[10], src2.xyz = temp[74]
>     MAD temp[20].xyz, src0.xyz, src1.xxx, src2.xyz
> 66: src0.xyz = temp[20], src0.w = temp[20]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = input[2]
>     MAD temp[63].x, src0.1__, src0.111, -src0.x__
>     RCP temp[42].w, src1.z
>  1: src0.xyz = input[2], src1.xyz = const[8]
>     DP3 temp[50].x, src0.xyz, src0.xyz
>     RCP temp[22].w, src1.y
>  2: src0.xyz = temp[42], src0.w = const[13], src1.xyz = temp[50], src1.w = temp[42]
>     MIN temp[43].x, src1.w__, src0.w__
>     RSQ temp[51].w, |src1.x|
>  3: src0.xyz = input[2], src0.w = input[0], src1.xyz = const[8]
>     MAD temp[21].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[15].w, src0.w
>  4: src0.xyz = input[2], src0.w = temp[51], src1.xyz = temp[51], src1.w = temp[22], src2.xyz = temp[21]
>     MAD temp[52].xyz, src0.xyz, src0.www, src0.000
>     MAD_SAT temp[23].w, src2.x, src1.w, src0.0
>  5: src0.xyz = input[0], src0.w = temp[15], src1.xyz = const[9], src2.xyz = temp[43]
>     MAD temp[16].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[44].w, src1.x, src2.x, src0.0
>  6: src0.xyz = temp[16], src0.w = temp[23], src1.xyz = const[14], src2.xyz = const[15]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src2.xy_
>     LG2 temp[24].w, src0.w
>  7: src0.xyz = temp[17], src0.w = temp[24], src1.xyz = const[6]
>     MAD temp[18].x, src0.x__, src0.111, src0.000
>     MAD temp[25].w, src0.w, src1.x, src0.0
>  8: src0.xyz = temp[17], src1.xyz = const[12]
>     MAD temp[18].y, src0._y_, src1._x_, src1._y_
>  9: src0.w = temp[25]
>     REPL_ALPHA temp[26].x
>     EX2, src0.w
> 10: src0.xyz = temp[26]
>     MAD temp[27].x, src0.1__, src0.111, -src0.x__
> 11: src0.xyz = const[7], src1.xyz = temp[26], src2.xyz = temp[27]
>     MAD temp[28].x, src0.x__, src1.x__, src2.x__
> 12: src0.xyz = temp[28], src1.xyz = const[11]
>     MAD temp[29].x, src0.x__, src1.x__, src0.000
> 13: src0.xyz = temp[29]
>     MAD temp[31].x, src0.1__, src0.111, -src0.x__
> 14: BEGIN_TEX;
> 15: TEX temp[19], input[1].xy__, 2D[3];
> 16: TEX temp[33].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 17: src0.xyz = temp[19], src0.w = temp[19], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[20].w, src0.w, src1.w, src0.0
> 18: src0.xyz = temp[33], src1.xyz = const[13]
>     MAD temp[35].xyz, src0.xyz, src1.yyy, -src0.111
> 19: src0.xyz = temp[35], src0.w = temp[44], src1.xyz = temp[44], src2.xyz = temp[18]
>     MAD temp[45].xy, src0.xy_, src0.ww_, src2.xy_
> 20: src0.xyz = temp[35], src1.xyz = input[5]
>     MAD temp[36].xyz, src0.xxx, src1.xyz, src0.000
> 21: src0.xyz = temp[20]
>     MAD temp[30].xyz, src0.111, src0.111, -src0.xyz
> 22: src0.xyz = temp[45], src1.xyz = const[16]
>     MAD temp[46].xy, src0.xy_, src1.xy_, src0.000
> 23: src0.xyz = temp[35], src1.xyz = input[4], src2.xyz = temp[36]
>     MAD temp[37].xyz, src0.yyy, src1.xyz, src2.xyz
> 24: src0.xyz = temp[30], src1.xyz = temp[31], src2.xyz = temp[20]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[35], src1.xyz = input[6], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.zzz, src1.xyz, src2.xyz
> 26: src0.xyz = temp[38]
>     DP3 temp[39].x, src0.xyz, src0.xyz
> 27: src0.xyz = temp[39]
>     REPL_ALPHA temp[40].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[38], src1.xyz = temp[40]
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[41], src1.xyz = const[17]
>     DP3 temp[75].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[52], src1.xyz = temp[41], src2.xyz = temp[75]
>     DP3 temp[64].x, -src0.xyz, src1.xyz
>     MAD temp[77].w, src0.1, |src0.1|, -|src2.x|
> 31: src0.xyz = temp[41], src0.w = temp[77], src1.xyz = temp[52]
>     DP3 temp[53].x, src0.xyz, src1.xyz
>     LG2 temp[78].w, src0.w
> 32: src0.xyz = temp[64], src0.w = temp[78], src1.xyz = const[0]
>     MAX temp[65].x, src0.x__, src0.0__
>     MAD temp[79].w, src0.w, src1.y, src0.0
> 33: src0.xyz = temp[53], src1.xyz = temp[41], src2.xyz = temp[65]
>     MAD temp[55].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[66].w, src0.1, src0.1, -src2.x
> 34: src0.xyz = temp[52], src0.w = temp[66], src1.xyz = temp[55]
>     MAD temp[56].xyz, src0.xyz, src0.111, -src1.xyz
>     LG2 temp[67].w, src0.w
> 35: src0.xyz = const[2], src0.w = temp[79], src1.xyz = temp[56]
>     MAD temp[57].xyz, src0.xyz, src1.xxx, src0.000
>     EX2 temp[80].w, src0.w
> 36: src0.xyz = temp[80], src0.w = temp[80], src1.xyz = const[0], src1.w = temp[67], src2.xyz = const[1]
>     MAD temp[81].x, src0.w__, src1.x__, src0.000
>     MAD temp[68].w, src1.w, src2.y, src0.0
> 37: src0.xyz = const[3], src1.xyz = temp[56], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.yyy, src2.xyz
> 38: src0.xyz = const[4], src1.xyz = temp[56], src2.xyz = temp[58]
>     MAD temp[59].xyz, src0.xyz, src1.zzz, src2.xyz
> 39: src0.w = temp[68]
>     REPL_ALPHA temp[69].x
>     EX2, src0.w
> 40: src0.xyz = temp[59], src1.xyz = const[5]
>     MAD temp[60].xyz, src0.xyz, src0.111, src1.xyz
> 41: src0.xyz = temp[63], src1.xyz = temp[69], src2.xyz = const[1]
>     MAD temp[70].w, src0.x, src1.x, src2.x
> 42: src0.xyz = temp[70], src0.w = temp[70]
>     MAX temp[71].x, src0.w__, src0.0__
> 43: BEGIN_TEX;
> 44: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 45: TEX temp[61].xyz, temp[60].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 46: src0.xyz = temp[32], src1.xyz = temp[47] SEM_WAIT
>     MAD temp[49].xyz, src0.xyz, src1.xyz, src0.000
> 47: src0.xyz = temp[61], src1.xyz = temp[81]
>     MAD temp[82].xyz, src0.xyz, src1.xxx, src0.000
> 48: src0.xyz = temp[61], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 49: src0.xyz = temp[82], src1.xyz = temp[29]
>     MAD temp[83].xyz, src0.xyz, src1.xxx, src0.000
> 50: src0.xyz = temp[72], src1.xyz = temp[29]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 51: src0.xyz = temp[73], src1.xyz = const[10], src2.xyz = temp[49]
>     MAD temp[74].xyz, src0.xyz, src1.xxx, src2.xyz
> 52: src0.xyz = temp[83], src1.xyz = const[10], src2.xyz = temp[74]
>     MAD temp[20].xyz, src0.xyz, src1.xxx, src2.xyz
> 53: src0.xyz = temp[20], src0.w = temp[20]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = input[2]
>     MAD temp[63].x, src0.1__, src0.111, -src0.x__
>     RCP temp[42].w, src1.z
>  1: src0.xyz = input[2], src1.xyz = const[8]
>     DP3 temp[50].x, src0.xyz, src0.xyz
>     RCP temp[22].w, src1.y
>  2: src0.w = const[13], src1.xyz = temp[50], src1.w = temp[42]
>     MIN temp[43].x, src1.w__, src0.w__
>     RSQ temp[51].w, |src1.x|
>  3: src0.xyz = input[2], src0.w = input[0], src1.xyz = const[8]
>     MAD temp[21].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[15].w, src0.w
>  4: src0.xyz = input[2], src0.w = temp[51], src1.w = temp[22], src2.xyz = temp[21]
>     MAD temp[52].xyz, src0.xyz, src0.www, src0.000
>     MAD_SAT temp[23].w, src2.x, src1.w, src0.0
>  5: src0.xyz = input[0], src0.w = temp[15], src1.xyz = const[9], src2.xyz = temp[43]
>     MAD temp[16].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[44].w, src1.x, src2.x, src0.0
>  6: src0.xyz = temp[16], src0.w = temp[23], src1.xyz = const[14], src2.xyz = const[15]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src2.xy_
>     LG2 temp[24].w, src0.w
>  7: src0.xyz = temp[17], src0.w = temp[24], src1.xyz = const[6]
>     MAD temp[18].x, src0.x__, src0.111, src0.000
>     MAD temp[25].w, src0.w, src1.x, src0.0
>  8: src0.xyz = temp[17], src1.xyz = const[12]
>     MAD temp[18].y, src0._y_, src1._x_, src1._y_
>  9: src0.w = temp[25]
>     REPL_ALPHA temp[26].x
>     EX2, src0.w
> 10: src0.xyz = temp[26]
>     MAD temp[27].x, src0.1__, src0.111, -src0.x__
> 11: src0.xyz = const[7], src1.xyz = temp[26], src2.xyz = temp[27]
>     MAD temp[28].x, src0.x__, src1.x__, src2.x__
> 12: src0.xyz = temp[28], src1.xyz = const[11]
>     MAD temp[29].x, src0.x__, src1.x__, src0.000
> 13: src0.xyz = temp[29]
>     MAD temp[31].x, src0.1__, src0.111, -src0.x__
> 14: BEGIN_TEX;
> 15: TEX temp[19], input[1].xy__, 2D[3];
> 16: TEX temp[33].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 17: src0.xyz = temp[19], src0.w = temp[19], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[20].w, src0.w, src1.w, src0.0
> 18: src0.xyz = temp[33], src1.xyz = const[13]
>     MAD temp[35].xyz, src0.xyz, src1.yyy, -src0.111
> 19: src0.xyz = temp[35], src0.w = temp[44], src2.xyz = temp[18]
>     MAD temp[45].xy, src0.xy_, src0.ww_, src2.xy_
> 20: src0.xyz = temp[35], src1.xyz = input[5]
>     MAD temp[36].xyz, src0.xxx, src1.xyz, src0.000
> 21: src0.xyz = temp[20]
>     MAD temp[30].xyz, src0.111, src0.111, -src0.xyz
> 22: src0.xyz = temp[45], src1.xyz = const[16]
>     MAD temp[46].xy, src0.xy_, src1.xy_, src0.000
> 23: src0.xyz = temp[35], src1.xyz = input[4], src2.xyz = temp[36]
>     MAD temp[37].xyz, src0.yyy, src1.xyz, src2.xyz
> 24: src0.xyz = temp[30], src1.xyz = temp[31], src2.xyz = temp[20]
>     MAD temp[32].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[35], src1.xyz = input[6], src2.xyz = temp[37]
>     MAD temp[38].xyz, src0.zzz, src1.xyz, src2.xyz
> 26: src0.xyz = temp[38]
>     DP3 temp[39].x, src0.xyz, src0.xyz
> 27: src0.xyz = temp[39]
>     REPL_ALPHA temp[40].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[38], src1.xyz = temp[40]
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[41], src1.xyz = const[17]
>     DP3 temp[75].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[52], src1.xyz = temp[41], src2.xyz = temp[75]
>     DP3 temp[64].x, -src0.xyz, src1.xyz
>     MAD temp[77].w, src0.1, |src0.1|, -|src2.x|
> 31: src0.xyz = temp[41], src0.w = temp[77], src1.xyz = temp[52]
>     DP3 temp[53].x, src0.xyz, src1.xyz
>     LG2 temp[78].w, src0.w
> 32: src0.xyz = temp[64], src0.w = temp[78], src1.xyz = const[0]
>     MAX temp[65].x, src0.x__, src0.0__
>     MAD temp[79].w, src0.w, src1.y, src0.0
> 33: src0.xyz = temp[53], src1.xyz = temp[41], src2.xyz = temp[65]
>     MAD temp[55].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[66].w, src0.1, src0.1, -src2.x
> 34: src0.xyz = temp[52], src0.w = temp[66], src1.xyz = temp[55]
>     MAD temp[56].xyz, src0.xyz, src0.111, -src1.xyz
>     LG2 temp[67].w, src0.w
> 35: src0.xyz = const[2], src0.w = temp[79], src1.xyz = temp[56]
>     MAD temp[57].xyz, src0.xyz, src1.xxx, src0.000
>     EX2 temp[80].w, src0.w
> 36: src0.w = temp[80], src1.xyz = const[0], src1.w = temp[67], src2.xyz = const[1]
>     MAD temp[81].x, src0.w__, src1.x__, src0.000
>     MAD temp[68].w, src1.w, src2.y, src0.0
> 37: src0.xyz = const[3], src1.xyz = temp[56], src2.xyz = temp[57]
>     MAD temp[58].xyz, src0.xyz, src1.yyy, src2.xyz
> 38: src0.xyz = const[4], src1.xyz = temp[56], src2.xyz = temp[58]
>     MAD temp[59].xyz, src0.xyz, src1.zzz, src2.xyz
> 39: src0.w = temp[68]
>     REPL_ALPHA temp[69].x
>     EX2, src0.w
> 40: src0.xyz = temp[59], src1.xyz = const[5]
>     MAD temp[60].xyz, src0.xyz, src0.111, src1.xyz
> 41: src0.xyz = temp[63], src1.xyz = temp[69], src2.xyz = const[1]
>     MAD temp[70].w, src0.x, src1.x, src2.x
> 42: src0.w = temp[70]
>     MAX temp[71].x, src0.w__, src0.0__
> 43: BEGIN_TEX;
> 44: TEX temp[47].xyz, temp[46].xy__, 2D[1];
> 45: TEX temp[61].xyz, temp[60].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 46: src0.xyz = temp[32], src1.xyz = temp[47] SEM_WAIT
>     MAD temp[49].xyz, src0.xyz, src1.xyz, src0.000
> 47: src0.xyz = temp[61], src1.xyz = temp[81]
>     MAD temp[82].xyz, src0.xyz, src1.xxx, src0.000
> 48: src0.xyz = temp[61], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 49: src0.xyz = temp[82], src1.xyz = temp[29]
>     MAD temp[83].xyz, src0.xyz, src1.xxx, src0.000
> 50: src0.xyz = temp[72], src1.xyz = temp[29]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 51: src0.xyz = temp[73], src1.xyz = const[10], src2.xyz = temp[49]
>     MAD temp[74].xyz, src0.xyz, src1.xxx, src2.xyz
> 52: src0.xyz = temp[83], src1.xyz = const[10], src2.xyz = temp[74]
>     MAD temp[20].xyz, src0.xyz, src1.xxx, src2.xyz
> 53: src0.xyz = temp[20], src0.w = temp[20]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = input[1]
>     MAD temp[0].z, src0.__1, src0.__1, -src0.__x
>     RCP temp[0].w, src1.z
>  1: src0.xyz = input[1], src1.xyz = const[8]
>     DP3 temp[6].z, src0.xyz, src0.xyz
>     RCP temp[1].w, src1.y
>  2: src0.w = const[13], src1.xyz = temp[6], src1.w = temp[0]
>     MIN temp[6].z, src1.__w, src0.__w
>     RSQ temp[0].w, |src1.z|
>  3: src0.xyz = input[1], src0.w = input[6], src1.xyz = const[8]
>     MAD temp[7].x, -src0.z__, src0.1__, -src1.x__
>     RCP temp[3].w, src0.w
>  4: src0.xyz = input[1], src0.w = temp[0], src1.w = temp[1], src2.xyz = temp[7]
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
>     MAD_SAT temp[0].w, src2.x, src1.w, src0.0
>  5: src0.xyz = input[6], src0.w = temp[3], src1.xyz = const[9], src2.xyz = temp[6]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.00_
>     MAD temp[1].w, src1.x, src2.z, src0.0
>  6: src0.xyz = temp[6], src0.w = temp[0], src1.xyz = const[14], src2.xyz = const[15]
>     MAD temp[6].xy, src0.xy_, src1.xy_, src2.xy_
>     LG2 temp[0].w, src0.w
>  7: src0.xyz = temp[6], src0.w = temp[0], src1.xyz = const[6]
>     MAD temp[7].x, src0.x__, src0.11_, src0.00_
>     MAD temp[0].w, src0.w, src1.x, src0.0
>  8: src0.xyz = temp[6], src1.xyz = const[12]
>     MAD temp[7].y, src0._y_, src1._x_, src1._y_
>  9: src0.w = temp[0]
>     REPL_ALPHA temp[6].x
>     EX2, src0.w
> 10: src0.xyz = temp[6]
>     MAD temp[6].y, src0._1_, src0._1_, -src0._x_
> 11: src0.xyz = const[7], src1.xyz = temp[6], src2.xyz = temp[6]
>     MAD temp[6].x, src0.x__, src1.x__, src2.y__
> 12: src0.xyz = temp[6], src1.xyz = const[11]
>     MAD temp[6].x, src0.x__, src1.x__, src0.0__
> 13: src0.xyz = temp[6]
>     MAD temp[6].y, src0._1_, src0._1_, -src0._x_
> 14: BEGIN_TEX;
> 15: TEX temp[8], input[0].xy__, 2D[3];
> 16: TEX temp[9].xyz, input[0].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 17: src0.xyz = temp[8], src0.w = temp[8], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
> 18: src0.xyz = temp[9], src1.xyz = const[13]
>     MAD temp[8].xyz, src0.xyz, src1.yyy, -src0.111
> 19: src0.xyz = temp[8], src0.w = temp[1], src2.xyz = temp[7]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src2.xy_
> 20: src0.xyz = temp[8], src1.xyz = input[4]
>     MAD temp[4].xyz, src0.xxx, src1.xyz, src0.000
> 21: src0.xyz = temp[2]
>     MAD temp[7].xyz, src0.111, src0.111, -src0.xyz
> 22: src0.xyz = temp[0], src1.xyz = const[16]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
> 23: src0.xyz = temp[8], src1.xyz = input[3], src2.xyz = temp[4]
>     MAD temp[3].xyz, src0.yyy, src1.xyz, src2.xyz
> 24: src0.xyz = temp[7], src1.xyz = temp[6], src2.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src1.yyy, src2.xyz
> 25: src0.xyz = temp[8], src1.xyz = input[5], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.zzz, src1.xyz, src2.xyz
> 26: src0.xyz = temp[3]
>     DP3 temp[4].x, src0.xyz, src0.xyz
> 27: src0.xyz = temp[4]
>     REPL_ALPHA temp[4].x
>     RSQ, |src0.x|
> 28: src0.xyz = temp[3], src1.xyz = temp[4]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[3], src1.xyz = const[17]
>     DP3 temp[4].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[1], src1.xyz = temp[3], src2.xyz = temp[4]
>     DP3 temp[4].x, -src0.xyz, src1.xyz
>     MAD temp[1].w, src0.1, |src0.1|, -|src2.x|
> 31: src0.xyz = temp[3], src0.w = temp[1], src1.xyz = temp[1]
>     DP3 temp[4].y, src0.xyz, src1.xyz
>     LG2 temp[1].w, src0.w
> 32: src0.xyz = temp[4], src0.w = temp[1], src1.xyz = const[0]
>     MAX temp[4].x, src0.x__, src0.0__
>     MAD temp[1].w, src0.w, src1.y, src0.0
> 33: src0.xyz = temp[4], src1.xyz = temp[3], src2.xyz = temp[4]
>     MAD temp[3].xyz * 2, src0.yyy, src1.xyz, src0.000
>     MAD temp[2].w, src0.1, src0.1, -src2.x
> 34: src0.xyz = temp[1], src0.w = temp[2], src1.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src0.111, -src1.xyz
>     LG2 temp[2].w, src0.w
> 35: src0.xyz = const[2], src0.w = temp[1], src1.xyz = temp[1]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
>     EX2 temp[1].w, src0.w
> 36: src0.w = temp[1], src1.xyz = const[0], src1.w = temp[2], src2.xyz = const[1]
>     MAD temp[4].x, src0.w__, src1.x__, src0.0__
>     MAD temp[1].w, src1.w, src2.y, src0.0
> 37: src0.xyz = const[3], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.xyz, src1.yyy, src2.xyz
> 38: src0.xyz = const[4], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src1.zzz, src2.xyz
> 39: src0.w = temp[1]
>     REPL_ALPHA temp[3].x
>     EX2, src0.w
> 40: src0.xyz = temp[1], src1.xyz = const[5]
>     MAD temp[1].xyz, src0.xyz, src0.111, src1.xyz
> 41: src0.xyz = temp[0], src1.xyz = temp[3], src2.xyz = const[1]
>     MAD temp[1].w, src0.z, src1.x, src2.x
> 42: src0.w = temp[1]
>     MAX temp[0].z, src0.__w, src0.__0
> 43: BEGIN_TEX;
> 44: TEX temp[3].xyz, temp[0].xy__, 2D[1];
> 45: TEX temp[1].xyz, temp[1].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 46: src0.xyz = temp[2], src1.xyz = temp[3] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
> 47: src0.xyz = temp[1], src1.xyz = temp[4]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
> 48: src0.xyz = temp[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 49: src0.xyz = temp[3], src1.xyz = temp[6]
>     MAD temp[1].xyz, src0.xyz, src1.xxx, src0.000
> 50: src0.xyz = temp[0], src1.xyz = temp[6]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 51: src0.xyz = temp[0], src1.xyz = const[10], src2.xyz = temp[2]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 52: src0.xyz = temp[1], src1.xyz = const[10], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 53: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=37*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 13, tex_end: 0  (code_addr: 00000340)
>  0: xyz:  c1  t1  t0  bias-> t0.z                 (02000061)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:      1.0      1.0  -c1.xxx    op: 00084a95 
>       w:     t1.z     c1.x     c1.x    op: 05000005
>  1: xyz:  t1  c8  t0  bias-> t6.z                 (02180a01)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   t1.xyz   t1.xxx    op: 00804000 
>       w:     c8.y     t1.x     t1.x    op: 05000004
>  2: xyz: t32  t6  t0  bias-> t6.z                 (02180180)
>       w: c13  t0  t0  bias-> t0.w                 (0080002d)
>     xyz:   t0.www  c13.www  t32.xxx    op: 0200460d 
>       w:   |t6.z|    t32.x    t32.x    op: 05800045
>  3: xyz:  t1  c8  t0  bias-> t7.x                 (009c0a01)
>       w:  t6  t0  t0  bias-> t3.w                 (008c0006)
>     xyz:  -t1.zzz      1.0  -c8.xyz    op: 00090aa3 
>       w:     t6.w     t1.x     t1.x    op: 05000009
>  4: xyz:  t1 t32  t7  bias-> t1.xyz               (03847001)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     t7.x     t1.w      0.0    op: 40040506
>  5: xyz:  t6  c9  t6  bias-> t6.xy                (01986a46)
>       w:  t3  t0  t0  bias-> t1.w                 (00840003)
>     xyz:   t6.xyz   t3.www      0.0    op: 00050600 
>       w:     c9.x     t6.z      0.0    op: 00040403
>  6: xyz:  t6 c14 c15  bias-> t6.xy                (019afb86)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t6.xyz  c14.xyz  c15.xyz    op: 00020200 
>       w:     t0.w     t6.x     t6.x    op: 04800009
>  7: xyz:  t6  c6  t0  bias-> t7.x                 (009c0986)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t6.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w     c6.x      0.0    op: 00040189
>  8: xyz:  t6 c12  t0  bias-> t7.y                 (011c0b06)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz  c12.xxx  c12.xyz    op: 00010280 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  9: xyz:  t0  t0  t0  bias-> t6.x                 (00980000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t0.w     t0.x     t0.x    op: 04000009
> 10: xyz:  t6  t0  t0  bias-> t6.y                 (01180006)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t6.xxx    op: 00084a95 
>       w:     t6.x     t6.x     t6.x    op: 00000000
> 11: xyz:  c7  t6  t6  bias-> t6.x                 (009861a7)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c7.xyz   t6.xyz   t6.yyy    op: 00028200 
>       w:     c7.x     c7.x     c7.x    op: 00000000
> 12: xyz:  t6 c11  t0  bias-> t6.x                 (00980ac6)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz  c11.xyz      0.0    op: 00050200 
>       w:     t6.x     t6.x     t6.x    op: 00000000
> 13: xyz:  t6  t0  t0  bias-> t6.y                 (01180006)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t6.xxx    op: 00084a95 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>NODE 1: alu_offset: 14, tex_offset: 0, alu_end: 25, tex_end: 1  (code_addr: 0002064e)
>  TEX:
>    TEX t8, t0, texture[3]   (00009a00)
>    TEX t9, t0, texture[2]   (00009240)
> 14: xyz:  t8  t2  t0  bias-> t2.xyz               (03880088)
>       w:  t8  t2  t0  bias-> t0.w                 (00800088)
>     xyz:   t8.xyz   t2.xyz      0.0    op: 00050200 
>       w:     t8.w     t2.w      0.0    op: 00040509
> 15: xyz:  t9 c13  t0  bias-> t8.xyz               (03a00b49)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t9.xyz  c13.yyy     -1.0    op: 000d4300 
>       w:     t9.x     t9.x     t9.x    op: 00000000
> 16: xyz:  t8 t32  t7  bias-> t0.xy                (01807008)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t8.xyz   t1.www   t7.xyz    op: 00020600 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 17: xyz:  t8  t4  t0  bias-> t4.xyz               (03900108)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.xxx   t4.xyz      0.0    op: 00050201 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 18: xyz:  t2  t0  t0  bias-> t7.xyz               (039c0002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t2.xyz    op: 00080a95 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 19: xyz:  t0 c16  t0  bias-> t0.xy                (01800c00)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c16.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 20: xyz:  t8  t3  t4  bias-> t3.xyz               (038c40c8)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.yyy   t3.xyz   t4.xyz    op: 00020202 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 21: xyz:  t7  t6  t2  bias-> t2.xyz               (03882187)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t7.xyz   t6.yyy   t2.xyz    op: 00020300 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 22: xyz:  t8  t5  t3  bias-> t3.xyz               (038c3148)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.zzz   t5.xyz   t3.xyz    op: 00020203 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 23: xyz:  t3  t0  t0  bias-> t4.x                 (00900003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 24: xyz:  t4  t0  t0  bias-> t4.x                 (00900004)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xxx   t4.xxx   t4.xxx    op: 05004081 
>       w:   |t4.x|     t4.x     t4.x    op: 05800040
> 25: xyz:  t3  t4  t0  bias-> t3.xyz               (038c0103)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 26: xyz:  t3 c17  t0  bias-> t4.x                 (00900c43)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz  c17.xyz   t3.xxx    op: 00804200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 27: xyz:  t1  t3  t4  bias-> t4.x                 (009040c1)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:  -t1.xyz   t3.xyz   t1.xxx    op: 00804220 
>       w:      1.0    |1.0|  -|t4.x|    op: 0019a891
> 28: xyz:  t3  t1  t0  bias-> t4.y                 (01100043)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t3.xyz   t1.xyz   t3.xxx    op: 00804200 
>       w:     t1.w     t3.x     t3.x    op: 04800009
> 29: xyz:  t4  c0  t0  bias-> t4.x                 (00900804)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t4.xyz      0.0   t4.xxx    op: 02804a00 
>       w:     t1.w     c0.y      0.0    op: 00040209
> 30: xyz:  t4  t3  t4  bias-> t3.xyz               (038c40c4)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t4.yyy   t3.xyz      0.0    op: 08050202 
>       w:      1.0      1.0    -t4.x    op: 00098891
> 31: xyz:  t1  t3  t0  bias-> t1.xyz               (038400c1)
>       w:  t2  t0  t0  bias-> t2.w                 (00880002)
>     xyz:   t1.xyz      1.0  -t3.xyz    op: 00090a80 
>       w:     t2.w     t1.x     t1.x    op: 04800009
> 32: xyz:  c2  t1  t0  bias-> t3.xyz               (038c0062)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   c2.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t1.w     c2.x     c2.x    op: 04000009
> 33: xyz: t32  c0  c1  bias-> t4.x                 (00921800)
>       w:  t1  t2  t0  bias-> t1.w                 (00840081)
>     xyz:   t1.www   c0.xyz      0.0    op: 0005020c 
>       w:     t2.w     c1.y      0.0    op: 0004038a
> 34: xyz:  c3  t1  t3  bias-> t3.xyz               (038c3063)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c3.xyz   t1.yyy   t3.xyz    op: 00020300 
>       w:     c3.x     c3.x     c3.x    op: 00000000
> 35: xyz:  c4  t1  t3  bias-> t1.xyz               (03843064)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c4.xyz   t1.zzz   t3.xyz    op: 00020380 
>       w:     c4.x     c4.x     c4.x    op: 00000000
> 36: xyz:  t0  t0  t0  bias-> t3.x                 (008c0000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 37: xyz:  t1  c5  t0  bias-> t1.xyz               (03840941)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0   c5.xyz    op: 00010a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 38: xyz:  t0  t3  c1  bias->                      (000210c0)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.z     t3.x     c1.x    op: 00018182
> 39: xyz: t32  t0  t0  bias-> t0.z                 (02000000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t1.www      0.0  t32.xxx    op: 02804a0c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
>NODE 2: alu_offset: 40, tex_offset: 2, alu_end: 7, tex_end: 1  (code_addr: 004221e8)
>  TEX:
>    TEX t3, t0, texture[1]   (000088c0)
>    TEX t1, t1, texture[0]   (00008041)
> 40: xyz:  t2  t3  t0  bias-> t2.xyz               (038800c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 41: xyz:  t1  t4  t0  bias-> t3.xyz               (038c0101)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 42: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t0.zzz      0.0    op: 00050380 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 43: xyz:  t3  t6  t0  bias-> t1.xyz               (03840183)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t6.xxx      0.0    op: 00050280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 44: xyz:  t0  t6  t0  bias-> t0.xyz               (03800180)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   t6.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 45: xyz:  t0 c10  t2  bias-> t0.xyz               (03802a80)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz  c10.xxx   t2.xyz    op: 00020280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 46: xyz:  t1 c10  t0  bias-> t0.xyz               (03800a81)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz  c10.xxx   t0.xyz    op: 00020280 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 47: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0..4]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0000,     0.0000,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[2]
>  2: ADD TEMP[1].x, -IN[1].zzzz, -CONST[2].xxxx
>  3: RCP TEMP[2].x, CONST[2].yyyy
>  4: MUL_SAT TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
>  5: POW TEMP[1].x, TEMP[1].xxxx, CONST[0].xxxx
>  6: ADD TEMP[2].x, IMM[0].xxxx, -TEMP[1].xxxx
>  7: MAD TEMP[1].x, CONST[1].xxxx, TEMP[1].xxxx, TEMP[2].xxxx
>  8: MUL TEMP[1].x, TEMP[1].xxxx, CONST[4].xxxx
>  9: MUL TEMP[1].x, TEMP[0].wwww, TEMP[1].xxxx
> 10: MOV TEMP[0].w, TEMP[1].xxxx
> 11: MUL TEMP[0].xyz, TEMP[0].xyzz, CONST[3].xxxx
> 12: MOV OUT[0], TEMP[0]
> 13: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: POW temp[1].x, temp[1].xxxx, const[0].xxxx;
>  6: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  7: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
>  8: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
>  9: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 10: MOV temp[0].w, temp[1].xxxx;
> 11: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 12: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].zzzz, -const[2].xxxx;
>  3: RCP temp[2].x, const[2].yyyy;
>  4: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  5: LG2 temp[3].w, temp[1].xxxx;
>  6: MUL temp[3].w, temp[3].wwww, const[0].xxxx;
>  7: EX2 temp[1].x, temp[3].wwww;
>  8: ADD temp[2].x, temp[0].1111, -temp[1].xxxx;
>  9: MAD temp[1].x, const[1].xxxx, temp[1].xxxx, temp[2].xxxx;
> 10: MUL temp[1].x, temp[1].xxxx, const[4].xxxx;
> 11: MUL temp[1].x, temp[0].wwww, temp[1].xxxx;
> 12: MOV temp[0].w, temp[1].xxxx;
> 13: MUL temp[0].xyz, temp[0].xyzz, const[3].xxxx;
> 14: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[2].x, const[2].y___;
>  4: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  5: LG2 temp[3].w, temp[1].___x;
>  6: MUL temp[3].w, temp[3].___w, const[0].___x;
>  7: EX2 temp[1].x, temp[3].w___;
>  8: ADD temp[2].x, temp[0].1___, -temp[1].x___;
>  9: MAD temp[1].x, const[1].x___, temp[1].x___, temp[2].x___;
> 10: MUL temp[1].x, temp[1].x___, const[4].x___;
> 11: MUL temp[1].x, temp[0].w___, temp[1].x___;
> 12: MOV temp[0].w, temp[1].___x;
> 13: MUL temp[0].xyz, temp[0].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[2];
>  2: ADD temp[1].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[2].x, const[2].y___;
>  4: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  5: LG2 temp[3].w, temp[1].___x;
>  6: MUL temp[3].w, temp[3].___w, const[0].___x;
>  7: EX2 temp[1].x, temp[3].w___;
>  8: ADD temp[2].x, temp[0].1___, -temp[1].x___;
>  9: MAD temp[1].x, const[1].x___, temp[1].x___, temp[2].x___;
> 10: MUL temp[1].x, temp[1].x___, const[4].x___;
> 11: MUL temp[1].x, temp[0].w___, temp[1].x___;
> 12: MOV temp[0].w, temp[1].___x;
> 13: MUL temp[0].xyz, temp[0].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, temp[0].1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[5].w___, temp[14].x___;
> 12: MOV temp[16].w, temp[15].___x;
> 13: MUL temp[16].xyz, temp[5].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[16];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[5].w___, temp[14].x___;
> 12: MOV temp[16].w, temp[15].___x;
> 13: MUL temp[16].xyz, temp[5].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[16];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[5].w___, temp[14].x___;
> 12: MOV temp[16].w, temp[15].___x;
> 13: MUL temp[16].xyz, temp[5].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[16];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: MUL temp[5], temp[4], input[2];
>  2: ADD temp[6].x, -input[1].z___, -const[2].x___;
>  3: RCP temp[7].x, const[2].y___;
>  4: MUL_SAT temp[8].x, temp[6].x___, temp[7].x___;
>  5: LG2 temp[9].w, temp[8].___x;
>  6: MUL temp[10].w, temp[9].___w, const[0].___x;
>  7: EX2 temp[11].x, temp[10].w___;
>  8: ADD temp[12].x, none.1___, -temp[11].x___;
>  9: MAD temp[13].x, const[1].x___, temp[11].x___, temp[12].x___;
> 10: MUL temp[14].x, temp[13].x___, const[4].x___;
> 11: MUL temp[15].x, temp[5].w___, temp[14].x___;
> 12: MOV temp[16].w, temp[15].___x;
> 13: MUL temp[16].xyz, temp[5].xyz_, const[3].xxx_;
> 14: MOV output[0], temp[16];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[4], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2]
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>  3: src0.xyz = const[2]
>     REPL_ALPHA temp[7].x
>     RCP, src0.y
>  4: src0.xyz = temp[6], src1.xyz = temp[7]
>     MAD_SAT temp[8].x, src0.x__, src1.x__, src0.000
>  5: src0.xyz = temp[8]
>     LG2 temp[9].w, src0.x
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src0.w = temp[5]
>     MAD temp[15].x, src0.w__, src0.x__, src0.000
> 12: src0.xyz = temp[15]
>     MAD temp[16].w, src0.x, src0.1, src0.0
> 13: src0.xyz = temp[5], src1.xyz = const[3]
>     MAD temp[16].xyz, src0.xyz, src1.xxx, src0.000
> 14: src0.xyz = temp[16], src0.w = temp[16]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[4], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[7].w, src1.y
>  3: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[5], src0.w = temp[7], src1.xyz = const[3], src2.xyz = temp[6]
>     MAD temp[16].xyz, src0.xyz, src1.xxx, src0.000
>     MAD_SAT temp[8].w, src2.x, src0.w, src0.0
>  5: src0.xyz = temp[8], src0.w = temp[8]
>     LG2 temp[9].w, src0.w
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src0.w = temp[5]
>     MAD temp[15].x, src0.w__, src0.x__, src0.000
> 12: src0.xyz = temp[15]
>     MAD temp[16].w, src0.x, src0.1, src0.0
> 13: src0.xyz = temp[16], src0.w = temp[16]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[4], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[6].x, -src0.z__, src0.111, -src1.x__
>     RCP temp[7].w, src1.y
>  3: src0.xyz = temp[4], src0.w = temp[4], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[5].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[5].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[5], src0.w = temp[7], src1.xyz = const[3], src2.xyz = temp[6]
>     MAD temp[16].xyz, src0.xyz, src1.xxx, src0.000
>     MAD_SAT temp[8].w, src2.x, src0.w, src0.0
>  5: src0.w = temp[8]
>     LG2 temp[9].w, src0.w
>  6: src0.xyz = const[0], src0.w = temp[9]
>     MAD temp[10].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[10]
>     REPL_ALPHA temp[11].x
>     EX2, src0.w
>  8: src0.xyz = temp[11]
>     MAD temp[12].x, src0.1__, src0.111, -src0.x__
>  9: src0.xyz = const[1], src1.xyz = temp[11], src2.xyz = temp[12]
>     MAD temp[13].x, src0.x__, src1.x__, src2.x__
> 10: src0.xyz = temp[13], src1.xyz = const[4]
>     MAD temp[14].x, src0.x__, src1.x__, src0.000
> 11: src0.xyz = temp[14], src0.w = temp[5]
>     MAD temp[15].x, src0.w__, src0.x__, src0.000
> 12: src0.xyz = temp[15]
>     MAD temp[16].w, src0.x, src0.1, src0.0
> 13: src0.xyz = temp[16], src0.w = temp[16]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[1].x, -src0.z__, src0.1__, -src1.x__
>     RCP temp[1].w, src1.y
>  3: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  4: src0.xyz = temp[0], src0.w = temp[1], src1.xyz = const[3], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
>     MAD_SAT temp[1].w, src2.x, src0.w, src0.0
>  5: src0.w = temp[1]
>     LG2 temp[1].w, src0.w
>  6: src0.xyz = const[0], src0.w = temp[1]
>     MAD temp[1].w, src0.w, src0.x, src0.0
>  7: src0.w = temp[1]
>     REPL_ALPHA temp[1].x
>     EX2, src0.w
>  8: src0.xyz = temp[1]
>     MAD temp[1].y, src0._1_, src0._1_, -src0._x_
>  9: src0.xyz = const[1], src1.xyz = temp[1], src2.xyz = temp[1]
>     MAD temp[1].x, src0.x__, src1.x__, src2.y__
> 10: src0.xyz = temp[1], src1.xyz = const[4]
>     MAD temp[1].x, src0.x__, src1.x__, src0.0__
> 11: src0.xyz = temp[1], src0.w = temp[0]
>     MAD temp[1].x, src0.w__, src0.x__, src0.0__
> 12: src0.xyz = temp[1]
>     MAD temp[0].w, src0.x, src0.1, src0.0
> 13: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=38*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 11, tex_end: 0  (code_addr: 004002c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c2  t0  bias-> t1.x                 (00840881)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:  -t1.zzz      1.0  -c2.xyz    op: 00090aa3 
>       w:     c2.y     t1.x     t1.x    op: 05000004
>  1: xyz:  t0  t2  t0  bias-> t0.xyz               (03800080)
>       w:  t0  t2  t0  bias-> t0.w                 (00800080)
>     xyz:   t0.xyz   t2.xyz      0.0    op: 00050200 
>       w:     t0.w     t2.w      0.0    op: 00040509
>  2: xyz:  t0  c3  t1  bias-> t0.xyz               (038018c0)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t0.xyz   c3.xxx      0.0    op: 00050280 
>       w:     t1.x     t1.w      0.0    op: 40040486
>  3: xyz:  t0  t0  t0  bias->                      (00000000)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t1.w     t0.x     t0.x    op: 04800009
>  4: xyz:  c0  t0  t0  bias->                      (00000020)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   c0.xxx   c0.xxx   c0.xxx    op: 00004081 
>       w:     t1.w     c0.x      0.0    op: 00040009
>  5: xyz:  t0  t0  t0  bias-> t1.x                 (00840000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
>  6: xyz:  t1  t0  t0  bias-> t1.y                 (01040001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t1.xxx    op: 00084a95 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  7: xyz:  c1  t1  t1  bias-> t1.x                 (00841061)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c1.xyz   t1.xyz   t1.yyy    op: 00028200 
>       w:     c1.x     c1.x     c1.x    op: 00000000
>  8: xyz:  t1  c4  t0  bias-> t1.x                 (00840901)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  9: xyz:  t1  t0  t0  bias-> t1.x                 (00840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.www   t1.xyz      0.0    op: 0005000c 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 10: xyz:  t1  t0  t0  bias->                      (00000001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xxx   t1.xxx   t1.xxx    op: 00004081 
>       w:     t1.x      1.0      0.0    op: 00040880
> 11: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[14], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL CONST[0..10]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    1.0000,     2.0000,     0.0000,    -1.0000}
>  0: MOV TEMP[0].w, IMM[0].xxxx
>  1: ADD TEMP[1].x, -IN[1].zzzz, -CONST[8].xxxx
>  2: RCP TEMP[2].x, CONST[8].yyyy
>  3: MUL_SAT TEMP[1].x, TEMP[1].xxxx, TEMP[2].xxxx
>  4: POW TEMP[1].x, TEMP[1].xxxx, CONST[6].xxxx
>  5: ADD TEMP[2].x, IMM[0].xxxx, -TEMP[1].xxxx
>  6: MAD TEMP[1].x, CONST[7].xxxx, TEMP[1].xxxx, TEMP[2].xxxx
>  7: MUL TEMP[1].x, TEMP[1].xxxx, CONST[10].xxxx
>  8: DP3 TEMP[2].x, IN[2].xyzz, IN[2].xyzz
>  9: RSQ TEMP[2].x, TEMP[2].xxxx
> 10: MUL TEMP[2].xyz, IN[2].xyzz, TEMP[2].xxxx
> 11: DP3 TEMP[3].x, IN[1].xyzz, IN[1].xyzz
> 12: RSQ TEMP[3].x, TEMP[3].xxxx
> 13: MUL TEMP[3].xyz, IN[1].xyzz, TEMP[3].xxxx
> 14: DP3 TEMP[4].x, TEMP[2].xyzz, TEMP[3].xyzz
> 15: MUL TEMP[5].xyz, TEMP[4].xxxx, TEMP[2].xyzz
> 16: MUL TEMP[5].xyz, IMM[0].yyyy, TEMP[5].xyzz
> 17: ADD TEMP[4].xyz, TEMP[3].xyzz, -TEMP[5].xyzz
> 18: MUL TEMP[5], CONST[2], TEMP[4].xxxx
> 19: MAD TEMP[5], CONST[3], TEMP[4].yyyy, TEMP[5]
> 20: MAD TEMP[4], CONST[4], TEMP[4].zzzz, TEMP[5]
> 21: ADD TEMP[4].xyz, TEMP[4], CONST[5]
> 22: TEX TEMP[4], TEMP[4].xyzz, SAMP[1], CUBE
> 23: TEX TEMP[5].w, IN[0].xyyy, SAMP[0], 2D
> 24: MUL TEMP[4].xyz, TEMP[4], TEMP[5].wwww
> 25: ADD TEMP[5].x, IMM[0].xxxx, -CONST[1].xxxx
> 26: DP3 TEMP[3].x, -TEMP[3].xyzz, TEMP[2].xyzz
> 27: MAX TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 28: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[3].xxxx
> 29: POW TEMP[3].x, TEMP[3].xxxx, CONST[1].yyyy
> 30: MAD TEMP[3].x, TEMP[5].xxxx, TEMP[3].xxxx, CONST[1].xxxx
> 31: MAX TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 32: MUL TEMP[3].xyz, TEMP[4].xyzz, TEMP[3].xxxx
> 33: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[1].xxxx
> 34: MUL TEMP[0].xyz, TEMP[3].xyzz, CONST[9].xxxx
> 35: DP3 TEMP[2].x, TEMP[2].xyzz, IMM[0].zzww
> 36: ABS TEMP[2].x, TEMP[2].xxxx
> 37: ADD TEMP[2].x, IMM[0].xxxx, -TEMP[2].xxxx
> 38: POW TEMP[2].x, TEMP[2].xxxx, CONST[0].yyyy
> 39: MUL TEMP[2].x, TEMP[2].xxxx, CONST[0].xxxx
> 40: MUL TEMP[2].xyz, TEMP[4].xyzz, TEMP[2].xxxx
> 41: MUL TEMP[1].xyz, TEMP[2].xyzz, TEMP[1].xxxx
> 42: MAD TEMP[0].xyz, TEMP[1].xyzz, CONST[9].xxxx, TEMP[0].xyzz
> 43: MOV OUT[0], TEMP[0]
> 44: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[5].w, input[0].xyyy, 2D[0];
> 24: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 25: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 26: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 28: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 29: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 30: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 31: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 32: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 34: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 35: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 36: ABS temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 38: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 39: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 40: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 41: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 42: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 43: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[5].w, input[0].xyyy, 2D[0];
> 24: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 25: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 26: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 28: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 29: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 30: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 31: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 32: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 34: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 35: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 36: ABS temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 38: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 39: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 40: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 41: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 42: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 43: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[5].w, input[0].xyyy, 2D[0];
> 24: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 25: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 26: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 28: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 29: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 30: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 31: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 32: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 34: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 35: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 36: ABS temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 38: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 39: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 40: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 41: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 42: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 43: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[5].w, input[0].xyyy, 2D[0];
> 24: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 25: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 26: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 28: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 29: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 30: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 31: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 32: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 34: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 35: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 36: ABS temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 38: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 39: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 40: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 41: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 42: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 43: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[5].w, input[0].xyyy, 2D[0];
> 24: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 25: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 26: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 27: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 28: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 29: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 30: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 31: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 32: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 33: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 34: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 35: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 36: ABS temp[2].x, temp[2].xxxx;
> 37: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 38: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 39: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 40: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 41: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 42: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 43: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: POW temp[1].x, temp[1].xxxx, const[6].xxxx;
>  5: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  6: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  7: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
>  8: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
>  9: RSQ temp[2].x, temp[2].xxxx;
> 10: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 11: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 12: RSQ temp[3].x, temp[3].xxxx;
> 13: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 14: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 15: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 16: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 17: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 18: MUL temp[5], const[2], temp[4].xxxx;
> 19: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 20: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 21: ADD temp[4].xyz, temp[4], const[5];
> 22: TEX temp[4], temp[4].xyzz, CUBE[1];
> 23: TEX temp[6], input[0].xyyy, 2D[0];
> 24: MOV temp[5].w, temp[6];
> 25: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 26: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 27: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 28: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 29: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 30: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 31: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 32: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 33: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 34: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 35: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 36: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 37: ABS temp[2].x, temp[2].xxxx;
> 38: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 39: POW temp[2].x, temp[2].xxxx, const[0].yyyy;
> 40: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 41: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 42: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 43: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 44: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].xxxx;
>  1: ADD temp[1].x, -input[1].zzzz, -const[8].xxxx;
>  2: RCP temp[2].x, const[8].yyyy;
>  3: MUL_SAT temp[1].x, temp[1].xxxx, temp[2].xxxx;
>  4: LG2 temp[7].w, temp[1].xxxx;
>  5: MUL temp[7].w, temp[7].wwww, const[6].xxxx;
>  6: EX2 temp[1].x, temp[7].wwww;
>  7: ADD temp[2].x, const[11].xxxx, -temp[1].xxxx;
>  8: MAD temp[1].x, const[7].xxxx, temp[1].xxxx, temp[2].xxxx;
>  9: MUL temp[1].x, temp[1].xxxx, const[10].xxxx;
> 10: DP3 temp[2].x, input[2].xyzz, input[2].xyzz;
> 11: RSQ temp[2].x, |temp[2].xxxx|;
> 12: MUL temp[2].xyz, input[2].xyzz, temp[2].xxxx;
> 13: DP3 temp[3].x, input[1].xyzz, input[1].xyzz;
> 14: RSQ temp[3].x, |temp[3].xxxx|;
> 15: MUL temp[3].xyz, input[1].xyzz, temp[3].xxxx;
> 16: DP3 temp[4].x, temp[2].xyzz, temp[3].xyzz;
> 17: MUL temp[5].xyz, temp[4].xxxx, temp[2].xyzz;
> 18: MUL temp[5].xyz, const[11].yyyy, temp[5].xyzz;
> 19: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 20: MUL temp[5], const[2], temp[4].xxxx;
> 21: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 22: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 23: ADD temp[4].xyz, temp[4], const[5];
> 24: TEX temp[4], temp[4].xyzz, CUBE[1];
> 25: TEX temp[6], input[0].xyyy, 2D[0];
> 26: MOV temp[5].w, temp[6];
> 27: MUL temp[4].xyz, temp[4], temp[5].wwww;
> 28: ADD temp[5].x, const[11].xxxx, -const[1].xxxx;
> 29: DP3 temp[3].x, -temp[3].xyzz, temp[2].xyzz;
> 30: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 31: ADD temp[3].x, const[11].xxxx, -temp[3].xxxx;
> 32: LG2 temp[8].w, temp[3].xxxx;
> 33: MUL temp[8].w, temp[8].wwww, const[1].yyyy;
> 34: EX2 temp[3].x, temp[8].wwww;
> 35: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 36: MAX temp[3].x, temp[3].xxxx, const[11].zzzz;
> 37: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 38: MUL temp[3].xyz, temp[3].xyzz, temp[1].xxxx;
> 39: MUL temp[0].xyz, temp[3].xyzz, const[9].xxxx;
> 40: DP3 temp[2].x, temp[2].xyzz, const[11].zzww;
> 41: MOV temp[2].x, |temp[2].xxxx|;
> 42: ADD temp[2].x, const[11].xxxx, -temp[2].xxxx;
> 43: LG2 temp[9].w, temp[2].xxxx;
> 44: MUL temp[9].w, temp[9].wwww, const[0].yyyy;
> 45: EX2 temp[2].x, temp[9].wwww;
> 46: MUL temp[2].x, temp[2].xxxx, const[0].xxxx;
> 47: MUL temp[2].xyz, temp[4].xyzz, temp[2].xxxx;
> 48: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 49: MAD temp[0].xyz, temp[1].xyzz, const[9].xxxx, temp[0].xyzz;
> 50: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].___x;
>  1: ADD temp[1].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[2].x, const[8].y___;
>  3: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  4: LG2 temp[7].w, temp[1].___x;
>  5: MUL temp[7].w, temp[7].___w, const[6].___x;
>  6: EX2 temp[1].x, temp[7].w___;
>  7: ADD temp[2].x, const[11].x___, -temp[1].x___;
>  8: MAD temp[1].x, const[7].x___, temp[1].x___, temp[2].x___;
>  9: MUL temp[1].x, temp[1].x___, const[10].x___;
> 10: DP3 temp[2].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[2].x, |temp[2].x___|;
> 12: MUL temp[2].xyz, input[2].xyz_, temp[2].xxx_;
> 13: DP3 temp[3].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[3].x, |temp[3].x___|;
> 15: MUL temp[3].xyz, input[1].xyz_, temp[3].xxx_;
> 16: DP3 temp[4].x, temp[2].xyz_, temp[3].xyz_;
> 17: MUL temp[5].xyz, temp[4].xxx_, temp[2].xyz_;
> 18: MUL temp[5].xyz, const[11].yyy_, temp[5].xyz_;
> 19: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 20: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 21: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 22: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 23: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 24: TEX temp[4].xyz, temp[4].xyz_, CUBE[1];
> 25: TEX temp[6].w, input[0].xy__, 2D[0];
> 26: MOV temp[5].w, temp[6].___w;
> 27: MUL temp[4].xyz, temp[4].xyz_, temp[5].www_;
> 28: ADD temp[5].x, const[11].x___, -const[1].x___;
> 29: DP3 temp[3].x, -temp[3].xyz_, temp[2].xyz_;
> 30: MAX temp[3].x, temp[3].x___, const[11].z___;
> 31: ADD temp[3].x, const[11].x___, -temp[3].x___;
> 32: LG2 temp[8].w, temp[3].___x;
> 33: MUL temp[8].w, temp[8].___w, const[1].___y;
> 34: EX2 temp[3].x, temp[8].w___;
> 35: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 36: MAX temp[3].x, temp[3].x___, const[11].z___;
> 37: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 38: MUL temp[3].xyz, temp[3].xyz_, temp[1].xxx_;
> 39: MUL temp[0].xyz, temp[3].xyz_, const[9].xxx_;
> 40: DP3 temp[2].x, temp[2].xyz_, const[11].zzw_;
> 41: MOV temp[2].x, |temp[2].x___|;
> 42: ADD temp[2].x, const[11].x___, -temp[2].x___;
> 43: LG2 temp[9].w, temp[2].___x;
> 44: MUL temp[9].w, temp[9].___w, const[0].___y;
> 45: EX2 temp[2].x, temp[9].w___;
> 46: MUL temp[2].x, temp[2].x___, const[0].x___;
> 47: MUL temp[2].xyz, temp[4].xyz_, temp[2].xxx_;
> 48: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 49: MAD temp[0].xyz, temp[1].xyz_, const[9].xxx_, temp[0].xyz_;
> 50: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV temp[0].w, const[11].___x;
>  1: ADD temp[1].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[2].x, const[8].y___;
>  3: MUL_SAT temp[1].x, temp[1].x___, temp[2].x___;
>  4: LG2 temp[7].w, temp[1].___x;
>  5: MUL temp[7].w, temp[7].___w, const[6].___x;
>  6: EX2 temp[1].x, temp[7].w___;
>  7: ADD temp[2].x, const[11].x___, -temp[1].x___;
>  8: MAD temp[1].x, const[7].x___, temp[1].x___, temp[2].x___;
>  9: MUL temp[1].x, temp[1].x___, const[10].x___;
> 10: DP3 temp[2].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[2].x, |temp[2].x___|;
> 12: MUL temp[2].xyz, input[2].xyz_, temp[2].xxx_;
> 13: DP3 temp[3].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[3].x, |temp[3].x___|;
> 15: MUL temp[3].xyz, input[1].xyz_, temp[3].xxx_;
> 16: DP3 temp[4].x, temp[2].xyz_, temp[3].xyz_;
> 17: MUL temp[5].xyz, temp[4].xxx_, temp[2].xyz_;
> 18: MUL temp[5].xyz, const[11].yyy_, temp[5].xyz_;
> 19: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 20: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 21: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 22: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 23: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 24: TEX temp[4].xyz, temp[4].xyz_, CUBE[1];
> 25: TEX temp[6].w, input[0].xy__, 2D[0];
> 26: MOV temp[5].w, temp[6].___w;
> 27: MUL temp[4].xyz, temp[4].xyz_, temp[5].www_;
> 28: ADD temp[5].x, const[11].x___, -const[1].x___;
> 29: DP3 temp[3].x, -temp[3].xyz_, temp[2].xyz_;
> 30: MAX temp[3].x, temp[3].x___, const[11].z___;
> 31: ADD temp[3].x, const[11].x___, -temp[3].x___;
> 32: LG2 temp[8].w, temp[3].___x;
> 33: MUL temp[8].w, temp[8].___w, const[1].___y;
> 34: EX2 temp[3].x, temp[8].w___;
> 35: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 36: MAX temp[3].x, temp[3].x___, const[11].z___;
> 37: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 38: MUL temp[3].xyz, temp[3].xyz_, temp[1].xxx_;
> 39: MUL temp[0].xyz, temp[3].xyz_, const[9].xxx_;
> 40: DP3 temp[2].x, temp[2].xyz_, const[11].zzw_;
> 41: MOV temp[2].x, |temp[2].x___|;
> 42: ADD temp[2].x, const[11].x___, -temp[2].x___;
> 43: LG2 temp[9].w, temp[2].___x;
> 44: MUL temp[9].w, temp[9].___w, const[0].___y;
> 45: EX2 temp[2].x, temp[9].w___;
> 46: MUL temp[2].x, temp[2].x___, const[0].x___;
> 47: MUL temp[2].xyz, temp[4].xyz_, temp[2].xxx_;
> 48: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 49: MAD temp[0].xyz, temp[1].xyz_, const[9].xxx_, temp[0].xyz_;
> 50: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV temp[10].w, const[11].___x;
>  1: ADD temp[11].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[12].x, const[8].y___;
>  3: MUL_SAT temp[13].x, temp[11].x___, temp[12].x___;
>  4: LG2 temp[14].w, temp[13].___x;
>  5: MUL temp[15].w, temp[14].___w, const[6].___x;
>  6: EX2 temp[16].x, temp[15].w___;
>  7: ADD temp[17].x, const[11].x___, -temp[16].x___;
>  8: MAD temp[18].x, const[7].x___, temp[16].x___, temp[17].x___;
>  9: MUL temp[19].x, temp[18].x___, const[10].x___;
> 10: DP3 temp[20].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[21].x, |temp[20].x___|;
> 12: MUL temp[22].xyz, input[2].xyz_, temp[21].xxx_;
> 13: DP3 temp[23].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[24].x, |temp[23].x___|;
> 15: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 16: DP3 temp[26].x, temp[22].xyz_, temp[25].xyz_;
> 17: MUL temp[27].xyz, temp[26].xxx_, temp[22].xyz_;
> 18: MUL temp[28].xyz, const[11].yyy_, temp[27].xyz_;
> 19: ADD temp[29].xyz, temp[25].xyz_, -temp[28].xyz_;
> 20: MUL temp[30].xyz, const[2].xyz_, temp[29].xxx_;
> 21: MAD temp[31].xyz, const[3].xyz_, temp[29].yyy_, temp[30].xyz_;
> 22: MAD temp[32].xyz, const[4].xyz_, temp[29].zzz_, temp[31].xyz_;
> 23: ADD temp[33].xyz, temp[32].xyz_, const[5].xyz_;
> 24: TEX temp[34].xyz, temp[33].xyz_, CUBE[1];
> 25: TEX temp[35].w, input[0].xy__, 2D[0];
> 26: MOV temp[36].w, temp[35].___w;
> 27: MUL temp[37].xyz, temp[34].xyz_, temp[36].www_;
> 28: ADD temp[38].x, const[11].x___, -const[1].x___;
> 29: DP3 temp[39].x, -temp[25].xyz_, temp[22].xyz_;
> 30: MAX temp[40].x, temp[39].x___, const[11].z___;
> 31: ADD temp[41].x, const[11].x___, -temp[40].x___;
> 32: LG2 temp[42].w, temp[41].___x;
> 33: MUL temp[43].w, temp[42].___w, const[1].___y;
> 34: EX2 temp[44].x, temp[43].w___;
> 35: MAD temp[45].x, temp[38].x___, temp[44].x___, const[1].x___;
> 36: MAX temp[46].x, temp[45].x___, const[11].z___;
> 37: MUL temp[47].xyz, temp[37].xyz_, temp[46].xxx_;
> 38: MUL temp[48].xyz, temp[47].xyz_, temp[19].xxx_;
> 39: MUL temp[49].xyz, temp[48].xyz_, const[9].xxx_;
> 40: DP3 temp[50].x, temp[22].xyz_, const[11].zzw_;
> 41: MOV temp[51].x, |temp[50].x___|;
> 42: ADD temp[52].x, const[11].x___, -temp[51].x___;
> 43: LG2 temp[53].w, temp[52].___x;
> 44: MUL temp[54].w, temp[53].___w, const[0].___y;
> 45: EX2 temp[55].x, temp[54].w___;
> 46: MUL temp[56].x, temp[55].x___, const[0].x___;
> 47: MUL temp[57].xyz, temp[37].xyz_, temp[56].xxx_;
> 48: MUL temp[58].xyz, temp[57].xyz_, temp[19].xxx_;
> 49: MAD temp[10].xyz, temp[58].xyz_, const[9].xxx_, temp[49].xyz_;
> 50: MOV output[0], temp[10];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV temp[10].w, none.___1;
>  1: ADD temp[11].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[12].x, const[8].y___;
>  3: MUL_SAT temp[13].x, temp[11].x___, temp[12].x___;
>  4: LG2 temp[14].w, temp[13].___x;
>  5: MUL temp[15].w, temp[14].___w, const[6].___x;
>  6: EX2 temp[16].x, temp[15].w___;
>  7: ADD temp[17].x, none.1___, -temp[16].x___;
>  8: MAD temp[18].x, const[7].x___, temp[16].x___, temp[17].x___;
>  9: MUL temp[19].x, temp[18].x___, const[10].x___;
> 10: DP3 temp[20].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[21].x, |temp[20].x___|;
> 12: MUL temp[22].xyz, input[2].xyz_, temp[21].xxx_;
> 13: DP3 temp[23].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[24].x, |temp[23].x___|;
> 15: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 16: DP3 temp[26].x, temp[22].xyz_, temp[25].xyz_;
> 17: MUL temp[28].xyz * 2, temp[26].xxx_, temp[22].xyz_;
> 18: ADD temp[29].xyz, temp[25].xyz_, -temp[28].xyz_;
> 19: MUL temp[30].xyz, const[2].xyz_, temp[29].xxx_;
> 20: MAD temp[31].xyz, const[3].xyz_, temp[29].yyy_, temp[30].xyz_;
> 21: MAD temp[32].xyz, const[4].xyz_, temp[29].zzz_, temp[31].xyz_;
> 22: ADD temp[33].xyz, temp[32].xyz_, const[5].xyz_;
> 23: TEX temp[34].xyz, temp[33].xyz_, CUBE[1];
> 24: TEX temp[35].w, input[0].xy__, 2D[0];
> 25: MUL temp[37].xyz, temp[34].xyz_, temp[35].www_;
> 26: ADD temp[38].x, none.1___, -const[1].x___;
> 27: DP3 temp[39].x, -temp[25].xyz_, temp[22].xyz_;
> 28: MAX temp[40].x, temp[39].x___, none.0___;
> 29: ADD temp[41].x, none.1___, -temp[40].x___;
> 30: LG2 temp[42].w, temp[41].___x;
> 31: MUL temp[43].w, temp[42].___w, const[1].___y;
> 32: EX2 temp[44].x, temp[43].w___;
> 33: MAD temp[45].x, temp[38].x___, temp[44].x___, const[1].x___;
> 34: MAX temp[46].x, temp[45].x___, none.0___;
> 35: MUL temp[47].xyz, temp[37].xyz_, temp[46].xxx_;
> 36: MUL temp[48].xyz, temp[47].xyz_, temp[19].xxx_;
> 37: MUL temp[49].xyz, temp[48].xyz_, const[9].xxx_;
> 38: DP3 temp[50].x, temp[22].xyz_, none.00-1_;
> 39: ADD temp[52].x, none.1___, -|temp[50].x___|;
> 40: LG2 temp[53].w, temp[52].___x;
> 41: MUL temp[54].w, temp[53].___w, const[0].___y;
> 42: EX2 temp[55].x, temp[54].w___;
> 43: MUL temp[56].x, temp[55].x___, const[0].x___;
> 44: MUL temp[57].xyz, temp[37].xyz_, temp[56].xxx_;
> 45: MUL temp[58].xyz, temp[57].xyz_, temp[19].xxx_;
> 46: MAD temp[10].xyz, temp[58].xyz_, const[9].xxx_, temp[49].xyz_;
> 47: MOV output[0], temp[10];
>CONST[11] = {     1.0000     2.0000     0.0000    -1.0000 }
>CONST[12] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV temp[10].w, none.___1;
>  1: ADD temp[11].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[12].x, const[8].y___;
>  3: MUL_SAT temp[13].x, temp[11].x___, temp[12].x___;
>  4: LG2 temp[14].w, temp[13].___x;
>  5: MUL temp[15].w, temp[14].___w, const[6].___x;
>  6: EX2 temp[16].x, temp[15].w___;
>  7: ADD temp[17].x, none.1___, -temp[16].x___;
>  8: MAD temp[18].x, const[7].x___, temp[16].x___, temp[17].x___;
>  9: MUL temp[19].x, temp[18].x___, const[10].x___;
> 10: DP3 temp[20].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[21].x, |temp[20].x___|;
> 12: MUL temp[22].xyz, input[2].xyz_, temp[21].xxx_;
> 13: DP3 temp[23].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[24].x, |temp[23].x___|;
> 15: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 16: DP3 temp[26].x, temp[22].xyz_, temp[25].xyz_;
> 17: MUL temp[28].xyz * 2, temp[26].xxx_, temp[22].xyz_;
> 18: ADD temp[29].xyz, temp[25].xyz_, -temp[28].xyz_;
> 19: MUL temp[30].xyz, const[2].xyz_, temp[29].xxx_;
> 20: MAD temp[31].xyz, const[3].xyz_, temp[29].yyy_, temp[30].xyz_;
> 21: MAD temp[32].xyz, const[4].xyz_, temp[29].zzz_, temp[31].xyz_;
> 22: ADD temp[33].xyz, temp[32].xyz_, const[5].xyz_;
> 23: TEX temp[34].xyz, temp[33].xyz_, CUBE[1];
> 24: TEX temp[35].w, input[0].xy__, 2D[0];
> 25: MUL temp[37].xyz, temp[34].xyz_, temp[35].www_;
> 26: ADD temp[38].x, none.1___, -const[1].x___;
> 27: DP3 temp[39].x, -temp[25].xyz_, temp[22].xyz_;
> 28: MAX temp[40].x, temp[39].x___, none.0___;
> 29: ADD temp[41].x, none.1___, -temp[40].x___;
> 30: LG2 temp[42].w, temp[41].___x;
> 31: MUL temp[43].w, temp[42].___w, const[1].___y;
> 32: EX2 temp[44].x, temp[43].w___;
> 33: MAD temp[45].x, temp[38].x___, temp[44].x___, const[1].x___;
> 34: MAX temp[46].x, temp[45].x___, none.0___;
> 35: MUL temp[47].xyz, temp[37].xyz_, temp[46].xxx_;
> 36: MUL temp[48].xyz, temp[47].xyz_, temp[19].xxx_;
> 37: MUL temp[49].xyz, temp[48].xyz_, const[9].xxx_;
> 38: DP3 temp[50].x, temp[22].xyz_, const[12].xyz_;
> 39: ADD temp[52].x, none.1___, -|temp[50].x___|;
> 40: LG2 temp[53].w, temp[52].___x;
> 41: MUL temp[54].w, temp[53].___w, const[0].___y;
> 42: EX2 temp[55].x, temp[54].w___;
> 43: MUL temp[56].x, temp[55].x___, const[0].x___;
> 44: MUL temp[57].xyz, temp[37].xyz_, temp[56].xxx_;
> 45: MUL temp[58].xyz, temp[57].xyz_, temp[19].xxx_;
> 46: MAD temp[10].xyz, temp[58].xyz_, const[9].xxx_, temp[49].xyz_;
> 47: MOV output[0], temp[10];
>CONST[11] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV temp[10].w, none.___1;
>  1: ADD temp[11].x, -input[1].z___, -const[8].x___;
>  2: RCP temp[12].x, const[8].y___;
>  3: MUL_SAT temp[13].x, temp[11].x___, temp[12].x___;
>  4: LG2 temp[14].w, temp[13].___x;
>  5: MUL temp[15].w, temp[14].___w, const[6].___x;
>  6: EX2 temp[16].x, temp[15].w___;
>  7: ADD temp[17].x, none.1___, -temp[16].x___;
>  8: MAD temp[18].x, const[7].x___, temp[16].x___, temp[17].x___;
>  9: MUL temp[19].x, temp[18].x___, const[10].x___;
> 10: DP3 temp[20].x, input[2].xyz_, input[2].xyz_;
> 11: RSQ temp[21].x, |temp[20].x___|;
> 12: MUL temp[22].xyz, input[2].xyz_, temp[21].xxx_;
> 13: DP3 temp[23].x, input[1].xyz_, input[1].xyz_;
> 14: RSQ temp[24].x, |temp[23].x___|;
> 15: MUL temp[25].xyz, input[1].xyz_, temp[24].xxx_;
> 16: DP3 temp[26].x, temp[22].xyz_, temp[25].xyz_;
> 17: MUL temp[28].xyz * 2, temp[26].xxx_, temp[22].xyz_;
> 18: ADD temp[29].xyz, temp[25].xyz_, -temp[28].xyz_;
> 19: MUL temp[30].xyz, const[2].xyz_, temp[29].xxx_;
> 20: MAD temp[31].xyz, const[3].xyz_, temp[29].yyy_, temp[30].xyz_;
> 21: MAD temp[32].xyz, const[4].xyz_, temp[29].zzz_, temp[31].xyz_;
> 22: ADD temp[33].xyz, temp[32].xyz_, const[5].xyz_;
> 23: TEX temp[34].xyz, temp[33].xyz_, CUBE[1];
> 24: TEX temp[35].w, input[0].xy__, 2D[0];
> 25: MUL temp[37].xyz, temp[34].xyz_, temp[35].www_;
> 26: ADD temp[38].x, none.1___, -const[1].x___;
> 27: DP3 temp[39].x, -temp[25].xyz_, temp[22].xyz_;
> 28: MAX temp[40].x, temp[39].x___, none.0___;
> 29: ADD temp[41].x, none.1___, -temp[40].x___;
> 30: LG2 temp[42].w, temp[41].___x;
> 31: MUL temp[43].w, temp[42].___w, const[1].___y;
> 32: EX2 temp[44].x, temp[43].w___;
> 33: MAD temp[45].x, temp[38].x___, temp[44].x___, const[1].x___;
> 34: MAX temp[46].x, temp[45].x___, none.0___;
> 35: MUL temp[47].xyz, temp[37].xyz_, temp[46].xxx_;
> 36: MUL temp[48].xyz, temp[47].xyz_, temp[19].xxx_;
> 37: MUL temp[49].xyz, temp[48].xyz_, const[9].xxx_;
> 38: DP3 temp[50].x, temp[22].xyz_, const[11].xyz_;
> 39: ADD temp[52].x, none.1___, -|temp[50].x___|;
> 40: LG2 temp[53].w, temp[52].___x;
> 41: MUL temp[54].w, temp[53].___w, const[0].___y;
> 42: EX2 temp[55].x, temp[54].w___;
> 43: MUL temp[56].x, temp[55].x___, const[0].x___;
> 44: MUL temp[57].xyz, temp[37].xyz_, temp[56].xxx_;
> 45: MUL temp[58].xyz, temp[57].xyz_, temp[19].xxx_;
> 46: MAD temp[10].xyz, temp[58].xyz_, const[9].xxx_, temp[49].xyz_;
> 47: MOV output[0], temp[10];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD temp[10].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[1], src1.xyz = const[8]
>     MAD temp[11].x, -src0.z__, src0.111, -src1.x__
>  2: src0.xyz = const[8]
>     REPL_ALPHA temp[12].x
>     RCP, src0.y
>  3: src0.xyz = temp[11], src1.xyz = temp[12]
>     MAD_SAT temp[13].x, src0.x__, src1.x__, src0.000
>  4: src0.xyz = temp[13]
>     LG2 temp[14].w, src0.x
>  5: src0.xyz = const[6], src0.w = temp[14]
>     MAD temp[15].w, src0.w, src0.x, src0.0
>  6: src0.w = temp[15]
>     REPL_ALPHA temp[16].x
>     EX2, src0.w
>  7: src0.xyz = temp[16]
>     MAD temp[17].x, src0.1__, src0.111, -src0.x__
>  8: src0.xyz = const[7], src1.xyz = temp[16], src2.xyz = temp[17]
>     MAD temp[18].x, src0.x__, src1.x__, src2.x__
>  9: src0.xyz = temp[18], src1.xyz = const[10]
>     MAD temp[19].x, src0.x__, src1.x__, src0.000
> 10: src0.xyz = input[2]
>     DP3 temp[20].x, src0.xyz, src0.xyz
> 11: src0.xyz = temp[20]
>     REPL_ALPHA temp[21].x
>     RSQ, |src0.x|
> 12: src0.xyz = input[2], src1.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src0.000
> 13: src0.xyz = input[1]
>     DP3 temp[23].x, src0.xyz, src0.xyz
> 14: src0.xyz = temp[23]
>     REPL_ALPHA temp[24].x
>     RSQ, |src0.x|
> 15: src0.xyz = input[1], src1.xyz = temp[24]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[22], src1.xyz = temp[25]
>     DP3 temp[26].x, src0.xyz, src1.xyz
> 17: src0.xyz = temp[26], src1.xyz = temp[22]
>     MAD temp[28].xyz * 2, src0.xxx, src1.xyz, src0.000
> 18: src0.xyz = temp[25], src1.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, -src1.xyz
> 19: src0.xyz = const[2], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
> 20: src0.xyz = const[3], src1.xyz = temp[29], src2.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.yyy, src2.xyz
> 21: src0.xyz = const[4], src1.xyz = temp[29], src2.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.zzz, src2.xyz
> 22: src0.xyz = temp[32], src1.xyz = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.111, src1.xyz
> 23: TEX temp[34].xyz, temp[33].xyz_, CUBE[1];
> 24: TEX temp[35].w, input[0].xy__, 2D[0];
> 25: src0.xyz = temp[34], src0.w = temp[35]
>     MAD temp[37].xyz, src0.xyz, src0.www, src0.000
> 26: src0.xyz = const[1]
>     MAD temp[38].x, src0.1__, src0.111, -src0.x__
> 27: src0.xyz = temp[25], src1.xyz = temp[22]
>     DP3 temp[39].x, -src0.xyz, src1.xyz
> 28: src0.xyz = temp[39]
>     MAX temp[40].x, src0.x__, src0.0__
> 29: src0.xyz = temp[40]
>     MAD temp[41].x, src0.1__, src0.111, -src0.x__
> 30: src0.xyz = temp[41]
>     LG2 temp[42].w, src0.x
> 31: src0.xyz = const[1], src0.w = temp[42]
>     MAD temp[43].w, src0.w, src0.y, src0.0
> 32: src0.w = temp[43]
>     REPL_ALPHA temp[44].x
>     EX2, src0.w
> 33: src0.xyz = temp[38], src1.xyz = temp[44], src2.xyz = const[1]
>     MAD temp[45].x, src0.x__, src1.x__, src2.x__
> 34: src0.xyz = temp[45]
>     MAX temp[46].x, src0.x__, src0.0__
> 35: src0.xyz = temp[37], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
> 36: src0.xyz = temp[47], src1.xyz = temp[19]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 37: src0.xyz = temp[48], src1.xyz = const[9]
>     MAD temp[49].xyz, src0.xyz, src1.xxx, src0.000
> 38: src0.xyz = temp[22], src1.xyz = const[11]
>     DP3 temp[50].x, src0.xyz, src1.xyz
> 39: src0.xyz = temp[50]
>     MAD temp[52].x, src0.1__, |src0.111|, -|src0.x__|
> 40: src0.xyz = temp[52]
>     LG2 temp[53].w, src0.x
> 41: src0.xyz = const[0], src0.w = temp[53]
>     MAD temp[54].w, src0.w, src0.y, src0.0
> 42: src0.w = temp[54]
>     REPL_ALPHA temp[55].x
>     EX2, src0.w
> 43: src0.xyz = temp[55], src1.xyz = const[0]
>     MAD temp[56].x, src0.x__, src1.x__, src0.000
> 44: src0.xyz = temp[37], src1.xyz = temp[56]
>     MAD temp[57].xyz, src0.xyz, src1.xxx, src0.000
> 45: src0.xyz = temp[57], src1.xyz = temp[19]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 46: src0.xyz = temp[58], src1.xyz = const[9], src2.xyz = temp[49]
>     MAD temp[10].xyz, src0.xyz, src1.xxx, src2.xyz
> 47: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src1.xyz = const[1]
>     DP3 temp[23].x, src0.xyz, src0.xyz
>     MAD temp[38].w, src0.1, src0.1, -src1.x
>  1: src0.xyz = input[2], src1.xyz = const[8]
>     DP3 temp[20].x, src0.xyz, src0.xyz
>     RCP temp[12].w, src1.y
>  2: src0.xyz = input[1], src1.xyz = const[8]
>     MAD temp[11].x, -src0.z__, src0.111, -src1.x__
>     MAD temp[10].w, src0.1, src0.1, src0.0
>  3: src0.xyz = temp[11], src0.w = temp[12], src1.xyz = temp[12], src2.xyz = temp[23]
>     MAD_SAT temp[13].x, src0.x__, src0.w__, src0.000
>     RSQ temp[24].w, |src2.x|
>  4: src0.xyz = input[1], src0.w = temp[24], src1.xyz = temp[24], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[21].w, |src2.x|
>  5: src0.xyz = input[2], src0.w = temp[21], src1.xyz = temp[21], src2.xyz = temp[13]
>     MAD temp[22].xyz, src0.xyz, src0.www, src0.000
>     LG2 temp[14].w, src2.x
>  6: src0.xyz = temp[22], src0.w = temp[14], src1.xyz = const[11], src2.xyz = const[6]
>     DP3 temp[50].x, src0.xyz, src1.xyz
>     MAD temp[15].w, src0.w, src2.x, src0.0
>  7: src0.xyz = temp[25], src1.xyz = temp[22], src2.xyz = temp[50]
>     DP3 temp[39].x, -src0.xyz, src1.xyz
>     MAD temp[52].w, src0.1, |src0.1|, -|src2.x|
>  8: src0.xyz = temp[22], src0.w = temp[52], src1.xyz = temp[25]
>     DP3 temp[26].x, src0.xyz, src1.xyz
>     LG2 temp[53].w, src0.w
>  9: src0.xyz = temp[26], src0.w = temp[53], src1.xyz = temp[22], src2.xyz = const[0]
>     MAD temp[28].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[54].w, src0.w, src2.y, src0.0
> 10: src0.xyz = temp[25], src0.w = temp[54], src1.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, -src1.xyz
>     EX2 temp[55].w, src0.w
> 11: src0.xyz = const[2], src0.w = temp[55], src1.xyz = temp[29], src2.xyz = const[0]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[56].w, src0.w, src2.x, src0.0
> 12: src0.xyz = temp[39], src0.w = temp[15]
>     MAX temp[40].x, src0.x__, src0.0__
>     EX2 temp[16].w, src0.w
> 13: src0.xyz = temp[16], src0.w = temp[16], src1.xyz = temp[40]
>     MAD temp[17].x, src0.1__, src0.111, -src0.w__
>     MAD temp[41].w, src0.1, src0.1, -src1.x
> 14: src0.xyz = const[3], src0.w = temp[41], src1.xyz = temp[29], src2.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.yyy, src2.xyz
>     LG2 temp[42].w, src0.w
> 15: src0.xyz = const[7], src0.w = temp[16], src1.xyz = temp[16], src2.xyz = temp[17]
>     MAD temp[18].w, src0.x, src0.w, src2.x
> 16: src0.xyz = temp[18], src0.w = temp[18], src1.xyz = const[10], src1.w = temp[42], src2.xyz = const[1]
>     MAD temp[19].x, src0.w__, src1.x__, src0.000
>     MAD temp[43].w, src1.w, src2.y, src0.0
> 17: src0.xyz = const[4], src1.xyz = temp[29], src2.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.zzz, src2.xyz
> 18: src0.w = temp[43]
>     REPL_ALPHA temp[44].x
>     EX2, src0.w
> 19: src0.xyz = temp[32], src1.xyz = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.111, src1.xyz
> 20: BEGIN_TEX;
> 21: TEX temp[35].w, input[0].xy__, 2D[0];
> 22: TEX temp[34].xyz, temp[33].xyz_, CUBE[1] SEM_WAIT SEM_ACQUIRE;
> 23: src0.xyz = temp[34], src0.w = temp[35], src1.xyz = temp[44], src1.w = temp[38], src2.xyz = const[1] SEM_WAIT
>     MAD temp[37].xyz, src0.xyz, src0.www, src0.000
>     MAD temp[45].w, src1.w, src1.x, src2.x
> 24: src0.xyz = temp[37], src0.w = temp[56], src1.xyz = temp[56], src1.w = temp[45]
>     MAD temp[57].xyz, src0.xyz, src0.www, src0.000
>     MAX temp[46].w, src1.w, src0.0
> 25: src0.xyz = temp[37], src0.w = temp[46], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src0.www, src0.000
> 26: src0.xyz = temp[57], src1.xyz = temp[19]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = temp[47], src1.xyz = temp[19]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[48], src1.xyz = const[9]
>     MAD temp[49].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[58], src1.xyz = const[9], src2.xyz = temp[49]
>     MAD temp[10].xyz, src0.xyz, src1.xxx, src2.xyz
> 30: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src1.xyz = const[1]
>     DP3 temp[23].x, src0.xyz, src0.xyz
>     MAD temp[38].w, src0.1, src0.1, -src1.x
>  1: src0.xyz = input[2], src1.xyz = const[8]
>     DP3 temp[20].x, src0.xyz, src0.xyz
>     RCP temp[12].w, src1.y
>  2: src0.xyz = input[1], src1.xyz = const[8]
>     MAD temp[11].x, -src0.z__, src0.111, -src1.x__
>     MAD temp[10].w, src0.1, src0.1, src0.0
>  3: src0.xyz = temp[11], src0.w = temp[12], src2.xyz = temp[23]
>     MAD_SAT temp[13].x, src0.x__, src0.w__, src0.000
>     RSQ temp[24].w, |src2.x|
>  4: src0.xyz = input[1], src0.w = temp[24], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[21].w, |src2.x|
>  5: src0.xyz = input[2], src0.w = temp[21], src2.xyz = temp[13]
>     MAD temp[22].xyz, src0.xyz, src0.www, src0.000
>     LG2 temp[14].w, src2.x
>  6: src0.xyz = temp[22], src0.w = temp[14], src1.xyz = const[11], src2.xyz = const[6]
>     DP3 temp[50].x, src0.xyz, src1.xyz
>     MAD temp[15].w, src0.w, src2.x, src0.0
>  7: src0.xyz = temp[25], src1.xyz = temp[22], src2.xyz = temp[50]
>     DP3 temp[39].x, -src0.xyz, src1.xyz
>     MAD temp[52].w, src0.1, |src0.1|, -|src2.x|
>  8: src0.xyz = temp[22], src0.w = temp[52], src1.xyz = temp[25]
>     DP3 temp[26].x, src0.xyz, src1.xyz
>     LG2 temp[53].w, src0.w
>  9: src0.xyz = temp[26], src0.w = temp[53], src1.xyz = temp[22], src2.xyz = const[0]
>     MAD temp[28].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[54].w, src0.w, src2.y, src0.0
> 10: src0.xyz = temp[25], src0.w = temp[54], src1.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, -src1.xyz
>     EX2 temp[55].w, src0.w
> 11: src0.xyz = const[2], src0.w = temp[55], src1.xyz = temp[29], src2.xyz = const[0]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[56].w, src0.w, src2.x, src0.0
> 12: src0.xyz = temp[39], src0.w = temp[15]
>     MAX temp[40].x, src0.x__, src0.0__
>     EX2 temp[16].w, src0.w
> 13: src0.w = temp[16], src1.xyz = temp[40]
>     MAD temp[17].x, src0.1__, src0.111, -src0.w__
>     MAD temp[41].w, src0.1, src0.1, -src1.x
> 14: src0.xyz = const[3], src0.w = temp[41], src1.xyz = temp[29], src2.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.yyy, src2.xyz
>     LG2 temp[42].w, src0.w
> 15: src0.xyz = const[7], src0.w = temp[16], src2.xyz = temp[17]
>     MAD temp[18].w, src0.x, src0.w, src2.x
> 16: src0.w = temp[18], src1.xyz = const[10], src1.w = temp[42], src2.xyz = const[1]
>     MAD temp[19].x, src0.w__, src1.x__, src0.000
>     MAD temp[43].w, src1.w, src2.y, src0.0
> 17: src0.xyz = const[4], src1.xyz = temp[29], src2.xyz = temp[31]
>     MAD temp[32].xyz, src0.xyz, src1.zzz, src2.xyz
> 18: src0.w = temp[43]
>     REPL_ALPHA temp[44].x
>     EX2, src0.w
> 19: src0.xyz = temp[32], src1.xyz = const[5]
>     MAD temp[33].xyz, src0.xyz, src0.111, src1.xyz
> 20: BEGIN_TEX;
> 21: TEX temp[35].w, input[0].xy__, 2D[0];
> 22: TEX temp[34].xyz, temp[33].xyz_, CUBE[1] SEM_WAIT SEM_ACQUIRE;
> 23: src0.xyz = temp[34], src0.w = temp[35], src1.xyz = temp[44], src1.w = temp[38], src2.xyz = const[1] SEM_WAIT
>     MAD temp[37].xyz, src0.xyz, src0.www, src0.000
>     MAD temp[45].w, src1.w, src1.x, src2.x
> 24: src0.xyz = temp[37], src0.w = temp[56], src1.w = temp[45]
>     MAD temp[57].xyz, src0.xyz, src0.www, src0.000
>     MAX temp[46].w, src1.w, src0.0
> 25: src0.xyz = temp[37], src0.w = temp[46]
>     MAD temp[47].xyz, src0.xyz, src0.www, src0.000
> 26: src0.xyz = temp[57], src1.xyz = temp[19]
>     MAD temp[58].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = temp[47], src1.xyz = temp[19]
>     MAD temp[48].xyz, src0.xyz, src1.xxx, src0.000
> 28: src0.xyz = temp[48], src1.xyz = const[9]
>     MAD temp[49].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[58], src1.xyz = const[9], src2.xyz = temp[49]
>     MAD temp[10].xyz, src0.xyz, src1.xxx, src2.xyz
> 30: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[1], src1.xyz = const[1]
>     DP3 temp[0].z, src0.xyz, src0.xyz
>     MAD temp[0].w, src0.1, src0.1, -src1.x
>  1: src0.xyz = input[2], src1.xyz = const[8]
>     DP3 temp[3].x, src0.xyz, src0.xyz
>     RCP temp[1].w, src1.y
>  2: src0.xyz = input[1], src1.xyz = const[8]
>     MAD temp[3].y, -src0._z_, src0._1_, -src1._x_
>     MAD temp[2].w, src0.1, src0.1, src0.0
>  3: src0.xyz = temp[3], src0.w = temp[1], src2.xyz = temp[0]
>     MAD_SAT temp[0].z, src0.__y, src0.__w, src0.__0
>     RSQ temp[1].w, |src2.z|
>  4: src0.xyz = input[1], src0.w = temp[1], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
>     RSQ temp[1].w, |src2.x|
>  5: src0.xyz = input[2], src0.w = temp[1], src2.xyz = temp[0]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>     LG2 temp[1].w, src2.z
>  6: src0.xyz = temp[2], src0.w = temp[1], src1.xyz = const[11], src2.xyz = const[6]
>     DP3 temp[0].z, src0.xyz, src1.xyz
>     MAD temp[1].w, src0.w, src2.x, src0.0
>  7: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[0]
>     DP3 temp[0].z, -src0.xyz, src1.xyz
>     MAD temp[3].w, src0.1, |src0.1|, -|src2.z|
>  8: src0.xyz = temp[2], src0.w = temp[3], src1.xyz = temp[1]
>     DP3 temp[3].x, src0.xyz, src1.xyz
>     LG2 temp[3].w, src0.w
>  9: src0.xyz = temp[3], src0.w = temp[3], src1.xyz = temp[2], src2.xyz = const[0]
>     MAD temp[2].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[3].w, src0.w, src2.y, src0.0
> 10: src0.xyz = temp[1], src0.w = temp[3], src1.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src0.111, -src1.xyz
>     EX2 temp[3].w, src0.w
> 11: src0.xyz = const[2], src0.w = temp[3], src1.xyz = temp[1], src2.xyz = const[0]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[3].w, src0.w, src2.x, src0.0
> 12: src0.xyz = temp[0], src0.w = temp[1]
>     MAX temp[0].z, src0.__z, src0.__0
>     EX2 temp[1].w, src0.w
> 13: src0.w = temp[1], src1.xyz = temp[0]
>     MAD temp[0].z, src0.__1, src0.__1, -src0.__w
>     MAD temp[4].w, src0.1, src0.1, -src1.z
> 14: src0.xyz = const[3], src0.w = temp[4], src1.xyz = temp[1], src2.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src1.yyy, src2.xyz
>     LG2 temp[4].w, src0.w
> 15: src0.xyz = const[7], src0.w = temp[1], src2.xyz = temp[0]
>     MAD temp[1].w, src0.x, src0.w, src2.z
> 16: src0.w = temp[1], src1.xyz = const[10], src1.w = temp[4], src2.xyz = const[1]
>     MAD temp[0].z, src0.__w, src1.__x, src0.__0
>     MAD temp[1].w, src1.w, src2.y, src0.0
> 17: src0.xyz = const[4], src1.xyz = temp[1], src2.xyz = temp[2]
>     MAD temp[1].xyz, src0.xyz, src1.zzz, src2.xyz
> 18: src0.w = temp[1]
>     REPL_ALPHA temp[2].x
>     EX2, src0.w
> 19: src0.xyz = temp[1], src1.xyz = const[5]
>     MAD temp[1].xyz, src0.xyz, src0.111, src1.xyz
> 20: BEGIN_TEX;
> 21: TEX temp[4].w, input[0].xy__, 2D[0];
> 22: TEX temp[1].xyz, temp[1].xyz_, CUBE[1] SEM_WAIT SEM_ACQUIRE;
> 23: src0.xyz = temp[1], src0.w = temp[4], src1.xyz = temp[2], src1.w = temp[0], src2.xyz = const[1] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
>     MAD temp[0].w, src1.w, src1.x, src2.x
> 24: src0.xyz = temp[1], src0.w = temp[3], src1.w = temp[0]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>     MAX temp[0].w, src1.w, src0.0
> 25: src0.xyz = temp[1], src0.w = temp[0]
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
> 26: src0.xyz = temp[2], src1.xyz = temp[0]
>     MAD temp[2].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 28: src0.xyz = temp[0], src1.xyz = const[9]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[2], src1.xyz = const[9], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 30: src0.xyz = temp[0], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=39*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 19, tex_end: 0  (code_addr: 000004c0)
>  0: xyz:  t1  c1  t0  bias-> t0.z                 (02000841)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t1.xyz   t1.xxx    op: 00804000 
>       w:      1.0      1.0    -c1.x    op: 0008c891
>  1: xyz:  t2  c8  t0  bias-> t3.x                 (008c0a02)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:     c8.y     t2.x     t2.x    op: 05000004
>  2: xyz:  t1  c8  t0  bias-> t3.y                 (010c0a01)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:  -t1.zzz      1.0  -c8.xxx    op: 00094aa3 
>       w:      1.0      1.0      0.0    op: 00040891
>  3: xyz:  t3  t0  t0  bias-> t0.z                 (02000003)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t3.yyy   t1.www      0.0    op: 40050602 
>       w:   |t0.z|     t3.x     t3.x    op: 05800048
>  4: xyz:  t1  t0  t3  bias-> t1.xyz               (03843001)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t1.xyz   t1.www      0.0    op: 00050600 
>       w:   |t3.x|     t1.x     t1.x    op: 05800046
>  5: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t2.xyz   t1.www      0.0    op: 00050600 
>       w:     t0.z     t2.x     t2.x    op: 04800008
>  6: xyz:  t2 c11  c6  bias-> t0.z                 (02026ac2)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t2.xyz  c11.xyz   t2.xxx    op: 00804200 
>       w:     t1.w     c6.x      0.0    op: 00040309
>  7: xyz:  t1  t2  t0  bias-> t0.z                 (02000081)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:  -t1.xyz   t2.xyz   t1.xxx    op: 00804220 
>       w:      1.0    |1.0|  -|t0.z|    op: 001a2891
>  8: xyz:  t2  t1  t0  bias-> t3.x                 (008c0042)
>       w:  t3  t0  t0  bias-> t3.w                 (008c0003)
>     xyz:   t2.xyz   t1.xyz   t2.xxx    op: 00804200 
>       w:     t3.w     t2.x     t2.x    op: 04800009
>  9: xyz:  t3  t2  c0  bias-> t2.xyz               (038a0083)
>       w:  t3  t0  t0  bias-> t3.w                 (008c0003)
>     xyz:   t3.xxx   t2.xyz      0.0    op: 08050201 
>       w:     t3.w     c0.y      0.0    op: 00040389
> 10: xyz:  t1  t2  t0  bias-> t1.xyz               (03840081)
>       w:  t3  t0  t0  bias-> t3.w                 (008c0003)
>     xyz:   t1.xyz      1.0  -t2.xyz    op: 00090a80 
>       w:     t3.w     t1.x     t1.x    op: 04000009
> 11: xyz:  c2  t1  c0  bias-> t2.xyz               (038a0062)
>       w:  t3  t0  t0  bias-> t3.w                 (008c0003)
>     xyz:   c2.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t3.w     c0.x      0.0    op: 00040309
> 12: xyz:  t0  t0  t0  bias-> t0.z                 (02000000)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t0.xyz      0.0   t0.xxx    op: 02804a00 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 13: xyz:  t0  t0  t0  bias-> t0.z                 (02000000)
>       w:  t1  t0  t0  bias-> t4.w                 (00900001)
>     xyz:      1.0      1.0  -t1.www    op: 000b0a95 
>       w:      1.0      1.0    -t0.z    op: 00094891
> 14: xyz:  c3  t1  t2  bias-> t2.xyz               (03882063)
>       w:  t4  t0  t0  bias-> t4.w                 (00900004)
>     xyz:   c3.xyz   t1.yyy   t2.xyz    op: 00020300 
>       w:     t4.w     c3.x     c3.x    op: 04800009
> 15: xyz:  c7  t0  t0  bias->                      (00000027)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   c7.xxx   c7.xxx   c7.xxx    op: 00004081 
>       w:     c7.x     t1.w     t0.z    op: 00020480
> 16: xyz:  t0 c10  c1  bias-> t0.z                 (02021a80)
>       w:  t1  t4  t0  bias-> t1.w                 (00840101)
>     xyz:   t1.www  c10.xxx      0.0    op: 0005028c 
>       w:     t4.w     c1.y      0.0    op: 0004038a
> 17: xyz:  c4  t1  t2  bias-> t1.xyz               (03842064)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c4.xyz   t1.zzz   t2.xyz    op: 00020380 
>       w:     c4.x     c4.x     c4.x    op: 00000000
> 18: xyz:  t0  t0  t0  bias-> t2.x                 (00880000)
>       w:  t1  t0  t0  bias->                      (00000001)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t1.w     t0.x     t0.x    op: 04000009
> 19: xyz:  t1  c5  t0  bias-> t1.xyz               (03840941)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0   c5.xyz    op: 00010a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 1: alu_offset: 20, tex_offset: 0, alu_end: 7, tex_end: 1  (code_addr: 004201d4)
>  TEX:
>    TEX t4, t0, texture[0]   (00008100)
>    TEX t1, t1, texture[1]   (00008841)
> 20: xyz:  t1  t2  c1  bias-> t1.xyz               (03861081)
>       w:  t4  t0  t0  bias-> t0.w                 (00800004)
>     xyz:   t1.xyz   t4.www      0.0    op: 00050600 
>       w:     t0.w     t2.x     c1.x    op: 0001818a
> 21: xyz:  t1 t32  t0  bias-> t2.xyz               (03880001)
>       w:  t3  t0  t0  bias-> t0.w                 (00800003)
>     xyz:   t1.xyz   t3.www      0.0    op: 00050600 
>       w:     t0.w      0.0     t1.x    op: 0180080a
> 22: xyz:  t1 t32  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t0.www      0.0    op: 00050600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 23: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t0.zzz      0.0    op: 00050380 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 24: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t0.zzz      0.0    op: 00050380 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 25: xyz:  t0  c9  t0  bias-> t0.xyz               (03800a40)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c9.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 26: xyz:  t2  c9  t0  bias-> t0.xyz               (03800a42)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c9.xxx   t0.xyz    op: 00020280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 27: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t2  t0  t0  bias-> o0.w                 (01000002)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>IMM[0] FLT32 {    0.0000,     1.0000,    -2.5000,     0.2500}
>IMM[1] FLT32 {   -0.7500,     0.3000,     0.5000,     0.7500}
>IMM[2] FLT32 {    2.5000,     0.6250,     0.0000,     0.0000}
>  0: MUL TEMP[0].xy, IMM[0].xyyy, CONST[0].xxxx
>  1: MAD TEMP[1].xy, IMM[0].zzzz, TEMP[0].xyyy, IN[0].xyyy
>  2: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  3: MUL TEMP[1].xyz, TEMP[1].xyzz, IMM[0].wwww
>  4: MAD TEMP[2].xy, IMM[1].xxxx, TEMP[0].xyyy, IN[0].xyyy
>  5: TEX TEMP[2].xyz, TEMP[2].xyyy, SAMP[0], RECT
>  6: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].yyyy, TEMP[1].xyzz
>  7: TEX TEMP[2].xyz, IN[0].xyyy, SAMP[0], RECT
>  8: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].zzzz, TEMP[1].xyzz
>  9: MAD TEMP[2].xy, IMM[1].wwww, TEMP[0].xyyy, IN[0].xyyy
> 10: TEX TEMP[2].xyz, TEMP[2].xyyy, SAMP[0], RECT
> 11: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].yyyy, TEMP[1].xyzz
> 12: MAD TEMP[0].xy, IMM[2].xxxx, TEMP[0].xyyy, IN[0].xyyy
> 13: TEX TEMP[0].xyz, TEMP[0].xyyy, SAMP[0], RECT
> 14: MAD TEMP[1].xyz, TEMP[0].xyzz, IMM[0].wwww, TEMP[1].xyzz
> 15: MUL TEMP[0].xyz, TEMP[1].xyzz, IMM[2].yyyy
> 16: MOV TEMP[0].xyz, TEMP[0].xyzx
> 17: MOV TEMP[0].w, IMM[0].yyyy
> 18: MOV OUT[0], TEMP[0]
> 19: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].yyyy;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].yyyy;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].yyyy;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].yyyy;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].yyyy;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: MUL temp[3], temp[1].xyyy, const[4];
>  3: TEX temp[4], temp[3], 2D[0];
>  4: MOV temp[1].xyz, temp[4];
>  5: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  6: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  7: MUL temp[5], temp[2].xyyy, const[4];
>  8: TEX temp[6], temp[5], 2D[0];
>  9: MOV temp[2].xyz, temp[6];
> 10: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 11: MUL temp[7], input[0].xyyy, const[4];
> 12: TEX temp[8], temp[7], 2D[0];
> 13: MOV temp[2].xyz, temp[8];
> 14: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
> 15: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 16: MUL temp[9], temp[2].xyyy, const[4];
> 17: TEX temp[10], temp[9], 2D[0];
> 18: MOV temp[2].xyz, temp[10];
> 19: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 20: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 21: MUL temp[11], temp[0].xyyy, const[4];
> 22: TEX temp[12], temp[11], 2D[0];
> 23: MOV temp[0].xyz, temp[12];
> 24: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 25: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 26: MOV temp[0].xyz, temp[0].xyzx;
> 27: MOV temp[0].w, const[1].yyyy;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: MUL temp[3], temp[1].xyyy, const[4];
>  3: TEX temp[4], temp[3], 2D[0];
>  4: MOV temp[1].xyz, temp[4];
>  5: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  6: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  7: MUL temp[5], temp[2].xyyy, const[4];
>  8: TEX temp[6], temp[5], 2D[0];
>  9: MOV temp[2].xyz, temp[6];
> 10: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 11: MUL temp[7], input[0].xyyy, const[4];
> 12: TEX temp[8], temp[7], 2D[0];
> 13: MOV temp[2].xyz, temp[8];
> 14: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
> 15: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 16: MUL temp[9], temp[2].xyyy, const[4];
> 17: TEX temp[10], temp[9], 2D[0];
> 18: MOV temp[2].xyz, temp[10];
> 19: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 20: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 21: MUL temp[11], temp[0].xyyy, const[4];
> 22: TEX temp[12], temp[11], 2D[0];
> 23: MOV temp[0].xyz, temp[12];
> 24: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 25: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 26: MOV temp[0].xyz, temp[0].xyzx;
> 27: MOV temp[0].w, const[1].yyyy;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[1].xy, const[1].zz__, temp[0].xy__, input[0].xy__;
>  2: MUL temp[3].xy, temp[1].xy__, const[4].xy__;
>  3: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  4: MOV temp[1].xyz, temp[4].xyz_;
>  5: MUL temp[1].xyz, temp[1].xyz_, const[1].www_;
>  6: MAD temp[2].xy, const[2].xx__, temp[0].xy__, input[0].xy__;
>  7: MUL temp[5].xy, temp[2].xy__, const[4].xy__;
>  8: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  9: MOV temp[2].xyz, temp[6].xyz_;
> 10: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 11: MUL temp[7].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[8].xyz, temp[7].xy__, 2D[0];
> 13: MOV temp[2].xyz, temp[8].xyz_;
> 14: MAD temp[1].xyz, temp[2].xyz_, const[2].zzz_, temp[1].xyz_;
> 15: MAD temp[2].xy, const[2].ww__, temp[0].xy__, input[0].xy__;
> 16: MUL temp[9].xy, temp[2].xy__, const[4].xy__;
> 17: TEX temp[10].xyz, temp[9].xy__, 2D[0];
> 18: MOV temp[2].xyz, temp[10].xyz_;
> 19: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 20: MAD temp[0].xy, const[3].xx__, temp[0].xy__, input[0].xy__;
> 21: MUL temp[11].xy, temp[0].xy__, const[4].xy__;
> 22: TEX temp[12].xyz, temp[11].xy__, 2D[0];
> 23: MOV temp[0].xyz, temp[12].xyz_;
> 24: MAD temp[1].xyz, temp[0].xyz_, const[1].www_, temp[1].xyz_;
> 25: MUL temp[0].xyz, temp[1].xyz_, const[3].yyy_;
> 26: MOV temp[0].xyz, temp[0].xyz_;
> 27: MOV temp[0].w, const[1].___y;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[1].xy, const[1].zz__, temp[0].xy__, input[0].xy__;
>  2: MUL temp[3].xy, temp[1].xy__, const[4].xy__;
>  3: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  4: MOV temp[1].xyz, temp[4].xyz_;
>  5: MUL temp[1].xyz, temp[1].xyz_, const[1].www_;
>  6: MAD temp[2].xy, const[2].xx__, temp[0].xy__, input[0].xy__;
>  7: MUL temp[5].xy, temp[2].xy__, const[4].xy__;
>  8: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  9: MOV temp[2].xyz, temp[6].xyz_;
> 10: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 11: MUL temp[7].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[8].xyz, temp[7].xy__, 2D[0];
> 13: MOV temp[2].xyz, temp[8].xyz_;
> 14: MAD temp[1].xyz, temp[2].xyz_, const[2].zzz_, temp[1].xyz_;
> 15: MAD temp[2].xy, const[2].ww__, temp[0].xy__, input[0].xy__;
> 16: MUL temp[9].xy, temp[2].xy__, const[4].xy__;
> 17: TEX temp[10].xyz, temp[9].xy__, 2D[0];
> 18: MOV temp[2].xyz, temp[10].xyz_;
> 19: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 20: MAD temp[0].xy, const[3].xx__, temp[0].xy__, input[0].xy__;
> 21: MUL temp[11].xy, temp[0].xy__, const[4].xy__;
> 22: TEX temp[12].xyz, temp[11].xy__, 2D[0];
> 23: MOV temp[0].xyz, temp[12].xyz_;
> 24: MAD temp[1].xyz, temp[0].xyz_, const[1].www_, temp[1].xyz_;
> 25: MUL temp[0].xyz, temp[1].xyz_, const[3].yyy_;
> 26: MOV temp[0].xyz, temp[0].xyz_;
> 27: MOV temp[0].w, const[1].___y;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MOV temp[17].xyz, temp[16].xyz_;
>  5: MUL temp[18].xyz, temp[17].xyz_, const[1].www_;
>  6: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  7: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  8: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  9: MOV temp[22].xyz, temp[21].xyz_;
> 10: MAD temp[23].xyz, temp[22].xyz_, const[2].yyy_, temp[18].xyz_;
> 11: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 13: MOV temp[26].xyz, temp[25].xyz_;
> 14: MAD temp[27].xyz, temp[26].xyz_, const[2].zzz_, temp[23].xyz_;
> 15: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 16: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 17: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 18: MOV temp[31].xyz, temp[30].xyz_;
> 19: MAD temp[32].xyz, temp[31].xyz_, const[2].yyy_, temp[27].xyz_;
> 20: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 21: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 22: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 23: MOV temp[36].xyz, temp[35].xyz_;
> 24: MAD temp[37].xyz, temp[36].xyz_, const[1].www_, temp[32].xyz_;
> 25: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 26: MOV temp[39].xyz, temp[38].xyz_;
> 27: MOV temp[39].w, const[1].___y;
> 28: MOV output[0], temp[39];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>CONST[1] = {     0.0000     1.0000    -2.5000     0.2500 }
>CONST[2] = {    -0.7500     0.3000     0.5000     0.7500 }
>CONST[3] = {     2.5000     0.6250     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>CONST[1] = {     0.0000     1.0000    -2.5000     0.2500 }
>CONST[2] = {    -0.7500     0.3000     0.5000     0.7500 }
>CONST[3] = {     2.5000     0.6250     0.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>  1: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  2: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: src0.xyz = temp[16], src0.w = const[1]
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
>  5: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
>  9: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 12: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
> 13: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 16: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
> 17: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 20: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 21: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 22: 
>     MAD temp[39].w, src0.1, src0.1, src0.0
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>     MAD temp[39].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  2: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
> 10: BEGIN_TEX;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 12: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 13: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 14: TEX temp[21].xyz, temp[20].xy__, 2D[0];
> 15: TEX temp[16].xyz, temp[15].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[16], src0.w = const[1] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>     MAD temp[39].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  2: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
> 10: BEGIN_TEX;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 12: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 13: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 14: TEX temp[21].xyz, temp[20].xy__, 2D[0];
> 15: TEX temp[16].xyz, temp[15].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[16], src0.w = const[1] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[1].xy, src0.xy_, src1.xx_, src0.00_
>     MAD temp[0].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.00_
>  2: src0.xyz = const[3], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[3].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[1], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[4].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[5].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[0].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[3], src1.xyz = const[4]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[4], src1.xyz = const[4]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[5], src1.xyz = const[4]
>     MAD temp[4].xy, src0.xy_, src1.xy_, src0.00_
>  9: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
> 10: BEGIN_TEX;
> 11: TEX temp[2].xyz, temp[2].xy__, 2D[0];
> 12: TEX temp[1].xyz, temp[1].xy__, 2D[0];
> 13: TEX temp[3].xyz, temp[3].xy__, 2D[0];
> 14: TEX temp[4].xyz, temp[4].xy__, 2D[0];
> 15: TEX temp[5].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[5], src0.w = const[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[4], src1.xyz = const[2], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[2], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[3], src1.xyz = const[2], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[1], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[0], src1.xyz = const[3]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=40*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 9, tex_end: 0  (code_addr: 00000240)
>  0: xyz:  c1  c0  t0  bias-> t1.xy                (01840821)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c1.xyz   c0.xxx      0.0    op: 00050280 
>       w:      1.0      1.0      0.0    op: 00040891
>  1: xyz:  t0  c4  t0  bias-> t2.xy                (01880900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  2: xyz:  c3  t1  t0  bias-> t3.xy                (018c0063)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c3.xxx   t1.xyz   t0.xyz    op: 00020201 
>       w:     c3.x     c3.x     c3.x    op: 00000000
>  3: xyz:  t1  t0  t0  bias-> t4.xy                (01900001)
>       w:  c2  t0  t0  bias->                      (00000022)
>     xyz:   c2.www   t1.xyz   t0.xyz    op: 0001000c 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  4: xyz:  c2  t1  t0  bias-> t5.xy                (01940062)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c2.xxx   t1.xyz   t0.xyz    op: 00020201 
>       w:     c2.x     c2.x     c2.x    op: 00000000
>  5: xyz:  c1  t1  t0  bias-> t0.xy                (01800061)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c1.zzz   t1.xyz   t0.xyz    op: 00020203 
>       w:     c1.x     c1.x     c1.x    op: 00000000
>  6: xyz:  t3  c4  t0  bias-> t1.xy                (01840903)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  7: xyz:  t4  c4  t0  bias-> t3.xy                (018c0904)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t4.x     t4.x     t4.x    op: 00000000
>  8: xyz:  t5  c4  t0  bias-> t4.xy                (01900905)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t5.x     t5.x     t5.x    op: 00000000
>  9: xyz:  t0  c4  t0  bias-> t0.xy                (01800900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 10, tex_offset: 0, alu_end: 7, tex_end: 4  (code_addr: 004801ca)
>  TEX:
>    TEX t2, t2, texture[0]   (00008082)
>    TEX t1, t1, texture[0]   (00008041)
>    TEX t3, t3, texture[0]   (000080c3)
>    TEX t4, t4, texture[0]   (00008104)
>    TEX t5, t0, texture[0]   (00008140)
> 10: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t5.xyz   c1.www      0.0    op: 00050600 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 11: xyz:  t4  c2  t0  bias-> t0.xyz               (03800884)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c2.yyy   t0.xyz    op: 00020300 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 12: xyz:  t2  t0  t0  bias-> t0.xyz               (03800002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      0.5   t0.xyz    op: 00010b00 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t3  c2  t0  bias-> t0.xyz               (03800883)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c2.yyy   t0.xyz    op: 00020300 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 14: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t1.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 15: xyz:  t0  c3  t0  bias-> t0.xyz               (038008c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c3.yyy      0.0    op: 00050300 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 17: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0000,    -2.5000,     0.2500}
>IMM[1] FLT32 {   -0.7500,     0.3000,     0.5000,     0.7500}
>IMM[2] FLT32 {    2.5000,     0.6250,     0.0000,     0.0000}
>  0: MUL TEMP[0].xy, IMM[0].xyyy, CONST[0].xxxx
>  1: MAD TEMP[1].xy, IMM[0].zzzz, TEMP[0].xyyy, IN[0].xyyy
>  2: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  3: MUL TEMP[1].xyz, TEMP[1].xyzz, IMM[0].wwww
>  4: MAD TEMP[2].xy, IMM[1].xxxx, TEMP[0].xyyy, IN[0].xyyy
>  5: TEX TEMP[2].xyz, TEMP[2].xyyy, SAMP[0], RECT
>  6: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].yyyy, TEMP[1].xyzz
>  7: TEX TEMP[2].xyz, IN[0].xyyy, SAMP[0], RECT
>  8: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].zzzz, TEMP[1].xyzz
>  9: MAD TEMP[2].xy, IMM[1].wwww, TEMP[0].xyyy, IN[0].xyyy
> 10: TEX TEMP[2].xyz, TEMP[2].xyyy, SAMP[0], RECT
> 11: MAD TEMP[1].xyz, TEMP[2].xyzz, IMM[1].yyyy, TEMP[1].xyzz
> 12: MAD TEMP[0].xy, IMM[2].xxxx, TEMP[0].xyyy, IN[0].xyyy
> 13: TEX TEMP[0].xyz, TEMP[0].xyyy, SAMP[0], RECT
> 14: MAD TEMP[1].xyz, TEMP[0].xyzz, IMM[0].wwww, TEMP[1].xyzz
> 15: MUL TEMP[0].xyz, TEMP[1].xyzz, IMM[2].yyyy
> 16: MOV TEMP[0].xyz, TEMP[0].xyzx
> 17: MOV TEMP[0].w, IMM[0].xxxx
> 18: MOV OUT[0], TEMP[0]
> 19: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].xxxx;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].xxxx;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].xxxx;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].xxxx;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  3: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  4: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  5: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
>  6: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
>  7: TEX temp[2].xyz, input[0].xyyy, RECT[0];
>  8: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
>  9: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 10: TEX temp[2].xyz, temp[2].xyyy, RECT[0];
> 11: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 12: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 13: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
> 14: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 15: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 16: MOV temp[0].xyz, temp[0].xyzx;
> 17: MOV temp[0].w, const[1].xxxx;
> 18: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: MUL temp[3], temp[1].xyyy, const[4];
>  3: TEX temp[4], temp[3], 2D[0];
>  4: MOV temp[1].xyz, temp[4];
>  5: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  6: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  7: MUL temp[5], temp[2].xyyy, const[4];
>  8: TEX temp[6], temp[5], 2D[0];
>  9: MOV temp[2].xyz, temp[6];
> 10: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 11: MUL temp[7], input[0].xyyy, const[4];
> 12: TEX temp[8], temp[7], 2D[0];
> 13: MOV temp[2].xyz, temp[8];
> 14: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
> 15: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 16: MUL temp[9], temp[2].xyyy, const[4];
> 17: TEX temp[10], temp[9], 2D[0];
> 18: MOV temp[2].xyz, temp[10];
> 19: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 20: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 21: MUL temp[11], temp[0].xyyy, const[4];
> 22: TEX temp[12], temp[11], 2D[0];
> 23: MOV temp[0].xyz, temp[12];
> 24: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 25: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 26: MOV temp[0].xyz, temp[0].xyzx;
> 27: MOV temp[0].w, const[1].xxxx;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xyyy, const[0].xxxx;
>  1: MAD temp[1].xy, const[1].zzzz, temp[0].xyyy, input[0].xyyy;
>  2: MUL temp[3], temp[1].xyyy, const[4];
>  3: TEX temp[4], temp[3], 2D[0];
>  4: MOV temp[1].xyz, temp[4];
>  5: MUL temp[1].xyz, temp[1].xyzz, const[1].wwww;
>  6: MAD temp[2].xy, const[2].xxxx, temp[0].xyyy, input[0].xyyy;
>  7: MUL temp[5], temp[2].xyyy, const[4];
>  8: TEX temp[6], temp[5], 2D[0];
>  9: MOV temp[2].xyz, temp[6];
> 10: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 11: MUL temp[7], input[0].xyyy, const[4];
> 12: TEX temp[8], temp[7], 2D[0];
> 13: MOV temp[2].xyz, temp[8];
> 14: MAD temp[1].xyz, temp[2].xyzz, const[2].zzzz, temp[1].xyzz;
> 15: MAD temp[2].xy, const[2].wwww, temp[0].xyyy, input[0].xyyy;
> 16: MUL temp[9], temp[2].xyyy, const[4];
> 17: TEX temp[10], temp[9], 2D[0];
> 18: MOV temp[2].xyz, temp[10];
> 19: MAD temp[1].xyz, temp[2].xyzz, const[2].yyyy, temp[1].xyzz;
> 20: MAD temp[0].xy, const[3].xxxx, temp[0].xyyy, input[0].xyyy;
> 21: MUL temp[11], temp[0].xyyy, const[4];
> 22: TEX temp[12], temp[11], 2D[0];
> 23: MOV temp[0].xyz, temp[12];
> 24: MAD temp[1].xyz, temp[0].xyzz, const[1].wwww, temp[1].xyzz;
> 25: MUL temp[0].xyz, temp[1].xyzz, const[3].yyyy;
> 26: MOV temp[0].xyz, temp[0].xyzx;
> 27: MOV temp[0].w, const[1].xxxx;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[1].xy, const[1].zz__, temp[0].xy__, input[0].xy__;
>  2: MUL temp[3].xy, temp[1].xy__, const[4].xy__;
>  3: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  4: MOV temp[1].xyz, temp[4].xyz_;
>  5: MUL temp[1].xyz, temp[1].xyz_, const[1].www_;
>  6: MAD temp[2].xy, const[2].xx__, temp[0].xy__, input[0].xy__;
>  7: MUL temp[5].xy, temp[2].xy__, const[4].xy__;
>  8: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  9: MOV temp[2].xyz, temp[6].xyz_;
> 10: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 11: MUL temp[7].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[8].xyz, temp[7].xy__, 2D[0];
> 13: MOV temp[2].xyz, temp[8].xyz_;
> 14: MAD temp[1].xyz, temp[2].xyz_, const[2].zzz_, temp[1].xyz_;
> 15: MAD temp[2].xy, const[2].ww__, temp[0].xy__, input[0].xy__;
> 16: MUL temp[9].xy, temp[2].xy__, const[4].xy__;
> 17: TEX temp[10].xyz, temp[9].xy__, 2D[0];
> 18: MOV temp[2].xyz, temp[10].xyz_;
> 19: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 20: MAD temp[0].xy, const[3].xx__, temp[0].xy__, input[0].xy__;
> 21: MUL temp[11].xy, temp[0].xy__, const[4].xy__;
> 22: TEX temp[12].xyz, temp[11].xy__, 2D[0];
> 23: MOV temp[0].xyz, temp[12].xyz_;
> 24: MAD temp[1].xyz, temp[0].xyz_, const[1].www_, temp[1].xyz_;
> 25: MUL temp[0].xyz, temp[1].xyz_, const[3].yyy_;
> 26: MOV temp[0].xyz, temp[0].xyz_;
> 27: MOV temp[0].w, const[1].___x;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[0].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[1].xy, const[1].zz__, temp[0].xy__, input[0].xy__;
>  2: MUL temp[3].xy, temp[1].xy__, const[4].xy__;
>  3: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  4: MOV temp[1].xyz, temp[4].xyz_;
>  5: MUL temp[1].xyz, temp[1].xyz_, const[1].www_;
>  6: MAD temp[2].xy, const[2].xx__, temp[0].xy__, input[0].xy__;
>  7: MUL temp[5].xy, temp[2].xy__, const[4].xy__;
>  8: TEX temp[6].xyz, temp[5].xy__, 2D[0];
>  9: MOV temp[2].xyz, temp[6].xyz_;
> 10: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 11: MUL temp[7].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[8].xyz, temp[7].xy__, 2D[0];
> 13: MOV temp[2].xyz, temp[8].xyz_;
> 14: MAD temp[1].xyz, temp[2].xyz_, const[2].zzz_, temp[1].xyz_;
> 15: MAD temp[2].xy, const[2].ww__, temp[0].xy__, input[0].xy__;
> 16: MUL temp[9].xy, temp[2].xy__, const[4].xy__;
> 17: TEX temp[10].xyz, temp[9].xy__, 2D[0];
> 18: MOV temp[2].xyz, temp[10].xyz_;
> 19: MAD temp[1].xyz, temp[2].xyz_, const[2].yyy_, temp[1].xyz_;
> 20: MAD temp[0].xy, const[3].xx__, temp[0].xy__, input[0].xy__;
> 21: MUL temp[11].xy, temp[0].xy__, const[4].xy__;
> 22: TEX temp[12].xyz, temp[11].xy__, 2D[0];
> 23: MOV temp[0].xyz, temp[12].xyz_;
> 24: MAD temp[1].xyz, temp[0].xyz_, const[1].www_, temp[1].xyz_;
> 25: MUL temp[0].xyz, temp[1].xyz_, const[3].yyy_;
> 26: MOV temp[0].xyz, temp[0].xyz_;
> 27: MOV temp[0].w, const[1].___x;
> 28: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MOV temp[17].xyz, temp[16].xyz_;
>  5: MUL temp[18].xyz, temp[17].xyz_, const[1].www_;
>  6: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  7: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  8: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  9: MOV temp[22].xyz, temp[21].xyz_;
> 10: MAD temp[23].xyz, temp[22].xyz_, const[2].yyy_, temp[18].xyz_;
> 11: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 13: MOV temp[26].xyz, temp[25].xyz_;
> 14: MAD temp[27].xyz, temp[26].xyz_, const[2].zzz_, temp[23].xyz_;
> 15: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 16: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 17: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 18: MOV temp[31].xyz, temp[30].xyz_;
> 19: MAD temp[32].xyz, temp[31].xyz_, const[2].yyy_, temp[27].xyz_;
> 20: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 21: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 22: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 23: MOV temp[36].xyz, temp[35].xyz_;
> 24: MAD temp[37].xyz, temp[36].xyz_, const[1].www_, temp[32].xyz_;
> 25: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 26: MOV temp[39].xyz, temp[38].xyz_;
> 27: MOV temp[39].w, const[1].___x;
> 28: MOV output[0], temp[39];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>CONST[1] = {     1.0000     0.0000    -2.5000     0.2500 }
>CONST[2] = {    -0.7500     0.3000     0.5000     0.7500 }
>CONST[3] = {     2.5000     0.6250     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>CONST[1] = {     1.0000     0.0000    -2.5000     0.2500 }
>CONST[2] = {    -0.7500     0.3000     0.5000     0.7500 }
>CONST[3] = {     2.5000     0.6250     0.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[13].xy, const[1].xy__, const[0].xx__;
>  1: MAD temp[14].xy, const[1].zz__, temp[13].xy__, input[0].xy__;
>  2: MUL temp[15].xy, temp[14].xy__, const[4].xy__;
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: MUL temp[18].xyz, temp[16].xyz_, const[1].www_;
>  5: MAD temp[19].xy, const[2].xx__, temp[13].xy__, input[0].xy__;
>  6: MUL temp[20].xy, temp[19].xy__, const[4].xy__;
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: MAD temp[23].xyz, temp[21].xyz_, const[2].yyy_, temp[18].xyz_;
>  9: MUL temp[24].xy, input[0].xy__, const[4].xy__;
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: MAD temp[27].xyz, temp[25].xyz_, none.HHH_, temp[23].xyz_;
> 12: MAD temp[28].xy, const[2].ww__, temp[13].xy__, input[0].xy__;
> 13: MUL temp[29].xy, temp[28].xy__, const[4].xy__;
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: MAD temp[32].xyz, temp[30].xyz_, const[2].yyy_, temp[27].xyz_;
> 16: MAD temp[33].xy, const[3].xx__, temp[13].xy__, input[0].xy__;
> 17: MUL temp[34].xy, temp[33].xy__, const[4].xy__;
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: MAD temp[37].xyz, temp[35].xyz_, const[1].www_, temp[32].xyz_;
> 20: MUL temp[38].xyz, temp[37].xyz_, const[3].yyy_;
> 21: MOV temp[39].xyz, temp[38].xyz_;
> 22: MOV temp[39].w, none.___1;
> 23: MOV output[0], temp[39];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>  1: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  2: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
>  3: TEX temp[16].xyz, temp[15].xy__, 2D[0];
>  4: src0.xyz = temp[16], src0.w = const[1]
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
>  5: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  7: TEX temp[21].xyz, temp[20].xy__, 2D[0];
>  8: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
>  9: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 11: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 12: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
> 13: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
> 14: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 15: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 16: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
> 17: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
> 18: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 19: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 20: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 21: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 22: 
>     MAD temp[39].w, src0.1, src0.1, src0.0
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>     MAD temp[39].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  2: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
> 10: BEGIN_TEX;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 12: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 13: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 14: TEX temp[21].xyz, temp[20].xy__, 2D[0];
> 15: TEX temp[16].xyz, temp[15].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[16], src0.w = const[1] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[13].xy, src0.xy_, src1.xx_, src0.000
>     MAD temp[39].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  2: src0.xyz = const[3], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[33].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[13], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[28].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[19].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[13], src2.xyz = input[0]
>     MAD temp[14].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[33], src1.xyz = const[4]
>     MAD temp[34].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[28], src1.xyz = const[4]
>     MAD temp[29].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[19], src1.xyz = const[4]
>     MAD temp[20].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[14], src1.xyz = const[4]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src0.000
> 10: BEGIN_TEX;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[0];
> 12: TEX temp[35].xyz, temp[34].xy__, 2D[0];
> 13: TEX temp[30].xyz, temp[29].xy__, 2D[0];
> 14: TEX temp[21].xyz, temp[20].xy__, 2D[0];
> 15: TEX temp[16].xyz, temp[15].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[16], src0.w = const[1] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = temp[18]
>     MAD temp[23].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[25], src1.xyz = temp[23]
>     MAD temp[27].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[30], src1.xyz = const[2], src2.xyz = temp[27]
>     MAD temp[32].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[35], src0.w = const[1], src1.xyz = temp[32]
>     MAD temp[37].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[37], src1.xyz = const[3]
>     MAD temp[38].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[38]
>     MAD temp[39].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[39], src0.w = temp[39]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = const[1], src1.xyz = const[0]
>     MAD temp[1].xy, src0.xy_, src1.xx_, src0.00_
>     MAD temp[0].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.00_
>  2: src0.xyz = const[3], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[3].xy, src0.xx_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[1], src0.w = const[2], src1.xyz = input[0]
>     MAD temp[4].xy, src0.ww_, src0.xy_, src1.xy_
>  4: src0.xyz = const[2], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[5].xy, src0.xx_, src1.xy_, src2.xy_
>  5: src0.xyz = const[1], src1.xyz = temp[1], src2.xyz = input[0]
>     MAD temp[0].xy, src0.zz_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[3], src1.xyz = const[4]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[4], src1.xyz = const[4]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[5], src1.xyz = const[4]
>     MAD temp[4].xy, src0.xy_, src1.xy_, src0.00_
>  9: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
> 10: BEGIN_TEX;
> 11: TEX temp[2].xyz, temp[2].xy__, 2D[0];
> 12: TEX temp[1].xyz, temp[1].xy__, 2D[0];
> 13: TEX temp[3].xyz, temp[3].xy__, 2D[0];
> 14: TEX temp[4].xyz, temp[4].xy__, 2D[0];
> 15: TEX temp[5].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 16: src0.xyz = temp[5], src0.w = const[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 17: src0.xyz = temp[4], src1.xyz = const[2], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src2.xyz
> 18: src0.xyz = temp[2], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src1.xyz
> 19: src0.xyz = temp[3], src1.xyz = const[2], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src2.xyz
> 20: src0.xyz = temp[1], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 21: src0.xyz = temp[0], src1.xyz = const[3]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src0.000
> 22: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
> 23: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=41*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 9, tex_end: 0  (code_addr: 00000240)
>  0: xyz:  c1  c0  t0  bias-> t1.xy                (01840821)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c1.xyz   c0.xxx      0.0    op: 00050280 
>       w:      1.0      1.0      0.0    op: 00040891
>  1: xyz:  t0  c4  t0  bias-> t2.xy                (01880900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  2: xyz:  c3  t1  t0  bias-> t3.xy                (018c0063)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c3.xxx   t1.xyz   t0.xyz    op: 00020201 
>       w:     c3.x     c3.x     c3.x    op: 00000000
>  3: xyz:  t1  t0  t0  bias-> t4.xy                (01900001)
>       w:  c2  t0  t0  bias->                      (00000022)
>     xyz:   c2.www   t1.xyz   t0.xyz    op: 0001000c 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  4: xyz:  c2  t1  t0  bias-> t5.xy                (01940062)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c2.xxx   t1.xyz   t0.xyz    op: 00020201 
>       w:     c2.x     c2.x     c2.x    op: 00000000
>  5: xyz:  c1  t1  t0  bias-> t0.xy                (01800061)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c1.zzz   t1.xyz   t0.xyz    op: 00020203 
>       w:     c1.x     c1.x     c1.x    op: 00000000
>  6: xyz:  t3  c4  t0  bias-> t1.xy                (01840903)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  7: xyz:  t4  c4  t0  bias-> t3.xy                (018c0904)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t4.x     t4.x     t4.x    op: 00000000
>  8: xyz:  t5  c4  t0  bias-> t4.xy                (01900905)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t5.x     t5.x     t5.x    op: 00000000
>  9: xyz:  t0  c4  t0  bias-> t0.xy                (01800900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 10, tex_offset: 0, alu_end: 7, tex_end: 4  (code_addr: 004801ca)
>  TEX:
>    TEX t2, t2, texture[0]   (00008082)
>    TEX t1, t1, texture[0]   (00008041)
>    TEX t3, t3, texture[0]   (000080c3)
>    TEX t4, t4, texture[0]   (00008104)
>    TEX t5, t0, texture[0]   (00008140)
> 10: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t5.xyz   c1.www      0.0    op: 00050600 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 11: xyz:  t4  c2  t0  bias-> t0.xyz               (03800884)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c2.yyy   t0.xyz    op: 00020300 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 12: xyz:  t2  t0  t0  bias-> t0.xyz               (03800002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      0.5   t0.xyz    op: 00010b00 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t3  c2  t0  bias-> t0.xyz               (03800883)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c2.yyy   t0.xyz    op: 00020300 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 14: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t1.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 15: xyz:  t0  c3  t0  bias-> t0.xyz               (038008c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c3.yyy      0.0    op: 00050300 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 17: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[1], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[1], RECT
>  1: DP3 TEMP[1].x, TEMP[0].xyzz, CONST[0].xyzz
>  2: MUL TEMP[1], TEMP[0], TEMP[1].xxxx
>  3: TEX TEMP[2], IN[1].xyyy, SAMP[0], RECT
>  4: MAD TEMP[0], TEMP[0], TEMP[1], TEMP[2]
>  5: MOV OUT[0], TEMP[0]
>  6: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, RECT[1];
>  1: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  2: MUL temp[1], temp[0], temp[1].xxxx;
>  3: TEX temp[2], input[1].xyyy, RECT[0];
>  4: MAD temp[0], temp[0], temp[1], temp[2];
>  5: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, RECT[1];
>  1: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  2: MUL temp[1], temp[0], temp[1].xxxx;
>  3: TEX temp[2], input[1].xyyy, RECT[0];
>  4: MAD temp[0], temp[0], temp[1], temp[2];
>  5: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, RECT[1];
>  1: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  2: MUL temp[1], temp[0], temp[1].xxxx;
>  3: TEX temp[2], input[1].xyyy, RECT[0];
>  4: MAD temp[0], temp[0], temp[1], temp[2];
>  5: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, RECT[1];
>  1: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  2: MUL temp[1], temp[0], temp[1].xxxx;
>  3: TEX temp[2], input[1].xyyy, RECT[0];
>  4: MAD temp[0], temp[0], temp[1], temp[2];
>  5: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, RECT[1];
>  1: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  2: MUL temp[1], temp[0], temp[1].xxxx;
>  3: TEX temp[2], input[1].xyyy, RECT[0];
>  4: MAD temp[0], temp[0], temp[1], temp[2];
>  5: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[3], input[0].xyyy, const[1];
>  1: TEX temp[0], temp[3], 2D[1];
>  2: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  3: MUL temp[1], temp[0], temp[1].xxxx;
>  4: MUL temp[4], input[1].xyyy, const[2];
>  5: TEX temp[2], temp[4], 2D[0];
>  6: MAD temp[0], temp[0], temp[1], temp[2];
>  7: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[3], input[0].xyyy, const[1];
>  1: TEX temp[0], temp[3], 2D[1];
>  2: DP3 temp[1].x, temp[0].xyzz, const[0].xyzz;
>  3: MUL temp[1], temp[0], temp[1].xxxx;
>  4: MUL temp[4], input[1].xyyy, const[2];
>  5: TEX temp[2], temp[4], 2D[0];
>  6: MAD temp[0], temp[0], temp[1], temp[2];
>  7: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[0], temp[3].xy__, 2D[1];
>  2: DP3 temp[1].x, temp[0].xyz_, const[0].xyz_;
>  3: MUL temp[1], temp[0], temp[1].xxxx;
>  4: MUL temp[4].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[2], temp[4].xy__, 2D[0];
>  6: MAD temp[0], temp[0], temp[1], temp[2];
>  7: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[0], temp[3].xy__, 2D[1];
>  2: DP3 temp[1].x, temp[0].xyz_, const[0].xyz_;
>  3: MUL temp[1], temp[0], temp[1].xxxx;
>  4: MUL temp[4].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[2], temp[4].xy__, 2D[0];
>  6: MAD temp[0], temp[0], temp[1], temp[2];
>  7: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[5].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[6], temp[5].xy__, 2D[1];
>  2: DP3 temp[7].x, temp[6].xyz_, const[0].xyz_;
>  3: MUL temp[8], temp[6], temp[7].xxxx;
>  4: MUL temp[9].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[10], temp[9].xy__, 2D[0];
>  6: MAD temp[11], temp[6], temp[8], temp[10];
>  7: MOV output[0], temp[11];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[5].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[6], temp[5].xy__, 2D[1];
>  2: DP3 temp[7].x, temp[6].xyz_, const[0].xyz_;
>  3: MUL temp[8], temp[6], temp[7].xxxx;
>  4: MUL temp[9].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[10], temp[9].xy__, 2D[0];
>  6: MAD temp[11], temp[6], temp[8], temp[10];
>  7: MOV output[0], temp[11];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[5].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[6], temp[5].xy__, 2D[1];
>  2: DP3 temp[7].x, temp[6].xyz_, const[0].xyz_;
>  3: MUL temp[8], temp[6], temp[7].xxxx;
>  4: MUL temp[9].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[10], temp[9].xy__, 2D[0];
>  6: MAD temp[11], temp[6], temp[8], temp[10];
>  7: MOV output[0], temp[11];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[5].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[6], temp[5].xy__, 2D[1];
>  2: DP3 temp[7].x, temp[6].xyz_, const[0].xyz_;
>  3: MUL temp[8], temp[6], temp[7].xxxx;
>  4: MUL temp[9].xy, input[1].xy__, const[2].xy__;
>  5: TEX temp[10], temp[9].xy__, 2D[0];
>  6: MAD temp[11], temp[6], temp[8], temp[10];
>  7: MOV output[0], temp[11];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[5].xy, src0.xy_, src1.xy_, src0.000
>  1: TEX temp[6], temp[5].xy__, 2D[1];
>  2: src0.xyz = temp[6], src1.xyz = const[0]
>     DP3 temp[7].x, src0.xyz, src1.xyz
>  3: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[7]
>     MAD temp[8].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[8].w, src0.w, src1.x, src0.0
>  4: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  5: TEX temp[10], temp[9].xy__, 2D[0];
>  6: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[8], src1.w = temp[8], src2.xyz = temp[10], src2.w = temp[10]
>     MAD temp[11].xyz, src0.xyz, src1.xyz, src2.xyz
>     MAD temp[11].w, src0.w, src1.w, src2.w
>  7: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[5].xy, src0.xy_, src1.xy_, src0.000
>  1: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  2: BEGIN_TEX;
>  3: TEX temp[6], temp[5].xy__, 2D[1];
>  4: TEX temp[10], temp[9].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  5: src0.xyz = temp[6], src1.xyz = const[0] SEM_WAIT
>     DP3 temp[7].x, src0.xyz, src1.xyz
>  6: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[7]
>     MAD temp[8].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[8].w, src0.w, src1.x, src0.0
>  7: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[8], src1.w = temp[8], src2.xyz = temp[10], src2.w = temp[10]
>     MAD temp[11].xyz, src0.xyz, src1.xyz, src2.xyz
>     MAD temp[11].w, src0.w, src1.w, src2.w
>  8: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[5].xy, src0.xy_, src1.xy_, src0.000
>  1: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[9].xy, src0.xy_, src1.xy_, src0.000
>  2: BEGIN_TEX;
>  3: TEX temp[6], temp[5].xy__, 2D[1];
>  4: TEX temp[10], temp[9].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  5: src0.xyz = temp[6], src1.xyz = const[0] SEM_WAIT
>     DP3 temp[7].x, src0.xyz, src1.xyz
>  6: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[7]
>     MAD temp[8].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[8].w, src0.w, src1.x, src0.0
>  7: src0.xyz = temp[6], src0.w = temp[6], src1.xyz = temp[8], src1.w = temp[8], src2.xyz = temp[10], src2.w = temp[10]
>     MAD temp[11].xyz, src0.xyz, src1.xyz, src2.xyz
>     MAD temp[11].w, src0.w, src1.w, src2.w
>  8: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>  1: src0.xyz = input[1], src1.xyz = const[2]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  2: BEGIN_TEX;
>  3: TEX temp[0], temp[0].xy__, 2D[1];
>  4: TEX temp[1], temp[1].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  5: src0.xyz = temp[0], src1.xyz = const[0] SEM_WAIT
>     DP3 temp[2].x, src0.xyz, src1.xyz
>  6: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
>     MAD temp[2].w, src0.w, src1.x, src0.0
>  7: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = temp[2], src1.w = temp[2], src2.xyz = temp[1], src2.w = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src2.xyz
>     MAD temp[0].w, src0.w, src1.w, src2.w
>  8: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=42*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00000040)
>  0: xyz:  t0  c1  t0  bias-> t0.xy                (01800840)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c1.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  1: xyz:  t1  c2  t0  bias-> t1.xy                (01840881)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c2.xyz      0.0    op: 00050200 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>NODE 1: alu_offset: 2, tex_offset: 0, alu_end: 3, tex_end: 1  (code_addr: 004200c2)
>  TEX:
>    TEX t0, t0, texture[1]   (00008800)
>    TEX t1, t1, texture[0]   (00008041)
>  2: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c0.xyz   t0.xxx    op: 00804200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  t2  t0  bias-> t2.xyz               (03880080)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz   t2.xxx      0.0    op: 00050280 
>       w:     t0.w     t2.x      0.0    op: 00040189
>  4: xyz:  t0  t2  t1  bias-> t0.xyz               (03801080)
>       w:  t0  t2  t1  bias-> t0.w                 (00801080)
>     xyz:   t0.xyz   t2.xyz   t1.xyz    op: 00020200 
>       w:     t0.w     t2.w     t1.w    op: 0002c509
>  5: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], LINEAR
>DCL OUT[0], COLOR
>  0: MOV OUT[0], IN[0]
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], input[0];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src0.w = input[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=43*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL CONST[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0].xyz, IN[0].xyyy, SAMP[0], RECT
>  1: MOV TEMP[0].xyz, TEMP[0].xyzx
>  2: MOV TEMP[0].w, CONST[0].xxxx
>  3: MOV OUT[0], TEMP[0]
>  4: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0].xyz, input[0].xyyy, RECT[0];
>  1: MOV temp[0].xyz, temp[0].xyzx;
>  2: MOV temp[0].w, const[0].xxxx;
>  3: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0].xyz, input[0].xyyy, RECT[0];
>  1: MOV temp[0].xyz, temp[0].xyzx;
>  2: MOV temp[0].w, const[0].xxxx;
>  3: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0].xyz, input[0].xyyy, RECT[0];
>  1: MOV temp[0].xyz, temp[0].xyzx;
>  2: MOV temp[0].w, const[0].xxxx;
>  3: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0].xyz, input[0].xyyy, RECT[0];
>  1: MOV temp[0].xyz, temp[0].xyzx;
>  2: MOV temp[0].w, const[0].xxxx;
>  3: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0].xyz, input[0].xyyy, RECT[0];
>  1: MOV temp[0].xyz, temp[0].xyzx;
>  2: MOV temp[0].w, const[0].xxxx;
>  3: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MUL temp[1], input[0].xyyy, const[1];
>  1: TEX temp[2], temp[1], 2D[0];
>  2: MOV temp[0].xyz, temp[2];
>  3: MOV temp[0].xyz, temp[0].xyzx;
>  4: MOV temp[0].w, const[0].xxxx;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MUL temp[1], input[0].xyyy, const[1];
>  1: TEX temp[2], temp[1], 2D[0];
>  2: MOV temp[0].xyz, temp[2];
>  3: MOV temp[0].xyz, temp[0].xyzx;
>  4: MOV temp[0].w, const[0].xxxx;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MUL temp[1].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[2].xyz, temp[1].xy__, 2D[0];
>  2: MOV temp[0].xyz, temp[2].xyz_;
>  3: MOV temp[0].xyz, temp[0].xyz_;
>  4: MOV temp[0].w, const[0].___x;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MUL temp[1].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[2].xyz, temp[1].xy__, 2D[0];
>  2: MOV temp[0].xyz, temp[2].xyz_;
>  3: MOV temp[0].xyz, temp[0].xyz_;
>  4: MOV temp[0].w, const[0].___x;
>  5: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  2: MOV temp[5].xyz, temp[4].xyz_;
>  3: MOV temp[6].xyz, temp[5].xyz_;
>  4: MOV temp[6].w, const[0].___x;
>  5: MOV output[0], temp[6];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  2: MOV temp[6].xyz, temp[4].xyz_;
>  3: MOV temp[6].w, const[0].___x;
>  4: MOV output[0], temp[6];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  2: MOV temp[6].xyz, temp[4].xyz_;
>  3: MOV temp[6].w, const[0].___x;
>  4: MOV output[0], temp[6];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[3].xy, input[0].xy__, const[1].xy__;
>  1: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  2: MOV temp[6].xyz, temp[4].xyz_;
>  3: MOV temp[6].w, const[0].___x;
>  4: MOV output[0], temp[6];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.000
>  1: TEX temp[4].xyz, temp[3].xy__, 2D[0];
>  2: src0.xyz = temp[4]
>     MAD temp[6].xyz, src0.xyz, src0.111, src0.000
>  3: src0.xyz = const[0]
>     MAD temp[6].w, src0.x, src0.1, src0.0
>  4: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1], src2.xyz = const[0]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[6].w, src2.x, src0.1, src0.0
>  1: BEGIN_TEX;
>  2: TEX temp[4].xyz, temp[3].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[4] SEM_WAIT
>     MAD temp[6].xyz, src0.xyz, src0.111, src0.000
>  4: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1], src2.xyz = const[0]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.000
>     MAD temp[6].w, src2.x, src0.1, src0.0
>  1: BEGIN_TEX;
>  2: TEX temp[4].xyz, temp[3].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[4] SEM_WAIT
>     MAD temp[6].xyz, src0.xyz, src0.111, src0.000
>  4: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[1], src2.xyz = const[0]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>     MAD temp[0].w, src2.x, src0.1, src0.0
>  1: BEGIN_TEX;
>  2: TEX temp[1].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = temp[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
>  4: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=44*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00000000)
>  0: xyz:  t0  c1  c0  bias-> t0.xy                (01820840)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xyz   c1.xyz      0.0    op: 00050200 
>       w:     c0.x      1.0      0.0    op: 00040886
>NODE 1: alu_offset: 1, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400041)
>  TEX:
>    TEX t1, t0, texture[0]   (00008040)
>  1: xyz:  t1  t0  t0  bias-> t0.xyz               (03800001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  2: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[6..7]
>DCL CONST[4..5]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>IMM[0] FLT32 {    1.0000,     0.0039,     0.0000,     2.0000}
>IMM[1] FLT32 {   -1.0000,     0.0000,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[7].xxxx, CONST[7].yyyy
>  2: TEX TEMP[1].xyz, TEMP[0].xyyy, SAMP[1], RECT
>  3: DP3 TEMP[1].x, TEMP[1].xyzz, IMM[0].xyzz
>  4: MUL TEMP[1].xyz, IN[1].xyzz, TEMP[1].xxxx
>  5: ADD TEMP[1].xyz, CONST[6].xyzz, -TEMP[1].xyzz
>  6: MUL TEMP[1].xyz, TEMP[1].xyzz, CONST[5].xxxx
>  7: TEX TEMP[2].xyz, TEMP[0].xyyy, SAMP[3], RECT
>  8: MUL TEMP[2].xyz, TEMP[2].xyzz, CONST[4].xyzz
>  9: DP3 TEMP[3].x, TEMP[1].xyzz, TEMP[1].xyzz
> 10: RSQ TEMP[3].x, TEMP[3].xxxx
> 11: MUL TEMP[3].xyz, TEMP[1].xyzz, TEMP[3].xxxx
> 12: TEX TEMP[4].xyz, TEMP[0].xyyy, SAMP[2], RECT
> 13: MAD TEMP[4].xyz, TEMP[4].xyzz, IMM[0].wwww, IMM[1].xxxx
> 14: DP3 TEMP[3].x, TEMP[3].xyzz, TEMP[4].xyzz
> 15: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].yyyy
> 16: MUL TEMP[2].xyz, TEMP[2].xyzz, TEMP[3].xxxx
> 17: DP3 TEMP[1].x, TEMP[1].xyzz, TEMP[1].xyzz
> 18: TEX TEMP[1].x, TEMP[1].xxxx, SAMP[0], 1D
> 19: MUL TEMP[1].xyz, TEMP[2].xyzz, TEMP[1].xxxx
> 20: MOV OUT[0], TEMP[1]
> 21: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[1];
>  6: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  7: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
>  8: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
>  9: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 10: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 11: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 12: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 13: RSQ temp[3].x, temp[3].xxxx;
> 14: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 15: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 16: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 18: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 19: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 20: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 21: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 22: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 23: MOV output[0], temp[1];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[1];
>  6: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  7: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
>  8: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
>  9: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 10: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 11: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 12: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 13: RSQ temp[3].x, temp[3].xxxx;
> 14: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 15: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 16: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 18: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 19: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 20: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 21: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 22: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 23: MOV output[0], temp[1];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[1];
>  6: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  7: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
>  8: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
>  9: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 10: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 11: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 12: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 13: RSQ temp[3].x, temp[3].xxxx;
> 14: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 15: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 16: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 18: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 19: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 20: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 21: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 22: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 23: MOV output[0], temp[1];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[1];
>  6: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  7: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
>  8: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
>  9: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 10: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 11: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 12: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 13: RSQ temp[3].x, temp[3].xxxx;
> 14: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 15: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 16: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 18: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 19: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 20: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 21: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 22: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 23: MOV output[0], temp[1];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: TEX temp[1].xyz, temp[0].xyyy, RECT[1];
>  6: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  7: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
>  8: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
>  9: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 10: TEX temp[2].xyz, temp[0].xyyy, RECT[3];
> 11: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 12: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 13: RSQ temp[3].x, temp[3].xxxx;
> 14: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 15: TEX temp[4].xyz, temp[0].xyyy, RECT[2];
> 16: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 17: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 18: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 19: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 20: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 21: TEX temp[1].x, temp[1].xxxx, 1D[0];
> 22: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 23: MOV output[0], temp[1];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: MUL temp[6], temp[0].xyyy, const[12];
>  6: TEX temp[7], temp[6], 2D[1];
>  7: MOV temp[1].xyz, temp[7];
>  8: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  9: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
> 10: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
> 11: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 12: MUL temp[8], temp[0].xyyy, const[13];
> 13: TEX temp[9], temp[8], 2D[3];
> 14: MOV temp[2].xyz, temp[9];
> 15: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: MUL temp[10], temp[0].xyyy, const[14];
> 20: TEX temp[11], temp[10], 2D[2];
> 21: MOV temp[4].xyz, temp[11];
> 22: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 23: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 24: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 25: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 26: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 27: TEX temp[12], temp[1].xxxx, 1D[0];
> 28: MOV temp[1].x, temp[12];
> 29: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 30: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].wwww;
>  1: MUL temp[5].xyz, input[0], temp[5].wwww;
>  2: MAD temp[5].xyz, temp[5].xyz0, const[10].xyz0, const[11].xyz0;
>  3: MOV temp[0], temp[5];
>  4: MAD temp[0].y, temp[5], const[7].xxxx, const[7].yyyy;
>  5: MUL temp[6], temp[0].xyyy, const[12];
>  6: TEX temp[7], temp[6], 2D[1];
>  7: MOV temp[1].xyz, temp[7];
>  8: DP3 temp[1].x, temp[1].xyzz, const[8].xyzz;
>  9: MUL temp[1].xyz, input[1].xyzz, temp[1].xxxx;
> 10: ADD temp[1].xyz, const[6].xyzz, -temp[1].xyzz;
> 11: MUL temp[1].xyz, temp[1].xyzz, const[5].xxxx;
> 12: MUL temp[8], temp[0].xyyy, const[13];
> 13: TEX temp[9], temp[8], 2D[3];
> 14: MOV temp[2].xyz, temp[9];
> 15: MUL temp[2].xyz, temp[2].xyzz, const[4].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, |temp[3].xxxx|;
> 18: MUL temp[3].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: MUL temp[10], temp[0].xyyy, const[14];
> 20: TEX temp[11], temp[10], 2D[2];
> 21: MOV temp[4].xyz, temp[11];
> 22: MAD temp[4].xyz, temp[4].xyzz, const[8].wwww, const[9].xxxx;
> 23: DP3 temp[3].x, temp[3].xyzz, temp[4].xyzz;
> 24: MAX temp[3].x, temp[3].xxxx, const[9].yyyy;
> 25: MUL temp[2].xyz, temp[2].xyzz, temp[3].xxxx;
> 26: DP3 temp[1].x, temp[1].xyzz, temp[1].xyzz;
> 27: TEX temp[12], temp[1].xxxx, 1D[0];
> 28: MOV temp[1].x, temp[12];
> 29: MUL temp[1].xyz, temp[2].xyzz, temp[1].xxxx;
> 30: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[5].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[5].xy, temp[5].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[0].x, temp[5].x___;
>  4: MAD temp[0].y, temp[5]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[6].xy, temp[0].xy__, const[12].xy__;
>  6: TEX temp[7].xyz, temp[6].xy__, 2D[1];
>  7: MOV temp[1].xyz, temp[7].xyz_;
>  8: DP3 temp[1].x, temp[1].xyz_, const[8].xyz_;
>  9: MUL temp[1].xyz, input[1].xyz_, temp[1].xxx_;
> 10: ADD temp[1].xyz, const[6].xyz_, -temp[1].xyz_;
> 11: MUL temp[1].xyz, temp[1].xyz_, const[5].xxx_;
> 12: MUL temp[8].xy, temp[0].xy__, const[13].xy__;
> 13: TEX temp[9].xyz, temp[8].xy__, 2D[3];
> 14: MOV temp[2].xyz, temp[9].xyz_;
> 15: MUL temp[2].xyz, temp[2].xyz_, const[4].xyz_;
> 16: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 17: RSQ temp[3].x, |temp[3].x___|;
> 18: MUL temp[3].xyz, temp[1].xyz_, temp[3].xxx_;
> 19: MUL temp[10].xy, temp[0].xy__, const[14].xy__;
> 20: TEX temp[11].xyz, temp[10].xy__, 2D[2];
> 21: MOV temp[4].xyz, temp[11].xyz_;
> 22: MAD temp[4].xyz, temp[4].xyz_, const[8].www_, const[9].xxx_;
> 23: DP3 temp[3].x, temp[3].xyz_, temp[4].xyz_;
> 24: MAX temp[3].x, temp[3].x___, const[9].y___;
> 25: MUL temp[2].xyz, temp[2].xyz_, temp[3].xxx_;
> 26: DP3 temp[1].x, temp[1].xyz_, temp[1].xyz_;
> 27: TEX temp[12].x, temp[1].x___, 1D[0];
> 28: MOV temp[1].x, temp[12].x___;
> 29: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 30: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[5].w, input[0].___w;
>  1: MUL temp[5].xy, input[0].xy__, temp[5].ww__;
>  2: MAD temp[5].xy, temp[5].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[0].x, temp[5].x___;
>  4: MAD temp[0].y, temp[5]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[6].xy, temp[0].xy__, const[12].xy__;
>  6: TEX temp[7].xyz, temp[6].xy__, 2D[1];
>  7: MOV temp[1].xyz, temp[7].xyz_;
>  8: DP3 temp[1].x, temp[1].xyz_, const[8].xyz_;
>  9: MUL temp[1].xyz, input[1].xyz_, temp[1].xxx_;
> 10: ADD temp[1].xyz, const[6].xyz_, -temp[1].xyz_;
> 11: MUL temp[1].xyz, temp[1].xyz_, const[5].xxx_;
> 12: MUL temp[8].xy, temp[0].xy__, const[13].xy__;
> 13: TEX temp[9].xyz, temp[8].xy__, 2D[3];
> 14: MOV temp[2].xyz, temp[9].xyz_;
> 15: MUL temp[2].xyz, temp[2].xyz_, const[4].xyz_;
> 16: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 17: RSQ temp[3].x, |temp[3].x___|;
> 18: MUL temp[3].xyz, temp[1].xyz_, temp[3].xxx_;
> 19: MUL temp[10].xy, temp[0].xy__, const[14].xy__;
> 20: TEX temp[11].xyz, temp[10].xy__, 2D[2];
> 21: MOV temp[4].xyz, temp[11].xyz_;
> 22: MAD temp[4].xyz, temp[4].xyz_, const[8].www_, const[9].xxx_;
> 23: DP3 temp[3].x, temp[3].xyz_, temp[4].xyz_;
> 24: MAX temp[3].x, temp[3].x___, const[9].y___;
> 25: MUL temp[2].xyz, temp[2].xyz_, temp[3].xxx_;
> 26: DP3 temp[1].x, temp[1].xyz_, temp[1].xyz_;
> 27: TEX temp[12].x, temp[1].x___, 1D[0];
> 28: MOV temp[1].x, temp[12].x___;
> 29: MUL temp[1].xyz, temp[2].xyz_, temp[1].xxx_;
> 30: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[17].xy, temp[16].xy__, const[12].xy__;
>  6: TEX temp[18].xyz, temp[17].xy__, 2D[1];
>  7: MOV temp[19].xyz, temp[18].xyz_;
>  8: DP3 temp[20].x, temp[19].xyz_, const[8].xyz_;
>  9: MUL temp[21].xyz, input[1].xyz_, temp[20].xxx_;
> 10: ADD temp[22].xyz, const[6].xyz_, -temp[21].xyz_;
> 11: MUL temp[23].xyz, temp[22].xyz_, const[5].xxx_;
> 12: MUL temp[24].xy, temp[16].xy__, const[13].xy__;
> 13: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 14: MOV temp[26].xyz, temp[25].xyz_;
> 15: MUL temp[27].xyz, temp[26].xyz_, const[4].xyz_;
> 16: DP3 temp[28].x, temp[23].xyz_, temp[23].xyz_;
> 17: RSQ temp[29].x, |temp[28].x___|;
> 18: MUL temp[30].xyz, temp[23].xyz_, temp[29].xxx_;
> 19: MUL temp[31].xy, temp[16].xy__, const[14].xy__;
> 20: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 21: MOV temp[33].xyz, temp[32].xyz_;
> 22: MAD temp[34].xyz, temp[33].xyz_, const[8].www_, const[9].xxx_;
> 23: DP3 temp[35].x, temp[30].xyz_, temp[34].xyz_;
> 24: MAX temp[36].x, temp[35].x___, const[9].y___;
> 25: MUL temp[37].xyz, temp[27].xyz_, temp[36].xxx_;
> 26: DP3 temp[38].x, temp[23].xyz_, temp[23].xyz_;
> 27: TEX temp[39].x, temp[38].x___, 1D[0];
> 28: MOV temp[40].x, temp[39].x___;
> 29: MUL temp[41].xyz, temp[37].xyz_, temp[40].xxx_;
> 30: MOV output[0], temp[41];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[17].xy, temp[16].xy__, const[12].xy__;
>  6: TEX temp[18].xyz, temp[17].xy__, 2D[1];
>  7: DP3 temp[20].x, temp[18].xyz_, const[8].xyz_;
>  8: MUL temp[21].xyz, input[1].xyz_, temp[20].xxx_;
>  9: MUL temp[23].xyz, (const[6] - temp[21]).xyz_, const[5].xxx_;
> 10: MUL temp[24].xy, temp[16].xy__, const[13].xy__;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 12: MUL temp[27].xyz, temp[25].xyz_, const[4].xyz_;
> 13: DP3 temp[28].x, temp[23].xyz_, temp[23].xyz_;
> 14: RSQ temp[29].x, |temp[28].x___|;
> 15: MUL temp[30].xyz, temp[23].xyz_, temp[29].xxx_;
> 16: MUL temp[31].xy, temp[16].xy__, const[14].xy__;
> 17: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 18: MAD temp[34].xyz, temp[32].xyz_, const[8].www_, none.-1-1-1_;
> 19: DP3 temp[35].x, temp[30].xyz_, temp[34].xyz_;
> 20: MAX temp[36].x, temp[35].x___, none.0___;
> 21: MUL temp[37].xyz, temp[27].xyz_, temp[36].xxx_;
> 22: DP3 temp[38].x, temp[23].xyz_, temp[23].xyz_;
> 23: TEX temp[39].x, temp[38].x___, 1D[0];
> 24: MUL temp[41].xyz, temp[37].xyz_, temp[39].xxx_;
> 25: MOV output[0], temp[41];
>CONST[8] = {     1.0000     0.0039     0.0000     2.0000 }
>CONST[9] = {    -1.0000     0.0000     0.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[10].xy__, const[11].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[7]._x__, const[7]._y__;
>  5: MUL temp[17].xy, temp[16].xy__, const[12].xy__;
>  6: TEX temp[18].xyz, temp[17].xy__, 2D[1];
>  7: DP3 temp[20].x, temp[18].xyz_, const[8].xyz_;
>  8: MUL temp[21].xyz, input[1].xyz_, temp[20].xxx_;
>  9: MUL temp[23].xyz, (const[6] - temp[21]).xyz_, const[5].xxx_;
> 10: MUL temp[24].xy, temp[16].xy__, const[13].xy__;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 12: MUL temp[27].xyz, temp[25].xyz_, const[4].xyz_;
> 13: DP3 temp[28].x, temp[23].xyz_, temp[23].xyz_;
> 14: RSQ temp[29].x, |temp[28].x___|;
> 15: MUL temp[30].xyz, temp[23].xyz_, temp[29].xxx_;
> 16: MUL temp[31].xy, temp[16].xy__, const[14].xy__;
> 17: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 18: MAD temp[34].xyz, temp[32].xyz_, const[8].www_, none.-1-1-1_;
> 19: DP3 temp[35].x, temp[30].xyz_, temp[34].xyz_;
> 20: MAX temp[36].x, temp[35].x___, none.0___;
> 21: MUL temp[37].xyz, temp[27].xyz_, temp[36].xxx_;
> 22: DP3 temp[38].x, temp[23].xyz_, temp[23].xyz_;
> 23: TEX temp[39].x, temp[38].x___, 1D[0];
> 24: MUL temp[41].xyz, temp[37].xyz_, temp[39].xxx_;
> 25: MOV output[0], temp[41];
>CONST[4] = {     1.0000     0.0039     0.0000     2.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[5].xy__, const[6].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[3]._x__, const[3]._y__;
>  5: MUL temp[17].xy, temp[16].xy__, const[7].xy__;
>  6: TEX temp[18].xyz, temp[17].xy__, 2D[1];
>  7: DP3 temp[20].x, temp[18].xyz_, const[4].xyz_;
>  8: MUL temp[21].xyz, input[1].xyz_, temp[20].xxx_;
>  9: MUL temp[23].xyz, (const[2] - temp[21]).xyz_, const[1].xxx_;
> 10: MUL temp[24].xy, temp[16].xy__, const[8].xy__;
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 12: MUL temp[27].xyz, temp[25].xyz_, const[0].xyz_;
> 13: DP3 temp[28].x, temp[23].xyz_, temp[23].xyz_;
> 14: RSQ temp[29].x, |temp[28].x___|;
> 15: MUL temp[30].xyz, temp[23].xyz_, temp[29].xxx_;
> 16: MUL temp[31].xy, temp[16].xy__, const[9].xy__;
> 17: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 18: MAD temp[34].xyz, temp[32].xyz_, const[4].www_, none.-1-1-1_;
> 19: DP3 temp[35].x, temp[30].xyz_, temp[34].xyz_;
> 20: MAX temp[36].x, temp[35].x___, none.0___;
> 21: MUL temp[37].xyz, temp[27].xyz_, temp[36].xxx_;
> 22: DP3 temp[38].x, temp[23].xyz_, temp[23].xyz_;
> 23: TEX temp[39].x, temp[38].x___, 1D[0];
> 24: MUL temp[41].xyz, temp[37].xyz_, temp[39].xxx_;
> 25: MOV output[0], temp[41];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[5], src2.xyz = const[6]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[3]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: src0.xyz = temp[16], src1.xyz = const[7]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src0.000
>  6: TEX temp[18].xyz, temp[17].xy__, 2D[1];
>  7: src0.xyz = temp[18], src1.xyz = const[4]
>     DP3 temp[20].x, src0.xyz, src1.xyz
>  8: src0.xyz = input[1], src1.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
>  9: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[23].xyz, srcp.xyz, src2.xxx, src0.000
> 10: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
> 11: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 12: src0.xyz = temp[25], src1.xyz = const[0]
>     MAD temp[27].xyz, src0.xyz, src1.xyz, src0.000
> 13: src0.xyz = temp[23]
>     DP3 temp[28].x, src0.xyz, src0.xyz
> 14: src0.xyz = temp[28]
>     REPL_ALPHA temp[29].x
>     RSQ, |src0.x|
> 15: src0.xyz = temp[23], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
> 17: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 18: src0.xyz = temp[32], src0.w = const[4]
>     MAD temp[34].xyz, src0.xyz, src0.www, -src0.111
> 19: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[35].x, src0.xyz, src1.xyz
> 20: src0.xyz = temp[35]
>     MAX temp[36].x, src0.x__, src0.0__
> 21: src0.xyz = temp[27], src1.xyz = temp[36]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
> 22: src0.xyz = temp[23]
>     DP3 temp[38].x, src0.xyz, src0.xyz
> 23: TEX temp[39].x, temp[38].x___, 1D[0];
> 24: src0.xyz = temp[37], src1.xyz = temp[39]
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 25: src0.xyz = temp[41], src0.w = temp[41]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[5], src2.xyz = const[6]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15], src1.xyz = const[3]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[16], src1.xyz = const[7]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 11: TEX temp[18].xyz, temp[17].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[32], src0.w = const[4] SEM_WAIT
>     MAD temp[34].xyz, src0.xyz, src0.www, -src0.111
> 13: src0.xyz = temp[25], src1.xyz = const[0]
>     MAD temp[27].xyz, src0.xyz, src1.xyz, src0.000
> 14: src0.xyz = temp[18], src1.xyz = const[4]
>     DP3 temp[20].x, src0.xyz, src1.xyz
> 15: src0.xyz = input[1], src1.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[23].xyz, srcp.xyz, src2.xxx, src0.000
> 17: src0.xyz = temp[23]
>     DP3 temp[38].x, src0.xyz, src0.xyz
> 18: src0.xyz = temp[23]
>     DP3 temp[28].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[28]
>     REPL_ALPHA temp[29].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[23], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[35].x, src0.xyz, src1.xyz
> 22: src0.xyz = temp[35]
>     MAX temp[36].x, src0.x__, src0.0__
> 23: src0.xyz = temp[27], src1.xyz = temp[36]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
> 24: BEGIN_TEX;
> 25: TEX temp[39].x, temp[38].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 26: src0.xyz = temp[37], src1.xyz = temp[39] SEM_WAIT
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = temp[41], src0.w = temp[41]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[5], src2.xyz = const[6]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15], src1.xyz = const[3]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  5: src0.xyz = temp[16], src1.xyz = const[9]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = temp[16], src1.xyz = const[8]
>     MAD temp[24].xy, src0.xy_, src1.xy_, src0.000
>  7: src0.xyz = temp[16], src1.xyz = const[7]
>     MAD temp[17].xy, src0.xy_, src1.xy_, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[32].xyz, temp[31].xy__, 2D[2];
> 10: TEX temp[25].xyz, temp[24].xy__, 2D[3];
> 11: TEX temp[18].xyz, temp[17].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[32], src0.w = const[4] SEM_WAIT
>     MAD temp[34].xyz, src0.xyz, src0.www, -src0.111
> 13: src0.xyz = temp[25], src1.xyz = const[0]
>     MAD temp[27].xyz, src0.xyz, src1.xyz, src0.000
> 14: src0.xyz = temp[18], src1.xyz = const[4]
>     DP3 temp[20].x, src0.xyz, src1.xyz
> 15: src0.xyz = input[1], src1.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[21], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[23].xyz, srcp.xyz, src2.xxx, src0.000
> 17: src0.xyz = temp[23]
>     DP3 temp[38].x, src0.xyz, src0.xyz
> 18: src0.xyz = temp[23]
>     DP3 temp[28].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[28]
>     REPL_ALPHA temp[29].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[23], src1.xyz = temp[29]
>     MAD temp[30].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[30], src1.xyz = temp[34]
>     DP3 temp[35].x, src0.xyz, src1.xyz
> 22: src0.xyz = temp[35]
>     MAX temp[36].x, src0.x__, src0.0__
> 23: src0.xyz = temp[27], src1.xyz = temp[36]
>     MAD temp[37].xyz, src0.xyz, src1.xxx, src0.000
> 24: BEGIN_TEX;
> 25: TEX temp[39].x, temp[38].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 26: src0.xyz = temp[37], src1.xyz = temp[39] SEM_WAIT
>     MAD temp[41].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = temp[41], src0.w = temp[41]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.w = input[1]
>     RCP temp[0].w, src0.w
>  1: src0.xyz = input[1], src0.w = temp[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.00_
>  2: src0.xyz = temp[1], src1.xyz = const[5], src2.xyz = const[6]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[1], src1.xyz = const[3]
>     MAD temp[2].y, src0._y_, src1._x_, src1._y_
>  4: src0.xyz = temp[1]
>     MAD temp[2].x, src0.x__, src0.11_, src0.00_
>  5: src0.xyz = temp[2], src1.xyz = const[9]
>     MAD temp[1].xy, src0.xy_, src1.xy_, src0.00_
>  6: src0.xyz = temp[2], src1.xyz = const[8]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  7: src0.xyz = temp[2], src1.xyz = const[7]
>     MAD temp[2].xy, src0.xy_, src1.xy_, src0.00_
>  8: BEGIN_TEX;
>  9: TEX temp[1].xyz, temp[1].xy__, 2D[2];
> 10: TEX temp[3].xyz, temp[3].xy__, 2D[3];
> 11: TEX temp[2].xyz, temp[2].xy__, 2D[1] SEM_WAIT SEM_ACQUIRE;
> 12: src0.xyz = temp[1], src0.w = const[4] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src0.www, -src0.111
> 13: src0.xyz = temp[3], src1.xyz = const[0]
>     MAD temp[3].xyz, src0.xyz, src1.xyz, src0.000
> 14: src0.xyz = temp[2], src1.xyz = const[4]
>     DP3 temp[2].x, src0.xyz, src1.xyz
> 15: src0.xyz = input[0], src1.xyz = temp[2]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 16: src0.xyz = temp[0], src1.xyz = const[2], src2.xyz = const[1], srcp.xyz = (src1 - src0)
>     MAD temp[0].xyz, srcp.xyz, src2.xxx, src0.000
> 17: src0.xyz = temp[0]
>     DP3 temp[2].x, src0.xyz, src0.xyz
> 18: src0.xyz = temp[0]
>     DP3 temp[2].y, src0.xyz, src0.xyz
> 19: src0.xyz = temp[2]
>     REPL_ALPHA temp[2].y
>     RSQ, |src0.y|
> 20: src0.xyz = temp[0], src1.xyz = temp[2]
>     MAD temp[0].xyz, src0.xyz, src1.yyy, src0.000
> 21: src0.xyz = temp[0], src1.xyz = temp[1]
>     DP3 temp[0].x, src0.xyz, src1.xyz
> 22: src0.xyz = temp[0]
>     MAX temp[0].x, src0.x__, src0.0__
> 23: src0.xyz = temp[3], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 24: BEGIN_TEX;
> 25: TEX temp[1].x, temp[2].x___, 1D[0] SEM_WAIT SEM_ACQUIRE;
> 26: src0.xyz = temp[0], src1.xyz = temp[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 27: src0.xyz = temp[0], src0.w = temp[41]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>r300compiler error: Too many hardware temporaries used.
>r300 FP: Compiler Error:
>Too many hardware temporaries used.
>Using a dummy shader instead.
>r300: Initial fragment program
>FRAG
>DCL OUT[0], COLOR
>IMM[0] FLT32 {    0.0000,     1.0000,     0.0000,     0.0000}
>  0: MOV OUT[0], IMM[0].xxxy
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: MOV output[0], temp[0].0001;
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MOV output[0], none.0001;
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: 
>     MAD color[0].xyz, src0.000, src0.111, src0.000
>     MAD color[0].w, src0.1, src0.1, src0.0
>pc=45*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:      0.0      1.0      0.0    op: 00050a94 
>       w:      1.0      1.0      0.0    op: 00040891
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], LINEAR
>DCL OUT[0], COLOR
>DCL SAMP[0]
>  0: TEX OUT[0], IN[0], SAMP[0], 2D
>  1: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX output[0], input[0], 2D[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX output[0], input[0], 2D[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX output[0], input[0], 2D[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX output[0], input[0], 2D[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX output[0], input[0], 2D[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[1], input[0], 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[1], input[0], 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=46*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>IMM[0] FLT32 {   -4.3500,     0.0000,     0.1000,    -2.5000}
>IMM[1] FLT32 {    0.2500,    -0.7500,     0.0000,     0.3000}
>IMM[2] FLT32 {    0.5000,     0.7500,     0.0000,     2.5000}
>IMM[3] FLT32 {    4.3500,     0.0000,     0.5556,     1.0000}
>  0: ADD TEMP[0].xy, IN[0].xyyy, IMM[0].xyyy
>  1: TEX TEMP[0].xyz, TEMP[0].xyyy, SAMP[0], RECT
>  2: MUL TEMP[0].xyz, TEMP[0].xyzz, IMM[0].zzzz
>  3: ADD TEMP[1].xy, IN[0].xyyy, IMM[0].wyyy
>  4: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  5: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].xxxx, TEMP[0].xyzz
>  6: ADD TEMP[1].xy, IN[0].xyyy, IMM[1].yzzz
>  7: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  8: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].wwww, TEMP[0].xyzz
>  9: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[0], RECT
> 10: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[2].xxxx, TEMP[0].xyzz
> 11: ADD TEMP[1].xy, IN[0].xyyy, IMM[2].yzzz
> 12: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 13: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].wwww, TEMP[0].xyzz
> 14: ADD TEMP[1].xy, IN[0].xyyy, IMM[2].wzzz
> 15: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 16: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].xxxx, TEMP[0].xyzz
> 17: ADD TEMP[1].xy, IN[0].xyyy, IMM[3].xyyy
> 18: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 19: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[0].zzzz, TEMP[0].xyzz
> 20: MUL TEMP[0].xyz, TEMP[0].xyzz, IMM[3].zzzz
> 21: MOV TEMP[0].xyz, TEMP[0].xyzx
> 22: MOV TEMP[0].w, IMM[3].wwww
> 23: MOV OUT[0], TEMP[0]
> 24: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: MUL temp[2], temp[0].xyyy, const[4];
>  2: TEX temp[3], temp[2], 2D[0];
>  3: MOV temp[0].xyz, temp[3];
>  4: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  5: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  6: MUL temp[4], temp[1].xyyy, const[4];
>  7: TEX temp[5], temp[4], 2D[0];
>  8: MOV temp[1].xyz, temp[5];
>  9: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 10: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
> 11: MUL temp[6], temp[1].xyyy, const[4];
> 12: TEX temp[7], temp[6], 2D[0];
> 13: MOV temp[1].xyz, temp[7];
> 14: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 15: MUL temp[8], input[0].xyyy, const[4];
> 16: TEX temp[9], temp[8], 2D[0];
> 17: MOV temp[1].xyz, temp[9];
> 18: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 19: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 20: MUL temp[10], temp[1].xyyy, const[4];
> 21: TEX temp[11], temp[10], 2D[0];
> 22: MOV temp[1].xyz, temp[11];
> 23: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 24: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 25: MUL temp[12], temp[1].xyyy, const[4];
> 26: TEX temp[13], temp[12], 2D[0];
> 27: MOV temp[1].xyz, temp[13];
> 28: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 29: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 30: MUL temp[14], temp[1].xyyy, const[4];
> 31: TEX temp[15], temp[14], 2D[0];
> 32: MOV temp[1].xyz, temp[15];
> 33: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 34: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 35: MOV temp[0].xyz, temp[0].xyzx;
> 36: MOV temp[0].w, const[3].wwww;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: MUL temp[2], temp[0].xyyy, const[4];
>  2: TEX temp[3], temp[2], 2D[0];
>  3: MOV temp[0].xyz, temp[3];
>  4: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  5: ADD temp[1].xy, input[0].xyyy, const[0].wyyy;
>  6: MUL temp[4], temp[1].xyyy, const[4];
>  7: TEX temp[5], temp[4], 2D[0];
>  8: MOV temp[1].xyz, temp[5];
>  9: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 10: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
> 11: MUL temp[6], temp[1].xyyy, const[4];
> 12: TEX temp[7], temp[6], 2D[0];
> 13: MOV temp[1].xyz, temp[7];
> 14: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 15: MUL temp[8], input[0].xyyy, const[4];
> 16: TEX temp[9], temp[8], 2D[0];
> 17: MOV temp[1].xyz, temp[9];
> 18: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 19: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 20: MUL temp[10], temp[1].xyyy, const[4];
> 21: TEX temp[11], temp[10], 2D[0];
> 22: MOV temp[1].xyz, temp[11];
> 23: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 24: ADD temp[1].xy, input[0].xyyy, const[2].wzzz;
> 25: MUL temp[12], temp[1].xyyy, const[4];
> 26: TEX temp[13], temp[12], 2D[0];
> 27: MOV temp[1].xyz, temp[13];
> 28: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 29: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 30: MUL temp[14], temp[1].xyyy, const[4];
> 31: TEX temp[15], temp[14], 2D[0];
> 32: MOV temp[1].xyz, temp[15];
> 33: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 34: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 35: MOV temp[0].xyz, temp[0].xyzx;
> 36: MOV temp[0].w, const[3].wwww;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[2].xy, temp[0].xy__, const[4].xy__;
>  2: TEX temp[3].xyz, temp[2].xy__, 2D[0];
>  3: MOV temp[0].xyz, temp[3].xyz_;
>  4: MUL temp[0].xyz, temp[0].xyz_, const[0].zzz_;
>  5: ADD temp[1].xy, input[0].xy__, const[0].wy__;
>  6: MUL temp[4].xy, temp[1].xy__, const[4].xy__;
>  7: TEX temp[5].xyz, temp[4].xy__, 2D[0];
>  8: MOV temp[1].xyz, temp[5].xyz_;
>  9: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 10: ADD temp[1].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[6].xy, temp[1].xy__, const[4].xy__;
> 12: TEX temp[7].xyz, temp[6].xy__, 2D[0];
> 13: MOV temp[1].xyz, temp[7].xyz_;
> 14: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 15: MUL temp[8].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[0];
> 17: MOV temp[1].xyz, temp[9].xyz_;
> 18: MAD temp[0].xyz, temp[1].xyz_, const[2].xxx_, temp[0].xyz_;
> 19: ADD temp[1].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[10].xy, temp[1].xy__, const[4].xy__;
> 21: TEX temp[11].xyz, temp[10].xy__, 2D[0];
> 22: MOV temp[1].xyz, temp[11].xyz_;
> 23: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 24: ADD temp[1].xy, input[0].xy__, const[2].wz__;
> 25: MUL temp[12].xy, temp[1].xy__, const[4].xy__;
> 26: TEX temp[13].xyz, temp[12].xy__, 2D[0];
> 27: MOV temp[1].xyz, temp[13].xyz_;
> 28: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 29: ADD temp[1].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[14].xy, temp[1].xy__, const[4].xy__;
> 31: TEX temp[15].xyz, temp[14].xy__, 2D[0];
> 32: MOV temp[1].xyz, temp[15].xyz_;
> 33: MAD temp[0].xyz, temp[1].xyz_, const[0].zzz_, temp[0].xyz_;
> 34: MUL temp[0].xyz, temp[0].xyz_, const[3].zzz_;
> 35: MOV temp[0].xyz, temp[0].xyz_;
> 36: MOV temp[0].w, const[3].___w;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[2].xy, temp[0].xy__, const[4].xy__;
>  2: TEX temp[3].xyz, temp[2].xy__, 2D[0];
>  3: MOV temp[0].xyz, temp[3].xyz_;
>  4: MUL temp[0].xyz, temp[0].xyz_, const[0].zzz_;
>  5: ADD temp[1].xy, input[0].xy__, const[0].wy__;
>  6: MUL temp[4].xy, temp[1].xy__, const[4].xy__;
>  7: TEX temp[5].xyz, temp[4].xy__, 2D[0];
>  8: MOV temp[1].xyz, temp[5].xyz_;
>  9: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 10: ADD temp[1].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[6].xy, temp[1].xy__, const[4].xy__;
> 12: TEX temp[7].xyz, temp[6].xy__, 2D[0];
> 13: MOV temp[1].xyz, temp[7].xyz_;
> 14: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 15: MUL temp[8].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[0];
> 17: MOV temp[1].xyz, temp[9].xyz_;
> 18: MAD temp[0].xyz, temp[1].xyz_, const[2].xxx_, temp[0].xyz_;
> 19: ADD temp[1].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[10].xy, temp[1].xy__, const[4].xy__;
> 21: TEX temp[11].xyz, temp[10].xy__, 2D[0];
> 22: MOV temp[1].xyz, temp[11].xyz_;
> 23: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 24: ADD temp[1].xy, input[0].xy__, const[2].wz__;
> 25: MUL temp[12].xy, temp[1].xy__, const[4].xy__;
> 26: TEX temp[13].xyz, temp[12].xy__, 2D[0];
> 27: MOV temp[1].xyz, temp[13].xyz_;
> 28: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 29: ADD temp[1].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[14].xy, temp[1].xy__, const[4].xy__;
> 31: TEX temp[15].xyz, temp[14].xy__, 2D[0];
> 32: MOV temp[1].xyz, temp[15].xyz_;
> 33: MAD temp[0].xyz, temp[1].xyz_, const[0].zzz_, temp[0].xyz_;
> 34: MUL temp[0].xyz, temp[0].xyz_, const[3].zzz_;
> 35: MOV temp[0].xyz, temp[0].xyz_;
> 36: MOV temp[0].w, const[3].___w;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: ADD temp[16].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[17].xy, temp[16].xy__, const[4].xy__;
>  2: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  3: MOV temp[19].xyz, temp[18].xyz_;
>  4: MUL temp[20].xyz, temp[19].xyz_, const[0].zzz_;
>  5: ADD temp[21].xy, input[0].xy__, const[0].wy__;
>  6: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  7: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  8: MOV temp[24].xyz, temp[23].xyz_;
>  9: MAD temp[25].xyz, temp[24].xyz_, const[1].xxx_, temp[20].xyz_;
> 10: ADD temp[26].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
> 12: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 13: MOV temp[29].xyz, temp[28].xyz_;
> 14: MAD temp[30].xyz, temp[29].xyz_, const[1].www_, temp[25].xyz_;
> 15: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 17: MOV temp[33].xyz, temp[32].xyz_;
> 18: MAD temp[34].xyz, temp[33].xyz_, const[2].xxx_, temp[30].xyz_;
> 19: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 21: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 22: MOV temp[38].xyz, temp[37].xyz_;
> 23: MAD temp[39].xyz, temp[38].xyz_, const[1].www_, temp[34].xyz_;
> 24: ADD temp[40].xy, input[0].xy__, const[2].wz__;
> 25: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 27: MOV temp[43].xyz, temp[42].xyz_;
> 28: MAD temp[44].xyz, temp[43].xyz_, const[1].xxx_, temp[39].xyz_;
> 29: ADD temp[45].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[46].xy, temp[45].xy__, const[4].xy__;
> 31: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 32: MOV temp[48].xyz, temp[47].xyz_;
> 33: MAD temp[49].xyz, temp[48].xyz_, const[0].zzz_, temp[44].xyz_;
> 34: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 35: MOV temp[51].xyz, temp[50].xyz_;
> 36: MOV temp[51].w, const[3].___w;
> 37: MOV output[0], temp[51];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[0].w0__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[2].wz__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>CONST[0] = {    -4.3500     0.0000     0.1000    -2.5000 }
>CONST[1] = {     0.2500    -0.7500     0.0000     0.3000 }
>CONST[2] = {     0.5000     0.7500     0.0000     2.5000 }
>CONST[3] = {     4.3500     0.0000     0.5556     1.0000 }
>CONST[5] = {     0.0000     0.0000     0.0000    -2.5000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[5].wz__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[2].wz__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>CONST[0] = {    -4.3500     0.0000     0.1000    -2.5000 }
>CONST[1] = {     0.2500    -0.7500     0.0000     0.3000 }
>CONST[2] = {     0.5000     0.7500     0.0000     2.5000 }
>CONST[3] = {     4.3500     0.0000     0.5556     1.0000 }
>CONST[5] = {     0.0000     0.0000     0.0000    -2.5000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[5].wz__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[2].wz__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: src0.xyz = temp[18], src1.xyz = const[0]
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
>  3: src0.xyz = input[0], src1.xyz = const[5], src1.w = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.wz_
>  4: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
>  7: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  8: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 11: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 14: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
> 15: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 18: src0.xyz = input[0], src1.xyz = const[2], src1.w = const[2]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.wz_
> 19: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 22: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 25: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 26: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 27: 
>     MAD temp[51].w, src0.1, src0.1, src0.0
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>     MAD temp[51].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
>  2: src0.xyz = input[0], src1.xyz = const[5], src1.w = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.wz_
>  3: src0.xyz = input[0], src1.xyz = const[2], src1.w = const[2]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.wz_
>  4: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  5: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
>  7: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
> 10: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 11: BEGIN_TEX;
> 12: TEX temp[18].xyz, temp[17].xy__, 2D[0];
> 13: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 14: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 15: TEX temp[23].xyz, temp[22].xy__, 2D[0];
> 16: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 17: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 18: TEX temp[37].xyz, temp[36].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[18], src1.xyz = const[0] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>     MAD temp[51].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
>  2: src0.xyz = input[0], src1.xyz = const[5], src1.w = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.wz_
>  3: src0.xyz = input[0], src1.xyz = const[2], src1.w = const[2]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.wz_
>  4: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  5: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  6: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
>  7: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
> 10: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 11: BEGIN_TEX;
> 12: TEX temp[18].xyz, temp[17].xy__, 2D[0];
> 13: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 14: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 15: TEX temp[23].xyz, temp[22].xy__, 2D[0];
> 16: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 17: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 18: TEX temp[37].xyz, temp[36].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[18], src1.xyz = const[0] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[1].xy, srcp.xy_, src2.xy_, src0.00_
>     MAD temp[0].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[2].xy, srcp.xy_, src2.xy_, src0.00_
>  2: src0.xyz = input[0], src1.xyz = const[5], src1.w = const[5]
>     MAD temp[3].xy, src0.xy_, src0.11_, src1.wz_
>  3: src0.xyz = input[0], src1.xyz = const[2], src1.w = const[2]
>     MAD temp[4].xy, src0.xy_, src0.11_, src1.wz_
>  4: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[5].xy, src0.xy_, src0.11_, src1.yz_
>  5: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[6].xy, src0.xy_, src1.xy_, src0.00_
>  6: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[0].xy, src0.xy_, src0.11_, src1.yz_
>  7: src0.xyz = temp[3], src1.xyz = const[4]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[4], src1.xyz = const[4]
>     MAD temp[4].xy, src0.xy_, src1.xy_, src0.00_
>  9: src0.xyz = temp[5], src1.xyz = const[4]
>     MAD temp[5].xy, src0.xy_, src1.xy_, src0.00_
> 10: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
> 11: BEGIN_TEX;
> 12: TEX temp[1].xyz, temp[1].xy__, 2D[0];
> 13: TEX temp[2].xyz, temp[2].xy__, 2D[0];
> 14: TEX temp[6].xyz, temp[6].xy__, 2D[0];
> 15: TEX temp[3].xyz, temp[3].xy__, 2D[0];
> 16: TEX temp[4].xyz, temp[4].xy__, 2D[0];
> 17: TEX temp[5].xyz, temp[5].xy__, 2D[0];
> 18: TEX temp[7].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[3], src1.xyz = const[1], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[5], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[6], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[7], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[4], src1.xyz = const[1], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[2], src1.xyz = const[0], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[0], src1.xyz = const[3]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=47*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 10, tex_end: 0  (code_addr: 00000280)
>  0: xyz:  t0  c0  c4   add-> t1.xy                (01864800)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz: srcp.xyz   c4.xyz      0.0    op: 0045040f 
>       w:      1.0      1.0      0.0    op: 00040891
>  1: xyz:  t0  c3  c4   add-> t2.xy                (018a48c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz   c4.xyz      0.0    op: 0045040f 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  2: xyz:  t0  c5  t0  bias-> t3.xy                (018c0940)
>       w:  t0  c5  t0  bias->                      (00000940)
>     xyz:   t0.xyz      1.0   c5.Wzy    op: 00078a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  c2  t0  bias-> t4.xy                (01900880)
>       w:  t0  c2  t0  bias->                      (00000880)
>     xyz:   t0.xyz      1.0   c2.Wzy    op: 00078a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t0  c1  t0  bias-> t5.xy                (01940840)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c1.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  5: xyz:  t0  c4  t0  bias-> t6.xy                (01980900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  6: xyz:  t0  c2  t0  bias-> t0.xy                (01800880)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c2.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t3  c4  t0  bias-> t3.xy                (018c0903)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  8: xyz:  t4  c4  t0  bias-> t4.xy                (01900904)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t4.x     t4.x     t4.x    op: 00000000
>  9: xyz:  t5  c4  t0  bias-> t5.xy                (01940905)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 10: xyz:  t0  c4  t0  bias-> t0.xy                (01800900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 11, tex_offset: 0, alu_end: 9, tex_end: 6  (code_addr: 004c024b)
>  TEX:
>    TEX t1, t1, texture[0]   (00008041)
>    TEX t2, t2, texture[0]   (00008082)
>    TEX t6, t6, texture[0]   (00008186)
>    TEX t3, t3, texture[0]   (000080c3)
>    TEX t4, t4, texture[0]   (00008104)
>    TEX t5, t5, texture[0]   (00008145)
>    TEX t7, t0, texture[0]   (000081c0)
> 11: xyz:  t1  c0  t0  bias-> t0.xyz               (03800801)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 12: xyz:  t3  c1  t0  bias-> t0.xyz               (03800843)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c1.xxx   t0.xyz    op: 00020280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 13: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t5.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 14: xyz:  t6  t0  t0  bias-> t0.xyz               (03800006)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz      0.5   t0.xyz    op: 00010b00 
>       w:     t6.x     t6.x     t6.x    op: 00000000
> 15: xyz:  t7  t0  t0  bias-> t0.xyz               (03800007)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t7.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 16: xyz:  t4  c1  t0  bias-> t0.xyz               (03800844)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c1.xxx   t0.xyz    op: 00020280 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 17: xyz:  t2  c0  t0  bias-> t0.xyz               (03800802)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c0.zzz   t0.xyz    op: 00020380 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 18: xyz:  t0  c3  t0  bias-> t0.xyz               (038008c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c3.zzz      0.0    op: 00050380 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 19: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 20: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>IMM[0] FLT32 {    0.0000,    -4.3500,     0.1000,    -2.5000}
>IMM[1] FLT32 {    0.2500,     0.0000,    -0.7500,     0.3000}
>IMM[2] FLT32 {    0.5000,     0.0000,     0.7500,     2.5000}
>IMM[3] FLT32 {    0.0000,     4.3500,     0.5556,     1.0000}
>  0: ADD TEMP[0].xy, IN[0].xyyy, IMM[0].xyyy
>  1: TEX TEMP[0].xyz, TEMP[0].xyyy, SAMP[0], RECT
>  2: MUL TEMP[0].xyz, TEMP[0].xyzz, IMM[0].zzzz
>  3: ADD TEMP[1].xy, IN[0].xyyy, IMM[0].xwww
>  4: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  5: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].xxxx, TEMP[0].xyzz
>  6: ADD TEMP[1].xy, IN[0].xyyy, IMM[1].yzzz
>  7: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
>  8: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].wwww, TEMP[0].xyzz
>  9: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[0], RECT
> 10: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[2].xxxx, TEMP[0].xyzz
> 11: ADD TEMP[1].xy, IN[0].xyyy, IMM[2].yzzz
> 12: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 13: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].wwww, TEMP[0].xyzz
> 14: ADD TEMP[1].xy, IN[0].xyyy, IMM[2].ywww
> 15: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 16: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[1].xxxx, TEMP[0].xyzz
> 17: ADD TEMP[1].xy, IN[0].xyyy, IMM[3].xyyy
> 18: TEX TEMP[1].xyz, TEMP[1].xyyy, SAMP[0], RECT
> 19: MAD TEMP[0].xyz, TEMP[1].xyzz, IMM[0].zzzz, TEMP[0].xyzz
> 20: MUL TEMP[0].xyz, TEMP[0].xyzz, IMM[3].zzzz
> 21: MOV TEMP[0].xyz, TEMP[0].xyzx
> 22: MOV TEMP[0].w, IMM[3].wwww
> 23: MOV OUT[0], TEMP[0]
> 24: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: TEX temp[0].xyz, temp[0].xyyy, RECT[0];
>  2: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  3: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  4: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  5: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
>  6: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
>  7: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
>  8: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
>  9: TEX temp[1].xyz, input[0].xyyy, RECT[0];
> 10: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 11: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 12: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 13: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 14: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 15: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 16: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 17: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 18: TEX temp[1].xyz, temp[1].xyyy, RECT[0];
> 19: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 20: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 21: MOV temp[0].xyz, temp[0].xyzx;
> 22: MOV temp[0].w, const[3].wwww;
> 23: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: MUL temp[2], temp[0].xyyy, const[4];
>  2: TEX temp[3], temp[2], 2D[0];
>  3: MOV temp[0].xyz, temp[3];
>  4: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  5: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  6: MUL temp[4], temp[1].xyyy, const[4];
>  7: TEX temp[5], temp[4], 2D[0];
>  8: MOV temp[1].xyz, temp[5];
>  9: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 10: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
> 11: MUL temp[6], temp[1].xyyy, const[4];
> 12: TEX temp[7], temp[6], 2D[0];
> 13: MOV temp[1].xyz, temp[7];
> 14: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 15: MUL temp[8], input[0].xyyy, const[4];
> 16: TEX temp[9], temp[8], 2D[0];
> 17: MOV temp[1].xyz, temp[9];
> 18: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 19: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 20: MUL temp[10], temp[1].xyyy, const[4];
> 21: TEX temp[11], temp[10], 2D[0];
> 22: MOV temp[1].xyz, temp[11];
> 23: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 24: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 25: MUL temp[12], temp[1].xyyy, const[4];
> 26: TEX temp[13], temp[12], 2D[0];
> 27: MOV temp[1].xyz, temp[13];
> 28: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 29: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 30: MUL temp[14], temp[1].xyyy, const[4];
> 31: TEX temp[15], temp[14], 2D[0];
> 32: MOV temp[1].xyz, temp[15];
> 33: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 34: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 35: MOV temp[0].xyz, temp[0].xyzx;
> 36: MOV temp[0].w, const[3].wwww;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xyyy, const[0].xyyy;
>  1: MUL temp[2], temp[0].xyyy, const[4];
>  2: TEX temp[3], temp[2], 2D[0];
>  3: MOV temp[0].xyz, temp[3];
>  4: MUL temp[0].xyz, temp[0].xyzz, const[0].zzzz;
>  5: ADD temp[1].xy, input[0].xyyy, const[0].xwww;
>  6: MUL temp[4], temp[1].xyyy, const[4];
>  7: TEX temp[5], temp[4], 2D[0];
>  8: MOV temp[1].xyz, temp[5];
>  9: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 10: ADD temp[1].xy, input[0].xyyy, const[1].yzzz;
> 11: MUL temp[6], temp[1].xyyy, const[4];
> 12: TEX temp[7], temp[6], 2D[0];
> 13: MOV temp[1].xyz, temp[7];
> 14: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 15: MUL temp[8], input[0].xyyy, const[4];
> 16: TEX temp[9], temp[8], 2D[0];
> 17: MOV temp[1].xyz, temp[9];
> 18: MAD temp[0].xyz, temp[1].xyzz, const[2].xxxx, temp[0].xyzz;
> 19: ADD temp[1].xy, input[0].xyyy, const[2].yzzz;
> 20: MUL temp[10], temp[1].xyyy, const[4];
> 21: TEX temp[11], temp[10], 2D[0];
> 22: MOV temp[1].xyz, temp[11];
> 23: MAD temp[0].xyz, temp[1].xyzz, const[1].wwww, temp[0].xyzz;
> 24: ADD temp[1].xy, input[0].xyyy, const[2].ywww;
> 25: MUL temp[12], temp[1].xyyy, const[4];
> 26: TEX temp[13], temp[12], 2D[0];
> 27: MOV temp[1].xyz, temp[13];
> 28: MAD temp[0].xyz, temp[1].xyzz, const[1].xxxx, temp[0].xyzz;
> 29: ADD temp[1].xy, input[0].xyyy, const[3].xyyy;
> 30: MUL temp[14], temp[1].xyyy, const[4];
> 31: TEX temp[15], temp[14], 2D[0];
> 32: MOV temp[1].xyz, temp[15];
> 33: MAD temp[0].xyz, temp[1].xyzz, const[0].zzzz, temp[0].xyzz;
> 34: MUL temp[0].xyz, temp[0].xyzz, const[3].zzzz;
> 35: MOV temp[0].xyz, temp[0].xyzx;
> 36: MOV temp[0].w, const[3].wwww;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[2].xy, temp[0].xy__, const[4].xy__;
>  2: TEX temp[3].xyz, temp[2].xy__, 2D[0];
>  3: MOV temp[0].xyz, temp[3].xyz_;
>  4: MUL temp[0].xyz, temp[0].xyz_, const[0].zzz_;
>  5: ADD temp[1].xy, input[0].xy__, const[0].xw__;
>  6: MUL temp[4].xy, temp[1].xy__, const[4].xy__;
>  7: TEX temp[5].xyz, temp[4].xy__, 2D[0];
>  8: MOV temp[1].xyz, temp[5].xyz_;
>  9: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 10: ADD temp[1].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[6].xy, temp[1].xy__, const[4].xy__;
> 12: TEX temp[7].xyz, temp[6].xy__, 2D[0];
> 13: MOV temp[1].xyz, temp[7].xyz_;
> 14: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 15: MUL temp[8].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[0];
> 17: MOV temp[1].xyz, temp[9].xyz_;
> 18: MAD temp[0].xyz, temp[1].xyz_, const[2].xxx_, temp[0].xyz_;
> 19: ADD temp[1].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[10].xy, temp[1].xy__, const[4].xy__;
> 21: TEX temp[11].xyz, temp[10].xy__, 2D[0];
> 22: MOV temp[1].xyz, temp[11].xyz_;
> 23: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 24: ADD temp[1].xy, input[0].xy__, const[2].yw__;
> 25: MUL temp[12].xy, temp[1].xy__, const[4].xy__;
> 26: TEX temp[13].xyz, temp[12].xy__, 2D[0];
> 27: MOV temp[1].xyz, temp[13].xyz_;
> 28: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 29: ADD temp[1].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[14].xy, temp[1].xy__, const[4].xy__;
> 31: TEX temp[15].xyz, temp[14].xy__, 2D[0];
> 32: MOV temp[1].xyz, temp[15].xyz_;
> 33: MAD temp[0].xyz, temp[1].xyz_, const[0].zzz_, temp[0].xyz_;
> 34: MUL temp[0].xyz, temp[0].xyz_, const[3].zzz_;
> 35: MOV temp[0].xyz, temp[0].xyz_;
> 36: MOV temp[0].w, const[3].___w;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: ADD temp[0].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[2].xy, temp[0].xy__, const[4].xy__;
>  2: TEX temp[3].xyz, temp[2].xy__, 2D[0];
>  3: MOV temp[0].xyz, temp[3].xyz_;
>  4: MUL temp[0].xyz, temp[0].xyz_, const[0].zzz_;
>  5: ADD temp[1].xy, input[0].xy__, const[0].xw__;
>  6: MUL temp[4].xy, temp[1].xy__, const[4].xy__;
>  7: TEX temp[5].xyz, temp[4].xy__, 2D[0];
>  8: MOV temp[1].xyz, temp[5].xyz_;
>  9: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 10: ADD temp[1].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[6].xy, temp[1].xy__, const[4].xy__;
> 12: TEX temp[7].xyz, temp[6].xy__, 2D[0];
> 13: MOV temp[1].xyz, temp[7].xyz_;
> 14: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 15: MUL temp[8].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[9].xyz, temp[8].xy__, 2D[0];
> 17: MOV temp[1].xyz, temp[9].xyz_;
> 18: MAD temp[0].xyz, temp[1].xyz_, const[2].xxx_, temp[0].xyz_;
> 19: ADD temp[1].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[10].xy, temp[1].xy__, const[4].xy__;
> 21: TEX temp[11].xyz, temp[10].xy__, 2D[0];
> 22: MOV temp[1].xyz, temp[11].xyz_;
> 23: MAD temp[0].xyz, temp[1].xyz_, const[1].www_, temp[0].xyz_;
> 24: ADD temp[1].xy, input[0].xy__, const[2].yw__;
> 25: MUL temp[12].xy, temp[1].xy__, const[4].xy__;
> 26: TEX temp[13].xyz, temp[12].xy__, 2D[0];
> 27: MOV temp[1].xyz, temp[13].xyz_;
> 28: MAD temp[0].xyz, temp[1].xyz_, const[1].xxx_, temp[0].xyz_;
> 29: ADD temp[1].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[14].xy, temp[1].xy__, const[4].xy__;
> 31: TEX temp[15].xyz, temp[14].xy__, 2D[0];
> 32: MOV temp[1].xyz, temp[15].xyz_;
> 33: MAD temp[0].xyz, temp[1].xyz_, const[0].zzz_, temp[0].xyz_;
> 34: MUL temp[0].xyz, temp[0].xyz_, const[3].zzz_;
> 35: MOV temp[0].xyz, temp[0].xyz_;
> 36: MOV temp[0].w, const[3].___w;
> 37: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: ADD temp[16].xy, input[0].xy__, const[0].xy__;
>  1: MUL temp[17].xy, temp[16].xy__, const[4].xy__;
>  2: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  3: MOV temp[19].xyz, temp[18].xyz_;
>  4: MUL temp[20].xyz, temp[19].xyz_, const[0].zzz_;
>  5: ADD temp[21].xy, input[0].xy__, const[0].xw__;
>  6: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  7: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  8: MOV temp[24].xyz, temp[23].xyz_;
>  9: MAD temp[25].xyz, temp[24].xyz_, const[1].xxx_, temp[20].xyz_;
> 10: ADD temp[26].xy, input[0].xy__, const[1].yz__;
> 11: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
> 12: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 13: MOV temp[29].xyz, temp[28].xyz_;
> 14: MAD temp[30].xyz, temp[29].xyz_, const[1].www_, temp[25].xyz_;
> 15: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 16: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 17: MOV temp[33].xyz, temp[32].xyz_;
> 18: MAD temp[34].xyz, temp[33].xyz_, const[2].xxx_, temp[30].xyz_;
> 19: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 20: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 21: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 22: MOV temp[38].xyz, temp[37].xyz_;
> 23: MAD temp[39].xyz, temp[38].xyz_, const[1].www_, temp[34].xyz_;
> 24: ADD temp[40].xy, input[0].xy__, const[2].yw__;
> 25: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 26: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 27: MOV temp[43].xyz, temp[42].xyz_;
> 28: MAD temp[44].xyz, temp[43].xyz_, const[1].xxx_, temp[39].xyz_;
> 29: ADD temp[45].xy, input[0].xy__, const[3].xy__;
> 30: MUL temp[46].xy, temp[45].xy__, const[4].xy__;
> 31: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 32: MOV temp[48].xyz, temp[47].xyz_;
> 33: MAD temp[49].xyz, temp[48].xyz_, const[0].zzz_, temp[44].xyz_;
> 34: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 35: MOV temp[51].xyz, temp[50].xyz_;
> 36: MOV temp[51].w, const[3].___w;
> 37: MOV output[0], temp[51];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[0].0w__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[2].0w__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>CONST[0] = {     0.0000    -4.3500     0.1000    -2.5000 }
>CONST[1] = {     0.2500     0.0000    -0.7500     0.3000 }
>CONST[2] = {     0.5000     0.0000     0.7500     2.5000 }
>CONST[3] = {     0.0000     4.3500     0.5556     1.0000 }
>CONST[5] = {     0.0000     0.0000    -2.5000     0.0000 }
>CONST[6] = {     0.0000     0.0000     2.5000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[5].yz__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[6].yz__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>CONST[0] = {     0.0000    -4.3500     0.1000    -2.5000 }
>CONST[1] = {     0.2500     0.0000    -0.7500     0.3000 }
>CONST[2] = {     0.5000     0.0000     0.7500     2.5000 }
>CONST[3] = {     0.0000     4.3500     0.5556     1.0000 }
>CONST[5] = {     0.0000     0.0000    -2.5000     0.0000 }
>CONST[6] = {     0.0000     0.0000     2.5000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: MUL temp[17].xy, (const[0] + input[0]).xy__, const[4].xy__;
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: MUL temp[20].xyz, temp[18].xyz_, const[0].zzz_;
>  3: ADD temp[21].xy, input[0].xy__, const[5].yz__;
>  4: MUL temp[22].xy, temp[21].xy__, const[4].xy__;
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: MAD temp[25].xyz, temp[23].xyz_, const[1].xxx_, temp[20].xyz_;
>  7: ADD temp[26].xy, input[0].xy__, const[1].yz__;
>  8: MUL temp[27].xy, temp[26].xy__, const[4].xy__;
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: MAD temp[30].xyz, temp[28].xyz_, const[1].www_, temp[25].xyz_;
> 11: MUL temp[31].xy, input[0].xy__, const[4].xy__;
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: MAD temp[34].xyz, temp[32].xyz_, none.HHH_, temp[30].xyz_;
> 14: ADD temp[35].xy, input[0].xy__, const[2].yz__;
> 15: MUL temp[36].xy, temp[35].xy__, const[4].xy__;
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: MAD temp[39].xyz, temp[37].xyz_, const[1].www_, temp[34].xyz_;
> 18: ADD temp[40].xy, input[0].xy__, const[6].yz__;
> 19: MUL temp[41].xy, temp[40].xy__, const[4].xy__;
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: MAD temp[44].xyz, temp[42].xyz_, const[1].xxx_, temp[39].xyz_;
> 22: MUL temp[46].xy, (const[3] + input[0]).xy__, const[4].xy__;
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: MAD temp[49].xyz, temp[47].xyz_, const[0].zzz_, temp[44].xyz_;
> 25: MUL temp[50].xyz, temp[49].xyz_, const[3].zzz_;
> 26: MOV temp[51].xyz, temp[50].xyz_;
> 27: MOV temp[51].w, none.___1;
> 28: MOV output[0], temp[51];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>  1: TEX temp[18].xyz, temp[17].xy__, 2D[0];
>  2: src0.xyz = temp[18], src1.xyz = const[0]
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
>  3: src0.xyz = input[0], src1.xyz = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.yz_
>  4: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  5: TEX temp[23].xyz, temp[22].xy__, 2D[0];
>  6: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
>  7: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  8: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
>  9: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 10: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 11: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
> 12: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 13: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 14: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
> 15: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 16: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 17: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 18: src0.xyz = input[0], src1.xyz = const[6]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.yz_
> 19: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
> 20: TEX temp[42].xyz, temp[41].xy__, 2D[0];
> 21: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 22: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
> 23: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 24: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 25: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 26: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 27: 
>     MAD temp[51].w, src0.1, src0.1, src0.0
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>     MAD temp[51].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
>  2: src0.xyz = input[0], src1.xyz = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.yz_
>  3: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  4: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  5: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
>  6: src0.xyz = input[0], src1.xyz = const[6]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.yz_
>  7: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 10: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
> 11: BEGIN_TEX;
> 12: TEX temp[18].xyz, temp[17].xy__, 2D[0];
> 13: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 14: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 15: TEX temp[23].xyz, temp[22].xy__, 2D[0];
> 16: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 17: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 18: TEX temp[42].xyz, temp[41].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[18], src1.xyz = const[0] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[17].xy, srcp.xy_, src2.xy_, src0.000
>     MAD temp[51].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[46].xy, srcp.xy_, src2.xy_, src0.000
>  2: src0.xyz = input[0], src1.xyz = const[5]
>     MAD temp[21].xy, src0.xy_, src0.111, src1.yz_
>  3: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[26].xy, src0.xy_, src0.111, src1.yz_
>  4: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[31].xy, src0.xy_, src1.xy_, src0.000
>  5: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[35].xy, src0.xy_, src0.111, src1.yz_
>  6: src0.xyz = input[0], src1.xyz = const[6]
>     MAD temp[40].xy, src0.xy_, src0.111, src1.yz_
>  7: src0.xyz = temp[21], src1.xyz = const[4]
>     MAD temp[22].xy, src0.xy_, src1.xy_, src0.000
>  8: src0.xyz = temp[26], src1.xyz = const[4]
>     MAD temp[27].xy, src0.xy_, src1.xy_, src0.000
>  9: src0.xyz = temp[35], src1.xyz = const[4]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 10: src0.xyz = temp[40], src1.xyz = const[4]
>     MAD temp[41].xy, src0.xy_, src1.xy_, src0.000
> 11: BEGIN_TEX;
> 12: TEX temp[18].xyz, temp[17].xy__, 2D[0];
> 13: TEX temp[47].xyz, temp[46].xy__, 2D[0];
> 14: TEX temp[32].xyz, temp[31].xy__, 2D[0];
> 15: TEX temp[23].xyz, temp[22].xy__, 2D[0];
> 16: TEX temp[28].xyz, temp[27].xy__, 2D[0];
> 17: TEX temp[37].xyz, temp[36].xy__, 2D[0];
> 18: TEX temp[42].xyz, temp[41].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[18], src1.xyz = const[0] SEM_WAIT
>     MAD temp[20].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[23], src1.xyz = const[1], src2.xyz = temp[20]
>     MAD temp[25].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[28], src0.w = const[1], src1.xyz = temp[25]
>     MAD temp[30].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[32], src1.xyz = temp[30]
>     MAD temp[34].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[37], src0.w = const[1], src1.xyz = temp[34]
>     MAD temp[39].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[42], src1.xyz = const[1], src2.xyz = temp[39]
>     MAD temp[44].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[47], src1.xyz = const[0], src2.xyz = temp[44]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[49], src1.xyz = const[3]
>     MAD temp[50].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[50]
>     MAD temp[51].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[51], src0.w = temp[51]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[0], src1.xyz = const[0], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[1].xy, srcp.xy_, src2.xy_, src0.00_
>     MAD temp[0].w, src0.1, src0.1, src0.0
>  1: src0.xyz = input[0], src1.xyz = const[3], src2.xyz = const[4], srcp.xyz = (src1 + src0)
>     MAD temp[2].xy, srcp.xy_, src2.xy_, src0.00_
>  2: src0.xyz = input[0], src1.xyz = const[5]
>     MAD temp[3].xy, src0.xy_, src0.11_, src1.yz_
>  3: src0.xyz = input[0], src1.xyz = const[1]
>     MAD temp[4].xy, src0.xy_, src0.11_, src1.yz_
>  4: src0.xyz = input[0], src1.xyz = const[4]
>     MAD temp[5].xy, src0.xy_, src1.xy_, src0.00_
>  5: src0.xyz = input[0], src1.xyz = const[2]
>     MAD temp[6].xy, src0.xy_, src0.11_, src1.yz_
>  6: src0.xyz = input[0], src1.xyz = const[6]
>     MAD temp[0].xy, src0.xy_, src0.11_, src1.yz_
>  7: src0.xyz = temp[3], src1.xyz = const[4]
>     MAD temp[3].xy, src0.xy_, src1.xy_, src0.00_
>  8: src0.xyz = temp[4], src1.xyz = const[4]
>     MAD temp[4].xy, src0.xy_, src1.xy_, src0.00_
>  9: src0.xyz = temp[6], src1.xyz = const[4]
>     MAD temp[6].xy, src0.xy_, src1.xy_, src0.00_
> 10: src0.xyz = temp[0], src1.xyz = const[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
> 11: BEGIN_TEX;
> 12: TEX temp[1].xyz, temp[1].xy__, 2D[0];
> 13: TEX temp[2].xyz, temp[2].xy__, 2D[0];
> 14: TEX temp[5].xyz, temp[5].xy__, 2D[0];
> 15: TEX temp[3].xyz, temp[3].xy__, 2D[0];
> 16: TEX temp[4].xyz, temp[4].xy__, 2D[0];
> 17: TEX temp[6].xyz, temp[6].xy__, 2D[0];
> 18: TEX temp[7].xyz, temp[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
> 19: src0.xyz = temp[1], src1.xyz = const[0] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 20: src0.xyz = temp[3], src1.xyz = const[1], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 21: src0.xyz = temp[4], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 22: src0.xyz = temp[5], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src1.xyz
> 23: src0.xyz = temp[6], src0.w = const[1], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.www, src1.xyz
> 24: src0.xyz = temp[7], src1.xyz = const[1], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 25: src0.xyz = temp[2], src1.xyz = const[0], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src2.xyz
> 26: src0.xyz = temp[0], src1.xyz = const[3]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 27: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, src0.000
> 28: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=48*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 10, tex_end: 0  (code_addr: 00000280)
>  0: xyz:  t0  c0  c4   add-> t1.xy                (01864800)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz: srcp.xyz   c4.xyz      0.0    op: 0045040f 
>       w:      1.0      1.0      0.0    op: 00040891
>  1: xyz:  t0  c3  c4   add-> t2.xy                (018a48c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz   c4.xyz      0.0    op: 0045040f 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  2: xyz:  t0  c5  t0  bias-> t3.xy                (018c0940)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c5.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  3: xyz:  t0  c1  t0  bias-> t4.xy                (01900840)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c1.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t0  c4  t0  bias-> t5.xy                (01940900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  5: xyz:  t0  c2  t0  bias-> t6.xy                (01980880)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c2.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  6: xyz:  t0  c6  t0  bias-> t0.xy                (01800980)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0   c6.yzx    op: 00060a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t3  c4  t0  bias-> t3.xy                (018c0903)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  8: xyz:  t4  c4  t0  bias-> t4.xy                (01900904)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t4.x     t4.x     t4.x    op: 00000000
>  9: xyz:  t6  c4  t0  bias-> t6.xy                (01980906)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t6.x     t6.x     t6.x    op: 00000000
> 10: xyz:  t0  c4  t0  bias-> t0.xy                (01800900)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c4.xyz      0.0    op: 00050200 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>NODE 1: alu_offset: 11, tex_offset: 0, alu_end: 9, tex_end: 6  (code_addr: 004c024b)
>  TEX:
>    TEX t1, t1, texture[0]   (00008041)
>    TEX t2, t2, texture[0]   (00008082)
>    TEX t5, t5, texture[0]   (00008145)
>    TEX t3, t3, texture[0]   (000080c3)
>    TEX t4, t4, texture[0]   (00008104)
>    TEX t6, t6, texture[0]   (00008186)
>    TEX t7, t0, texture[0]   (000081c0)
> 11: xyz:  t1  c0  t0  bias-> t0.xyz               (03800801)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 12: xyz:  t3  c1  t0  bias-> t0.xyz               (03800843)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   c1.xxx   t0.xyz    op: 00020280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 13: xyz:  t4  t0  t0  bias-> t0.xyz               (03800004)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t4.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 14: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz      0.5   t0.xyz    op: 00010b00 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 15: xyz:  t6  t0  t0  bias-> t0.xyz               (03800006)
>       w:  c1  t0  t0  bias->                      (00000021)
>     xyz:   t6.xyz   c1.www   t0.xyz    op: 00010600 
>       w:     t6.x     t6.x     t6.x    op: 00000000
> 16: xyz:  t7  c1  t0  bias-> t0.xyz               (03800847)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t7.xyz   c1.xxx   t0.xyz    op: 00020280 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 17: xyz:  t2  c0  t0  bias-> t0.xyz               (03800802)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c0.zzz   t0.xyz    op: 00020380 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 18: xyz:  t0  c3  t0  bias-> t0.xyz               (038008c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c3.zzz      0.0    op: 00050380 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 19: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 20: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MOV OUT[0], TEMP[0]
>  2: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MOV output[0], temp[1];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=49*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 0, tex_end: 0  (code_addr: 00400000)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    0.5000,   255.0000,     0.0039,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: DP3 TEMP[1].x, IN[2].xyzz, IN[2].xyzz
>  2: RSQ TEMP[2].x, TEMP[1].xxxx
>  3: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[2].xxxx
>  4: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].xxxx, IMM[0].xxxx
>  5: MUL TEMP[3].x, IN[1].xxxx, IMM[0].yyyy
>  6: FLR TEMP[4].x, TEMP[3].xxxx
>  7: MOV TEMP[5].x, TEMP[4].xxxx
>  8: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
>  9: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].yyyy
> 10: FLR TEMP[4].x, TEMP[3].xxxx
> 11: MOV TEMP[5].y, TEMP[4].xxxx
> 12: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 13: MOV TEMP[5].z, TEMP[3].xxxx
> 14: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].zzzz
> 15: MOV TEMP[2].xyz, TEMP[5].xyzx
> 16: MOV TEMP[1].w, IMM[0].wwww
> 17: MOV TEMP[2].w, IMM[0].wwww
> 18: MOV OUT[0], TEMP[0]
> 19: MOV OUT[1], TEMP[1]
> 20: MOV OUT[2], TEMP[2]
> 21: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: MOV temp[1].w, const[0].wwww;
> 17: MOV temp[2].w, const[0].wwww;
> 18: MOV output[0], temp[0];
> 19: MOV output[1], temp[1];
> 20: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, |temp[1].xxxx|;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FRC temp[4].x, temp[3].xxxx;
>  7: ADD temp[4].x, temp[3].xxxx, -temp[4];
>  8: MOV temp[5].x, temp[4].xxxx;
>  9: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 10: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 11: FRC temp[4].x, temp[3].xxxx;
> 12: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 13: MOV temp[5].y, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MOV temp[5].z, temp[3].xxxx;
> 16: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 17: MOV temp[2].xyz, temp[5].xyzx;
> 18: MOV temp[1].w, const[0].wwww;
> 19: MOV temp[2].w, const[0].wwww;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: MOV temp[1].w, const[0].___w;
> 19: MOV temp[2].w, const[0].___w;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: ADD temp[13].x, temp[11].x___, -temp[12].x___;
>  8: MOV temp[14].x, temp[13].x___;
>  9: ADD temp[15].x, temp[11].x___, -temp[13].x___;
> 10: MUL temp[16].x, temp[15].x___, const[0].y___;
> 11: FRC temp[17].x, temp[16].x___;
> 12: ADD temp[18].x, temp[16].x___, -temp[17].x___;
> 13: MOV temp[14].y, temp[18]._x__;
> 14: ADD temp[19].x, temp[16].x___, -temp[18].x___;
> 15: MOV temp[20].z, temp[19].__x_;
> 16: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 17: MOV temp[21].xyz, temp[20].xyz_;
> 18: MOV temp[10].w, const[0].___w;
> 19: MOV temp[21].w, const[0].___w;
> 20: MOV output[0], temp[6];
> 21: MOV output[1], temp[10];
> 22: MOV output[2], temp[21];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: DP3 temp[7].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[8].x, |temp[7].x___|;
>  3: MUL temp[9].xyz, input[2].xyz_, temp[8].xxx_;
>  4: MAD temp[10].xyz, temp[9].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[11].x, input[1].x___, const[0].y___;
>  6: FRC temp[12].x, temp[11].x___;
>  7: MOV temp[14].x, (temp[11] - temp[12]).x___;
>  8: ADD temp[15].x, temp[11].x___, -(temp[11] - temp[12]).x___;
>  9: MUL temp[16].x, temp[15].x___, const[0].y___;
> 10: FRC temp[17].x, temp[16].x___;
> 11: MOV temp[14].y, (temp[16] - temp[17])._x__;
> 12: ADD temp[19].x, temp[16].x___, -(temp[16] - temp[17]).x___;
> 13: MOV temp[20].z, temp[19].__x_;
> 14: MUL temp[20].xy, temp[14].xy__, const[0].zz__;
> 15: MOV temp[21].xyz, temp[20].xyz_;
> 16: MOV temp[10].w, none.___0;
> 17: MOV temp[21].w, none.___0;
> 18: MOV output[0], temp[6];
> 19: MOV output[1], temp[10];
> 20: MOV output[2], temp[21];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[6], input[0].xy__, 2D[0];
>  1: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>  2: src0.xyz = temp[7]
>     REPL_ALPHA temp[8].x
>     RSQ, |src0.x|
>  3: src0.xyz = input[2], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src1.xxx, src0.000
>  4: src0.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.HHH, src0.HHH
>  5: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>  6: src0.xyz = temp[11]
>     FRC temp[12].x, src0.x__
>  7: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  8: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, src1.x__, src0.111, -srcp.x__
>  9: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.x__, src1.y__, src0.000
> 10: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].x, src1.x__, src0.111, -srcp.x__
> 13: src0.xyz = temp[19]
>     MAD temp[20].z, src0.__x, src0.111, src0.000
> 14: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 15: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 16: 
>     MAD temp[10].w, src0.0, src0.1, src0.0
> 17: 
>     MAD temp[21].w, src0.0, src0.1, src0.0
> 18: src0.xyz = temp[6], src0.w = temp[6]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 19: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8], src1.xyz = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.xyz = temp[15], src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.xyz = temp[19], src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[6], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[11].x, src0.x__, src1.y__, src0.000
>     MAD temp[21].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[7].x, src0.xyz, src0.xyz
>     MAD temp[10].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[11], src1.xyz = temp[7]
>     FRC temp[12].x, src0.x__
>     RSQ temp[8].w, |src1.x|
>  5: src0.xyz = temp[12], src1.xyz = temp[11], srcp.xyz = (src1 - src0)
>     MAD temp[14].x, srcp.x__, src0.111, src0.000
>  6: src0.xyz = input[2], src0.w = temp[8]
>     MAD temp[9].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[12], src1.xyz = temp[11], src2.xyz = temp[9], srcp.xyz = (src1 - src0)
>     MAD temp[10].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[15].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[15], src1.xyz = const[0]
>     MAD temp[16].x, src0.w__, src1.y__, src0.000
>  9: src0.xyz = temp[16]
>     FRC temp[17].x, src0.x__
> 10: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[14].y, srcp._x_, src0.111, src0.000
> 11: src0.xyz = temp[17], src1.xyz = temp[16], srcp.xyz = (src1 - src0)
>     MAD temp[19].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[14], src1.xyz = const[0]
>     MAD temp[20].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[19]
>     MAD temp[20].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[20]
>     MAD temp[21].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[6], src0.w = temp[6] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[10], src0.w = temp[10]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[21], src0.w = temp[21]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[1].x, src0.x__, src1.y__, src0.0__
>     MAD temp[1].w, src0.0, src0.1, src0.0
>  3: src0.xyz = input[2]
>     DP3 temp[1].y, src0.xyz, src0.xyz
>     MAD temp[2].w, src0.0, src0.1, src0.0
>  4: src0.xyz = temp[1], src1.xyz = temp[1]
>     FRC temp[3].x, src0.x__
>     RSQ temp[3].w, |src1.y|
>  5: src0.xyz = temp[3], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].x, srcp.x__, src0.11_, src0.00_
>  6: src0.xyz = input[2], src0.w = temp[3]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>  7: src0.xyz = temp[3], src1.xyz = temp[1], src2.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[1].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
>  8: src0.w = temp[3], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.y__, src0.0__
>  9: src0.xyz = temp[2]
>     FRC temp[3].x, src0.x__
> 10: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[4].y, srcp._x_, src0.11_, src0.00_
> 11: src0.xyz = temp[3], src1.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[3].w, src1.x, src0.1, -srcp.x
> 12: src0.xyz = temp[4], src1.xyz = const[0]
>     MAD temp[2].xy, src0.xy_, src1.zz_, src0.000
> 13: src0.w = temp[3]
>     MAD temp[2].z, src0.__w, src0.111, src0.000
> 14: src0.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src0.111, src0.000
> 15: src0.xyz = temp[0], src0.w = temp[0] SEM_WAIT
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 16: src0.xyz = temp[1], src0.w = temp[2]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[2], src0.w = temp[1]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=50*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 15, tex_end: 0  (code_addr: 004003c0)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.yyy      0.0    op: 00050300 
>       w:      0.0      1.0      0.0    op: 00040890
>  1: xyz:  t2  t0  t0  bias-> t1.y                 (01040002)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:      0.0      1.0      0.0    op: 00040890
>  2: xyz:  t1  t1  t0  bias-> t3.x                 (008c0041)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:   |t1.y|     t1.x     t1.x    op: 05800044
>  3: xyz:  t3  t1  t0   sub-> t4.x                 (00900043)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  4: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t2.xyz   t3.www      0.0    op: 00050600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  5: xyz:  t3  t1  t2   sub-> t1.xyz               (03842043)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t2.xyz      0.5      0.5    op: 00258b08 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www   c0.yyy      0.0    op: 0005030c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t2  t0  t0  bias-> t3.x                 (008c0002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  8: xyz:  t3  t2  t0   sub-> t4.y                 (01100083)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t3.x     t3.x     t3.x    op: 00000000
>  9: xyz:  t3  t2  t0   sub->                      (00000083)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz:   t3.xxx   t3.xxx   t3.xxx    op: 00204081 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 10: xyz:  t4  c0  t0  bias-> t2.xy                (01880804)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> t2.z                 (02080000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t3.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
> 14: xyz:  t1  t0  t0  bias-> o1.xyz               (3c000001)
>       w:  t2  t0  t0  bias-> o1.w                 (03000002)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t2.w      1.0      0.0    op: 00040889
> 15: xyz:  t2  t0  t0  bias-> o2.xyz               (5c000002)
>       w:  t1  t0  t0  bias-> o2.w                 (05000001)
>     xyz:   t2.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL IN[3], GENERIC[14], PERSPECTIVE
>DCL IN[4], GENERIC[15], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {   -0.5000,     0.5000,   255.0000,     0.0039}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[2], 2D
>  1: TEX TEMP[1].xyz, IN[0].xyyy, SAMP[1], 2D
>  2: ADD TEMP[2].xyz, TEMP[1].xyzz, IMM[0].xxxx
>  3: MUL TEMP[3].xyz, TEMP[2].xxxx, IN[3].xyzz
>  4: MAD TEMP[3].xyz, TEMP[2].yyyy, IN[2].xyzz, TEMP[3].xyzz
>  5: MAD TEMP[3].xyz, TEMP[2].zzzz, IN[4].xyzz, TEMP[3].xyzz
>  6: DP3 TEMP[4].x, TEMP[3].xyzz, TEMP[3].xyzz
>  7: RSQ TEMP[4].x, TEMP[4].xxxx
>  8: MUL TEMP[3].xyz, TEMP[3].xyzz, TEMP[4].xxxx
>  9: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].yyyy, IMM[0].yyyy
> 10: MUL TEMP[3].x, IN[1].xxxx, IMM[0].zzzz
> 11: FLR TEMP[4].x, TEMP[3].xxxx
> 12: MOV TEMP[5].x, TEMP[4].xxxx
> 13: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 14: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].zzzz
> 15: FLR TEMP[4].x, TEMP[3].xxxx
> 16: MOV TEMP[5].y, TEMP[4].xxxx
> 17: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 18: MOV TEMP[5].z, TEMP[3].xxxx
> 19: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].wwww
> 20: MOV TEMP[2].xyz, TEMP[5].xyzx
> 21: TEX TEMP[3].xy, IN[0].xyyy, SAMP[0], 2D
> 22: MOV TEMP[1].w, TEMP[3].xxxx
> 23: MOV TEMP[2].w, TEMP[3].yyyy
> 24: MOV OUT[0], TEMP[0]
> 25: MOV OUT[1], TEMP[1]
> 26: MOV OUT[2], TEMP[2]
> 27: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[1].xyz, input[0].xyyy, 2D[1];
>  2: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  3: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  4: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  6: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  7: RSQ temp[4].x, temp[4].xxxx;
>  8: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
>  9: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 10: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 11: FLR temp[4].x, temp[3].xxxx;
> 12: MOV temp[5].x, temp[4].xxxx;
> 13: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 14: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 15: FLR temp[4].x, temp[3].xxxx;
> 16: MOV temp[5].y, temp[4].xxxx;
> 17: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 18: MOV temp[5].z, temp[3].xxxx;
> 19: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 20: MOV temp[2].xyz, temp[5].xyzx;
> 21: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 22: MOV temp[1].w, temp[3].xxxx;
> 23: MOV temp[2].w, temp[3].yyyy;
> 24: MOV output[0], temp[0];
> 25: MOV output[1], temp[1];
> 26: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, temp[4].xxxx;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FLR temp[4].x, temp[3].xxxx;
> 13: MOV temp[5].x, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 16: FLR temp[4].x, temp[3].xxxx;
> 17: MOV temp[5].y, temp[4].xxxx;
> 18: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 19: MOV temp[5].z, temp[3].xxxx;
> 20: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 21: MOV temp[2].xyz, temp[5].xyzx;
> 22: TEX temp[7], input[0].xyyy, 2D[0];
> 23: MOV temp[3].xy, temp[7];
> 24: MOV temp[1].w, temp[3].xxxx;
> 25: MOV temp[2].w, temp[3].yyyy;
> 26: MOV output[0], temp[0];
> 27: MOV output[1], temp[1];
> 28: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[2];
>  1: TEX temp[6], input[0].xyyy, 2D[1];
>  2: MOV temp[1].xyz, temp[6];
>  3: ADD temp[2].xyz, temp[1].xyzz, const[0].xxxx;
>  4: MUL temp[3].xyz, temp[2].xxxx, input[3].xyzz;
>  5: MAD temp[3].xyz, temp[2].yyyy, input[2].xyzz, temp[3].xyzz;
>  6: MAD temp[3].xyz, temp[2].zzzz, input[4].xyzz, temp[3].xyzz;
>  7: DP3 temp[4].x, temp[3].xyzz, temp[3].xyzz;
>  8: RSQ temp[4].x, |temp[4].xxxx|;
>  9: MUL temp[3].xyz, temp[3].xyzz, temp[4].xxxx;
> 10: MAD temp[1].xyz, temp[3].xyzz, const[0].yyyy, const[0].yyyy;
> 11: MUL temp[3].x, input[1].xxxx, const[0].zzzz;
> 12: FRC temp[4].x, temp[3].xxxx;
> 13: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 14: MOV temp[5].x, temp[4].xxxx;
> 15: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 16: MUL temp[3].x, temp[3].xxxx, const[0].zzzz;
> 17: FRC temp[4].x, temp[3].xxxx;
> 18: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 19: MOV temp[5].y, temp[4].xxxx;
> 20: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 21: MOV temp[5].z, temp[3].xxxx;
> 22: MUL temp[5].xy, temp[5].xyyy, const[0].wwww;
> 23: MOV temp[2].xyz, temp[5].xyzx;
> 24: TEX temp[7], input[0].xyyy, 2D[0];
> 25: MOV temp[3].xy, temp[7];
> 26: MOV temp[1].w, temp[3].xxxx;
> 27: MOV temp[2].w, temp[3].yyyy;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[2];
>  1: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[1].xyz, temp[6].xyz_;
>  3: ADD temp[2].xyz, temp[1].xyz_, const[0].xxx_;
>  4: MUL temp[3].xyz, temp[2].xxx_, input[3].xyz_;
>  5: MAD temp[3].xyz, temp[2].yyy_, input[2].xyz_, temp[3].xyz_;
>  6: MAD temp[3].xyz, temp[2].zzz_, input[4].xyz_, temp[3].xyz_;
>  7: DP3 temp[4].x, temp[3].xyz_, temp[3].xyz_;
>  8: RSQ temp[4].x, |temp[4].x___|;
>  9: MUL temp[3].xyz, temp[3].xyz_, temp[4].xxx_;
> 10: MAD temp[1].xyz, temp[3].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[3].x, input[1].x___, const[0].z___;
> 12: FRC temp[4].x, temp[3].x___;
> 13: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 14: MOV temp[5].x, temp[4].x___;
> 15: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 16: MUL temp[3].x, temp[3].x___, const[0].z___;
> 17: FRC temp[4].x, temp[3].x___;
> 18: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 19: MOV temp[5].y, temp[4]._x__;
> 20: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 21: MOV temp[5].z, temp[3].__x_;
> 22: MUL temp[5].xy, temp[5].xy__, const[0].ww__;
> 23: MOV temp[2].xyz, temp[5].xyz_;
> 24: TEX temp[7].xy, input[0].xy__, 2D[0];
> 25: MOV temp[3].xy, temp[7].xy__;
> 26: MOV temp[1].w, temp[3].___x;
> 27: MOV temp[2].w, temp[3].___y;
> 28: MOV output[0], temp[0];
> 29: MOV output[1], temp[1];
> 30: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: MOV temp[10].xyz, temp[9].xyz_;
>  3: ADD temp[11].xyz, temp[10].xyz_, const[0].xxx_;
>  4: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  5: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  6: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  7: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  8: RSQ temp[16].x, |temp[15].x___|;
>  9: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
> 10: MAD temp[18].xyz, temp[17].xyz_, const[0].yyy_, const[0].yyy_;
> 11: MUL temp[19].x, input[1].x___, const[0].z___;
> 12: FRC temp[20].x, temp[19].x___;
> 13: ADD temp[21].x, temp[19].x___, -temp[20].x___;
> 14: MOV temp[22].x, temp[21].x___;
> 15: ADD temp[23].x, temp[19].x___, -temp[21].x___;
> 16: MUL temp[24].x, temp[23].x___, const[0].z___;
> 17: FRC temp[25].x, temp[24].x___;
> 18: ADD temp[26].x, temp[24].x___, -temp[25].x___;
> 19: MOV temp[22].y, temp[26]._x__;
> 20: ADD temp[27].x, temp[24].x___, -temp[26].x___;
> 21: MOV temp[28].z, temp[27].__x_;
> 22: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 23: MOV temp[29].xyz, temp[28].xyz_;
> 24: TEX temp[30].xy, input[0].xy__, 2D[0];
> 25: MOV temp[31].xy, temp[30].xy__;
> 26: MOV temp[18].w, temp[31].___x;
> 27: MOV temp[29].w, temp[31].___y;
> 28: MOV output[0], temp[8];
> 29: MOV output[1], temp[18];
> 30: MOV output[2], temp[29];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>CONST[0] = {    -0.5000     0.5000   255.0000     0.0039 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: ADD temp[11].xyz, temp[9].xyz_, none.-H-H-H_;
>  3: MUL temp[12].xyz, temp[11].xxx_, input[3].xyz_;
>  4: MAD temp[13].xyz, temp[11].yyy_, input[2].xyz_, temp[12].xyz_;
>  5: MAD temp[14].xyz, temp[11].zzz_, input[4].xyz_, temp[13].xyz_;
>  6: DP3 temp[15].x, temp[14].xyz_, temp[14].xyz_;
>  7: RSQ temp[16].x, |temp[15].x___|;
>  8: MUL temp[17].xyz, temp[14].xyz_, temp[16].xxx_;
>  9: MAD temp[18].xyz, temp[17].xyz_, none.HHH_, none.HHH_;
> 10: MUL temp[19].x, input[1].x___, const[0].z___;
> 11: FRC temp[20].x, temp[19].x___;
> 12: MOV temp[22].x, (temp[19] - temp[20]).x___;
> 13: ADD temp[23].x, temp[19].x___, -(temp[19] - temp[20]).x___;
> 14: MUL temp[24].x, temp[23].x___, const[0].z___;
> 15: FRC temp[25].x, temp[24].x___;
> 16: MOV temp[22].y, (temp[24] - temp[25])._x__;
> 17: ADD temp[27].x, temp[24].x___, -(temp[24] - temp[25]).x___;
> 18: MOV temp[28].z, temp[27].__x_;
> 19: MUL temp[28].xy, temp[22].xy__, const[0].ww__;
> 20: MOV temp[29].xyz, temp[28].xyz_;
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: MOV temp[18].w, temp[30].___x;
> 23: MOV temp[29].w, temp[30].___y;
> 24: MOV output[0], temp[8];
> 25: MOV output[1], temp[18];
> 26: MOV output[2], temp[29];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[8], input[0].xy__, 2D[2];
>  1: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  2: src0.xyz = temp[9]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>  3: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  4: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
>  5: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
>  6: src0.xyz = temp[14]
>     DP3 temp[15].x, src0.xyz, src0.xyz
>  7: src0.xyz = temp[15]
>     REPL_ALPHA temp[16].x
>     RSQ, |src0.x|
>  8: src0.xyz = temp[14], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src1.xxx, src0.000
>  9: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 10: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
> 11: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
> 12: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
> 13: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].x, src1.x__, src0.111, -srcp.x__
> 14: src0.xyz = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.x__, src1.z__, src0.000
> 15: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 16: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
> 17: src0.xyz = temp[25], src1.xyz = temp[24], srcp.xyz = (src1 - src0)
>     MAD temp[27].x, src1.x__, src0.111, -srcp.x__
> 18: src0.xyz = temp[27]
>     MAD temp[28].z, src0.__x, src0.111, src0.000
> 19: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: TEX temp[30].xy, input[0].xy__, 2D[0];
> 22: src0.xyz = temp[30]
>     MAD temp[18].w, src0.x, src0.1, src0.0
> 23: src0.xyz = temp[30]
>     MAD temp[29].w, src0.y, src0.1, src0.0
> 24: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 25: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 26: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.xyz = temp[23], src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.xyz = temp[27], src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16], src1.xyz = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[8], input[0].xy__, 2D[2];
>  2: TEX temp[9].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[30].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[30] SEM_WAIT
>     MAD temp[19].x, src0.x__, src1.z__, src0.000
>     MAD temp[18].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[9], src1.xyz = temp[30]
>     MAD temp[11].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[29].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[19]
>     FRC temp[20].x, src0.x__
>  7: src0.xyz = temp[11], src1.xyz = input[3]
>     MAD temp[12].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[22].x, srcp.x__, src0.111, src0.000
>  9: src0.xyz = temp[20], src1.xyz = temp[19], srcp.xyz = (src1 - src0)
>     MAD temp[23].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[11], src1.xyz = input[2], src2.xyz = temp[12]
>     MAD temp[13].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[23], src1.xyz = const[0]
>     MAD temp[24].x, src0.w__, src1.z__, src0.000
> 12: src0.xyz = temp[11], src1.xyz = input[4], src2.xyz = temp[13]
>     MAD temp[14].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[24]
>     FRC temp[25].x, src0.x__
> 14: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[14], srcp.xyz = (src1 - src0)
>     DP3 temp[15].x, src2.xyz, src2.xyz
>     MAD temp[27].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[25], src1.xyz = temp[24], src2.xyz = temp[15], srcp.xyz = (src1 - src0)
>     MAD temp[22].y, srcp._x_, src0.111, src0.000
>     RSQ temp[16].w, |src2.x|
> 16: src0.w = temp[27]
>     MAD temp[28].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[14], src0.w = temp[16]
>     MAD temp[17].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[22], src0.w = const[0]
>     MAD temp[28].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[17]
>     MAD temp[18].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[28]
>     MAD temp[29].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[8], src0.w = temp[8]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[18], src0.w = temp[18]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[29], src0.w = temp[29]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[5], input[0].xy__, 2D[2];
>  2: TEX temp[6].xyz, input[0].xy__, 2D[1];
>  3: TEX temp[0].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  4: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[0] SEM_WAIT
>     MAD temp[1].x, src0.x__, src1.z__, src0.0__
>     MAD temp[1].w, src2.x, src0.1, src0.0
>  5: src0.xyz = temp[6], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.111, -src0.HHH
>     MAD temp[0].w, src1.y, src0.1, src0.0
>  6: src0.xyz = temp[1]
>     FRC temp[6].x, src0.x__
>  7: src0.xyz = temp[0], src1.xyz = input[3]
>     MAD temp[3].xyz, src0.xxx, src1.xyz, src0.000
>  8: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].x, srcp.x__, src0.11_, src0.00_
>  9: src0.xyz = temp[6], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 10: src0.xyz = temp[0], src1.xyz = input[2], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.yyy, src1.xyz, src2.xyz
> 11: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[2].x, src0.w__, src1.z__, src0.0__
> 12: src0.xyz = temp[0], src1.xyz = input[4], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.zzz, src1.xyz, src2.xyz
> 13: src0.xyz = temp[2]
>     FRC temp[1].x, src0.x__
> 14: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[0], srcp.xyz = (src1 - src0)
>     DP3 temp[1].y, src2.xyz, src2.xyz
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 15: src0.xyz = temp[1], src1.xyz = temp[2], src2.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[7].y, srcp._x_, src0.11_, src0.00_
>     RSQ temp[3].w, |src2.y|
> 16: src0.w = temp[2]
>     MAD temp[1].z, src0.__w, src0.111, src0.000
> 17: src0.xyz = temp[0], src0.w = temp[3]
>     MAD temp[0].xyz, src0.xyz, src0.www, src0.000
> 18: src0.xyz = temp[7], src0.w = const[0]
>     MAD temp[1].xy, src0.xy_, src0.ww_, src0.000
> 19: src0.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src0.HHH, src0.HHH
> 20: src0.xyz = temp[1]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 21: src0.xyz = temp[5], src0.w = temp[5]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 22: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 23: src0.xyz = temp[1], src0.w = temp[0]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=51*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 19, tex_end: 2  (code_addr: 004404c0)
>  TEX:
>    TEX t5, t0, texture[2]   (00009140)
>    TEX t6, t0, texture[1]   (00008980)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t0.x      1.0      0.0    op: 00040886
>  1: xyz:  t6  t0  t0  bias-> t0.xyz               (03800006)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t6.xyz      1.0     -0.5    op: 000d8a80 
>       w:     t0.y      1.0      0.0    op: 00040884
>  2: xyz:  t1  t0  t0  bias-> t6.x                 (00980001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 04804080 
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  3: xyz:  t0  t3  t0  bias-> t3.xyz               (038c00c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t3.xyz      0.0    op: 00050201 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t6  t1  t0   sub-> t7.x                 (009c0046)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  5: xyz:  t6  t1  t0   sub->                      (00000046)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t6.xxx   t6.xxx   t6.xxx    op: 00204081 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  t2  t3  bias-> t1.xyz               (03843080)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.yyy   t2.xyz   t3.xyz    op: 00020202 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t0  c0  t0  bias-> t2.x                 (00880800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.zzz      0.0    op: 0005038c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  8: xyz:  t0  t4  t1  bias-> t0.xyz               (03801100)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.zzz   t4.xyz   t1.xyz    op: 00020203 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  9: xyz:  t2  t0  t0  bias-> t1.x                 (00840002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t2.xxx   t2.xxx    op: 84804080 NOP
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 10: xyz:  t1  t2  t0   sub-> t1.y                 (01040081)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz   t0.xyz   t1.xxx    op: 00a04408 
>       w:     t2.x      1.0  -srcp.x    op: 000b0883
> 11: xyz:  t1  t2  t1   sub-> t7.y                 (011c1081)
>       w:  t0  t0  t0  bias-> t3.w                 (008c0000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:   |t1.y|     t1.x     t1.x    op: 05800047
> 12: xyz:  t0  t0  t0  bias-> t1.z                 (02040000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 13: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t3  t0  t0  bias->                      (00000003)
>     xyz:   t0.xyz   t3.www      0.0    op: 00050600 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 14: xyz:  t7  t0  t0  bias-> t1.xy                (01840007)
>       w:  c0  t0  t0  bias->                      (00000020)
>     xyz:   t7.xyz   c0.www      0.0    op: 00050600 
>       w:     t7.x     t7.x     t7.x    op: 00000000
> 15: xyz:  t0  t0  t0  bias-> t0.xyz               (03800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz      0.5      0.5    op: 00058b00 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 16: xyz:  t1  t0  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 17: xyz:  t5  t0  t0  bias-> o0.xyz               (1c000005)
>       w:  t5  t0  t0  bias-> o0.w                 (01000005)
>     xyz:   t5.xyz      1.0      0.0    op: 00050a80 
>       w:     t5.w      1.0      0.0    op: 00040889
> 18: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t1  t0  t0  bias-> o1.w                 (03000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
> 19: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t0  t0  t0  bias-> o2.w                 (05000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL IN[2], GENERIC[13], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL OUT[1], COLOR[1]
>DCL OUT[2], COLOR[2]
>DCL SAMP[0]
>DCL SAMP[1]
>DCL TEMP[0], LOCAL
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    0.5000,   255.0000,     0.0039,     0.0000}
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[1], 2D
>  1: DP3 TEMP[1].x, IN[2].xyzz, IN[2].xyzz
>  2: RSQ TEMP[2].x, TEMP[1].xxxx
>  3: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[2].xxxx
>  4: MAD TEMP[1].xyz, TEMP[3].xyzz, IMM[0].xxxx, IMM[0].xxxx
>  5: MUL TEMP[3].x, IN[1].xxxx, IMM[0].yyyy
>  6: FLR TEMP[4].x, TEMP[3].xxxx
>  7: MOV TEMP[5].x, TEMP[4].xxxx
>  8: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
>  9: MUL TEMP[3].x, TEMP[3].xxxx, IMM[0].yyyy
> 10: FLR TEMP[4].x, TEMP[3].xxxx
> 11: MOV TEMP[5].y, TEMP[4].xxxx
> 12: ADD TEMP[3].x, TEMP[3].xxxx, -TEMP[4].xxxx
> 13: MOV TEMP[5].z, TEMP[3].xxxx
> 14: MUL TEMP[5].xy, TEMP[5].xyyy, IMM[0].zzzz
> 15: MOV TEMP[2].xyz, TEMP[5].xyzx
> 16: TEX TEMP[3].xy, IN[0].xyyy, SAMP[0], 2D
> 17: MOV TEMP[1].w, TEMP[3].xxxx
> 18: MOV TEMP[2].w, TEMP[3].yyyy
> 19: MOV OUT[0], TEMP[0]
> 20: MOV OUT[1], TEMP[1]
> 21: MOV OUT[2], TEMP[2]
> 22: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[3].xy, input[0].xyyy, 2D[0];
> 17: MOV temp[1].w, temp[3].xxxx;
> 18: MOV temp[2].w, temp[3].yyyy;
> 19: MOV output[0], temp[0];
> 20: MOV output[1], temp[1];
> 21: MOV output[2], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, temp[1].xxxx;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FLR temp[4].x, temp[3].xxxx;
>  7: MOV temp[5].x, temp[4].xxxx;
>  8: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
>  9: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 10: FLR temp[4].x, temp[3].xxxx;
> 11: MOV temp[5].y, temp[4].xxxx;
> 12: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 13: MOV temp[5].z, temp[3].xxxx;
> 14: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 15: MOV temp[2].xyz, temp[5].xyzx;
> 16: TEX temp[6], input[0].xyyy, 2D[0];
> 17: MOV temp[3].xy, temp[6];
> 18: MOV temp[1].w, temp[3].xxxx;
> 19: MOV temp[2].w, temp[3].yyyy;
> 20: MOV output[0], temp[0];
> 21: MOV output[1], temp[1];
> 22: MOV output[2], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[1];
>  1: DP3 temp[1].x, input[2].xyzz, input[2].xyzz;
>  2: RSQ temp[2].x, |temp[1].xxxx|;
>  3: MUL temp[3].xyz, input[2].xyzz, temp[2].xxxx;
>  4: MAD temp[1].xyz, temp[3].xyzz, const[0].xxxx, const[0].xxxx;
>  5: MUL temp[3].x, input[1].xxxx, const[0].yyyy;
>  6: FRC temp[4].x, temp[3].xxxx;
>  7: ADD temp[4].x, temp[3].xxxx, -temp[4];
>  8: MOV temp[5].x, temp[4].xxxx;
>  9: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 10: MUL temp[3].x, temp[3].xxxx, const[0].yyyy;
> 11: FRC temp[4].x, temp[3].xxxx;
> 12: ADD temp[4].x, temp[3].xxxx, -temp[4];
> 13: MOV temp[5].y, temp[4].xxxx;
> 14: ADD temp[3].x, temp[3].xxxx, -temp[4].xxxx;
> 15: MOV temp[5].z, temp[3].xxxx;
> 16: MUL temp[5].xy, temp[5].xyyy, const[0].zzzz;
> 17: MOV temp[2].xyz, temp[5].xyzx;
> 18: TEX temp[6], input[0].xyyy, 2D[0];
> 19: MOV temp[3].xy, temp[6];
> 20: MOV temp[1].w, temp[3].xxxx;
> 21: MOV temp[2].w, temp[3].yyyy;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: TEX temp[6].xy, input[0].xy__, 2D[0];
> 19: MOV temp[3].xy, temp[6].xy__;
> 20: MOV temp[1].w, temp[3].___x;
> 21: MOV temp[2].w, temp[3].___y;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[1];
>  1: DP3 temp[1].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[2].x, |temp[1].x___|;
>  3: MUL temp[3].xyz, input[2].xyz_, temp[2].xxx_;
>  4: MAD temp[1].xyz, temp[3].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[3].x, input[1].x___, const[0].y___;
>  6: FRC temp[4].x, temp[3].x___;
>  7: ADD temp[4].x, temp[3].x___, -temp[4].x___;
>  8: MOV temp[5].x, temp[4].x___;
>  9: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 10: MUL temp[3].x, temp[3].x___, const[0].y___;
> 11: FRC temp[4].x, temp[3].x___;
> 12: ADD temp[4].x, temp[3].x___, -temp[4].x___;
> 13: MOV temp[5].y, temp[4]._x__;
> 14: ADD temp[3].x, temp[3].x___, -temp[4].x___;
> 15: MOV temp[5].z, temp[3].__x_;
> 16: MUL temp[5].xy, temp[5].xy__, const[0].zz__;
> 17: MOV temp[2].xyz, temp[5].xyz_;
> 18: TEX temp[6].xy, input[0].xy__, 2D[0];
> 19: MOV temp[3].xy, temp[6].xy__;
> 20: MOV temp[1].w, temp[3].___x;
> 21: MOV temp[2].w, temp[3].___y;
> 22: MOV output[0], temp[0];
> 23: MOV output[1], temp[1];
> 24: MOV output[2], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, const[0].xxx_, const[0].xxx_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: ADD temp[14].x, temp[12].x___, -temp[13].x___;
>  8: MOV temp[15].x, temp[14].x___;
>  9: ADD temp[16].x, temp[12].x___, -temp[14].x___;
> 10: MUL temp[17].x, temp[16].x___, const[0].y___;
> 11: FRC temp[18].x, temp[17].x___;
> 12: ADD temp[19].x, temp[17].x___, -temp[18].x___;
> 13: MOV temp[15].y, temp[19]._x__;
> 14: ADD temp[20].x, temp[17].x___, -temp[19].x___;
> 15: MOV temp[21].z, temp[20].__x_;
> 16: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 17: MOV temp[22].xyz, temp[21].xyz_;
> 18: TEX temp[23].xy, input[0].xy__, 2D[0];
> 19: MOV temp[24].xy, temp[23].xy__;
> 20: MOV temp[11].w, temp[24].___x;
> 21: MOV temp[22].w, temp[24].___y;
> 22: MOV output[0], temp[7];
> 23: MOV output[1], temp[11];
> 24: MOV output[2], temp[22];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>CONST[0] = {     0.5000   255.0000     0.0039     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: DP3 temp[8].x, input[2].xyz_, input[2].xyz_;
>  2: RSQ temp[9].x, |temp[8].x___|;
>  3: MUL temp[10].xyz, input[2].xyz_, temp[9].xxx_;
>  4: MAD temp[11].xyz, temp[10].xyz_, none.HHH_, none.HHH_;
>  5: MUL temp[12].x, input[1].x___, const[0].y___;
>  6: FRC temp[13].x, temp[12].x___;
>  7: MOV temp[15].x, (temp[12] - temp[13]).x___;
>  8: ADD temp[16].x, temp[12].x___, -(temp[12] - temp[13]).x___;
>  9: MUL temp[17].x, temp[16].x___, const[0].y___;
> 10: FRC temp[18].x, temp[17].x___;
> 11: MOV temp[15].y, (temp[17] - temp[18])._x__;
> 12: ADD temp[20].x, temp[17].x___, -(temp[17] - temp[18]).x___;
> 13: MOV temp[21].z, temp[20].__x_;
> 14: MUL temp[21].xy, temp[15].xy__, const[0].zz__;
> 15: MOV temp[22].xyz, temp[21].xyz_;
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: MOV temp[11].w, temp[23].___x;
> 18: MOV temp[22].w, temp[23].___y;
> 19: MOV output[0], temp[7];
> 20: MOV output[1], temp[11];
> 21: MOV output[2], temp[22];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[7], input[0].xy__, 2D[1];
>  1: src0.xyz = input[2]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>  2: src0.xyz = temp[8]
>     REPL_ALPHA temp[9].x
>     RSQ, |src0.x|
>  3: src0.xyz = input[2], src1.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src1.xxx, src0.000
>  4: src0.xyz = temp[10]
>     MAD temp[11].xyz, src0.xyz, src0.HHH, src0.HHH
>  5: src0.xyz = input[1], src1.xyz = const[0]
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>  6: src0.xyz = temp[12]
>     FRC temp[13].x, src0.x__
>  7: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[16].x, src1.x__, src0.111, -srcp.x__
>  9: src0.xyz = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.x__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].x, src1.x__, src0.111, -srcp.x__
> 13: src0.xyz = temp[20]
>     MAD temp[21].z, src0.__x, src0.111, src0.000
> 14: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: TEX temp[23].xy, input[0].xy__, 2D[0];
> 17: src0.xyz = temp[23]
>     MAD temp[11].w, src0.x, src0.1, src0.0
> 18: src0.xyz = temp[23]
>     MAD temp[22].w, src0.y, src0.1, src0.0
> 19: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 20: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 21: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[23].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[23] SEM_WAIT
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>     MAD temp[22].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[23]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>     MAD temp[11].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[12], src1.xyz = temp[8]
>     FRC temp[13].x, src0.x__
>     RSQ temp[9].w, |src1.x|
>  6: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  7: src0.xyz = input[2], src0.w = temp[9], src1.xyz = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], src2.xyz = temp[10], srcp.xyz = (src1 - src0)
>     MAD temp[11].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[16].w, src1.x, src0.1, -srcp.x
>  9: src0.xyz = temp[16], src0.w = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.w__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.xyz = temp[20], src0.w = temp[20]
>     MAD temp[21].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[7], input[0].xy__, 2D[1];
>  2: TEX temp[23].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[23] SEM_WAIT
>     MAD temp[12].x, src0.x__, src1.y__, src0.000
>     MAD temp[22].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[23]
>     DP3 temp[8].x, src0.xyz, src0.xyz
>     MAD temp[11].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[12], src1.xyz = temp[8]
>     FRC temp[13].x, src0.x__
>     RSQ temp[9].w, |src1.x|
>  6: src0.xyz = temp[13], src1.xyz = temp[12], srcp.xyz = (src1 - src0)
>     MAD temp[15].x, srcp.x__, src0.111, src0.000
>  7: src0.xyz = input[2], src0.w = temp[9]
>     MAD temp[10].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[13], src1.xyz = temp[12], src2.xyz = temp[10], srcp.xyz = (src1 - src0)
>     MAD temp[11].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[16].w, src1.x, src0.1, -srcp.x
>  9: src0.w = temp[16], src1.xyz = const[0]
>     MAD temp[17].x, src0.w__, src1.y__, src0.000
> 10: src0.xyz = temp[17]
>     FRC temp[18].x, src0.x__
> 11: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[15].y, srcp._x_, src0.111, src0.000
> 12: src0.xyz = temp[18], src1.xyz = temp[17], srcp.xyz = (src1 - src0)
>     MAD temp[20].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[15], src1.xyz = const[0]
>     MAD temp[21].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.w = temp[20]
>     MAD temp[21].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[21]
>     MAD temp[22].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[7], src0.w = temp[7]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[11], src0.w = temp[11]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[22], src0.w = temp[22]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[3], input[0].xy__, 2D[1];
>  2: TEX temp[0].xy, input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  3: src0.xyz = input[1], src1.xyz = const[0], src2.xyz = temp[0] SEM_WAIT
>     MAD temp[1].x, src0.x__, src1.y__, src0.0__
>     MAD temp[1].w, src2.y, src0.1, src0.0
>  4: src0.xyz = input[2], src1.xyz = temp[0]
>     DP3 temp[0].x, src0.xyz, src0.xyz
>     MAD temp[0].w, src1.x, src0.1, src0.0
>  5: src0.xyz = temp[1], src1.xyz = temp[0]
>     FRC temp[0].x, src0.x__
>     RSQ temp[2].w, |src1.x|
>  6: src0.xyz = temp[0], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].x, srcp.x__, src0.11_, src0.00_
>  7: src0.xyz = input[2], src0.w = temp[2]
>     MAD temp[2].xyz, src0.xyz, src0.www, src0.000
>  8: src0.xyz = temp[0], src1.xyz = temp[1], src2.xyz = temp[2], srcp.xyz = (src1 - src0)
>     MAD temp[0].xyz, src2.xyz, src0.HHH, src0.HHH
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
>  9: src0.w = temp[2], src1.xyz = const[0]
>     MAD temp[1].x, src0.w__, src1.y__, src0.0__
> 10: src0.xyz = temp[1]
>     FRC temp[2].x, src0.x__
> 11: src0.xyz = temp[2], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[4].y, srcp._x_, src0.11_, src0.00_
> 12: src0.xyz = temp[2], src1.xyz = temp[1], srcp.xyz = (src1 - src0)
>     MAD temp[2].w, src1.x, src0.1, -srcp.x
> 13: src0.xyz = temp[4], src1.xyz = const[0]
>     MAD temp[1].xy, src0.xy_, src1.zz_, src0.000
> 14: src0.w = temp[2]
>     MAD temp[1].z, src0.__w, src0.111, src0.000
> 15: src0.xyz = temp[1]
>     MAD temp[1].xyz, src0.xyz, src0.111, src0.000
> 16: src0.xyz = temp[3], src0.w = temp[3]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
> 17: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[1].xyz, src0.xyz, src0.111, src0.000
>     MAD color[1].w, src0.w, src0.1, src0.0
> 18: src0.xyz = temp[1], src0.w = temp[1]
>     MAD color[2].xyz, src0.xyz, src0.111, src0.000
>     MAD color[2].w, src0.w, src0.1, src0.0
>pc=52*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 15, tex_end: 1  (code_addr: 004203c0)
>  TEX:
>    TEX t3, t0, texture[1]   (000088c0)
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t1  c0  t0  bias-> t1.x                 (00840801)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:   t1.xyz   c0.yyy      0.0    op: 00050300 
>       w:     t0.y      1.0      0.0    op: 00040887
>  1: xyz:  t2  t0  t0  bias-> t0.x                 (00800002)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t2.xyz   t2.xyz   t2.xxx    op: 00804000 
>       w:     t0.x      1.0      0.0    op: 00040883
>  2: xyz:  t1  t0  t0  bias-> t0.x                 (00800001)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:   |t0.x|     t1.x     t1.x    op: 05800043
>  3: xyz:  t0  t1  t0   sub-> t4.x                 (00900040)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xyz      1.0      0.0    op: 00250a8f 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  4: xyz:  t2  t0  t0  bias-> t2.xyz               (03880002)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.xyz   t2.www      0.0    op: 00050600 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  5: xyz:  t0  t1  t2   sub-> t0.xyz               (03802040)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xyz      0.5      0.5    op: 00258b08 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
>  6: xyz:  t0  c0  t0  bias-> t1.x                 (00840800)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www   c0.yyy      0.0    op: 0005030c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
>  7: xyz:  t1  t0  t0  bias-> t2.x                 (00880001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz   t1.xxx   t1.xxx    op: 84804080 NOP
>       w:     t1.x     t1.x     t1.x    op: 00000000
>  8: xyz:  t2  t1  t0   sub-> t4.y                 (01100042)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz: srcp.xxx      1.0      0.0    op: 00250a90 
>       w:     t2.x     t2.x     t2.x    op: 00000000
>  9: xyz:  t2  t1  t0   sub->                      (00000042)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t2.xxx   t2.xxx   t2.xxx    op: 00204081 
>       w:     t1.x      1.0  -srcp.x    op: 000b0883
> 10: xyz:  t4  c0  t0  bias-> t1.xy                (01840804)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t4.xyz   c0.zzz      0.0    op: 00050380 
>       w:     t4.x     t4.x     t4.x    op: 00000000
> 11: xyz:  t0  t0  t0  bias-> t1.z                 (02040000)
>       w:  t2  t0  t0  bias->                      (00000002)
>     xyz:   t2.www      1.0      0.0    op: 00050a8c 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 12: xyz:  t1  t0  t0  bias-> t1.xyz               (03840001)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 13: xyz:  t3  t0  t0  bias-> o0.xyz               (1c000003)
>       w:  t3  t0  t0  bias-> o0.w                 (01000003)
>     xyz:   t3.xyz      1.0      0.0    op: 00050a80 
>       w:     t3.w      1.0      0.0    op: 00040889
> 14: xyz:  t0  t0  t0  bias-> o1.xyz               (3c000000)
>       w:  t0  t0  t0  bias-> o1.w                 (03000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
> 15: xyz:  t1  t0  t0  bias-> o2.xyz               (5c000001)
>       w:  t1  t0  t0  bias-> o2.w                 (05000001)
>     xyz:   t1.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], GENERIC[0], PERSPECTIVE
>DCL IN[1], GENERIC[12], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL TEMP[0], LOCAL
>  0: TEX TEMP[0], IN[0].xyyy, SAMP[0], 2D
>  1: MUL TEMP[0], TEMP[0], IN[1]
>  2: MOV OUT[0], TEMP[0]
>  3: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xyyy, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: TEX temp[0], input[0].xy__, 2D[0];
>  1: MUL temp[0], temp[0], input[1];
>  2: MOV output[0], temp[0];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: MUL temp[2], temp[1], input[1];
>  2: MOV output[0], temp[2];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: TEX temp[1], input[0].xy__, 2D[0];
>  1: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1]
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  2: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[1], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[1], src0.w = temp[1], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[2].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[2], src0.w = temp[2]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: BEGIN_TEX;
>  1: TEX temp[0], input[0].xy__, 2D[0] SEM_WAIT SEM_ACQUIRE;
>  2: src0.xyz = temp[0], src0.w = temp[0], src1.xyz = input[1], src1.w = input[1] SEM_WAIT
>     MAD temp[0].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src1.w, src0.0
>  3: src0.xyz = temp[0], src0.w = temp[0]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=53*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 1, tex_end: 0  (code_addr: 00400040)
>  TEX:
>    TEX t0, t0, texture[0]   (00008000)
>  0: xyz:  t0  t1  t0  bias-> t0.xyz               (03800040)
>       w:  t0  t1  t0  bias-> t0.w                 (00800040)
>     xyz:   t0.xyz   t1.xyz      0.0    op: 00050200 
>       w:     t0.w     t1.w      0.0    op: 00040509
>  1: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t0  t0  t0  bias-> o0.w                 (01000000)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t0.w      1.0      0.0    op: 00040889
>r300: Initial fragment program
>FRAG
>PROPERTY FS_COLOR0_WRITES_ALL_CBUFS 1
>DCL IN[0], POSITION, LINEAR
>DCL IN[1], GENERIC[0], PERSPECTIVE
>DCL IN[2], GENERIC[12], PERSPECTIVE
>DCL IN[3], GENERIC[13], PERSPECTIVE
>DCL IN[4], GENERIC[14], PERSPECTIVE
>DCL IN[5], GENERIC[15], PERSPECTIVE
>DCL IN[6], GENERIC[16], PERSPECTIVE
>DCL OUT[0], COLOR
>DCL SAMP[0]
>DCL SAMP[1]
>DCL SAMP[2]
>DCL SAMP[3]
>DCL CONST[13]
>DCL CONST[0..8]
>DCL TEMP[0]
>DCL TEMP[1], LOCAL
>DCL TEMP[2], LOCAL
>DCL TEMP[3], LOCAL
>DCL TEMP[4], LOCAL
>DCL TEMP[5], LOCAL
>IMM[0] FLT32 {    1.0000,     2.0000,    -1.0000,    10.0000}
>IMM[1] FLT32 {    0.0000,    -1.0000,     0.0000,     0.0000}
>  0: MOV TEMP[0], IN[0]
>  1: MAD TEMP[0].y, IN[0], CONST[13].xxxx, CONST[13].yyyy
>  2: TEX TEMP[1], IN[1].xyyy, SAMP[3], 2D
>  3: MUL TEMP[1], TEMP[1], IN[3]
>  4: MOV TEMP[2].w, TEMP[1].wwww
>  5: ADD TEMP[3].xyz, IMM[0].xxxx, -TEMP[1].xyzz
>  6: ADD TEMP[4].x, IMM[0].xxxx, -CONST[8].xxxx
>  7: MAD TEMP[2].xyz, TEMP[3].xyzz, TEMP[4].xxxx, TEMP[1].xyzz
>  8: TEX TEMP[1].xyz, IN[1].xyyy, SAMP[2], 2D
>  9: MAD TEMP[1].xyz, TEMP[1].xyzz, IMM[0].yyyy, IMM[0].zzzz
> 10: MUL TEMP[3].xyz, TEMP[1].xxxx, IN[5].xyzz
> 11: MAD TEMP[3].xyz, TEMP[1].yyyy, IN[4].xyzz, TEMP[3].xyzz
> 12: MAD TEMP[1].xyz, TEMP[1].zzzz, IN[6].xyzz, TEMP[3].xyzz
> 13: DP3 TEMP[3].x, TEMP[1].xyzz, TEMP[1].xyzz
> 14: RSQ TEMP[3].x, TEMP[3].xxxx
> 15: MUL TEMP[1].xyz, TEMP[1].xyzz, TEMP[3].xxxx
> 16: RCP TEMP[3].x, IN[2].zzzz
> 17: MIN TEMP[3].x, TEMP[3].xxxx, IMM[0].wwww
> 18: MUL TEMP[3].x, CONST[6].xxxx, TEMP[3].xxxx
> 19: MAD TEMP[3].xy, TEMP[1].xyyy, TEMP[3].xxxx, TEMP[0].xyyy
> 20: TEX TEMP[3].xyz, TEMP[3].xyyy, SAMP[1], RECT
> 21: MUL TEMP[2].xyz, TEMP[2].xyzz, TEMP[3].xyzz
> 22: DP3 TEMP[3].x, IN[2].xyzz, IN[2].xyzz
> 23: RSQ TEMP[3].x, TEMP[3].xxxx
> 24: MUL TEMP[3].xyz, IN[2].xyzz, TEMP[3].xxxx
> 25: DP3 TEMP[5].x, TEMP[1].xyzz, TEMP[3].xyzz
> 26: MUL TEMP[5].xyz, TEMP[5].xxxx, TEMP[1].xyzz
> 27: MUL TEMP[5].xyz, IMM[0].yyyy, TEMP[5].xyzz
> 28: ADD TEMP[4].xyz, TEMP[3].xyzz, -TEMP[5].xyzz
> 29: MUL TEMP[5], CONST[2], TEMP[4].xxxx
> 30: MAD TEMP[5], CONST[3], TEMP[4].yyyy, TEMP[5]
> 31: MAD TEMP[4], CONST[4], TEMP[4].zzzz, TEMP[5]
> 32: ADD TEMP[4].xyz, TEMP[4], CONST[5]
> 33: TEX TEMP[4].xyz, TEMP[4].xyzz, SAMP[0], CUBE
> 34: ADD TEMP[5].x, IMM[0].xxxx, -CONST[1].xxxx
> 35: DP3 TEMP[3].x, -TEMP[3].xyzz, TEMP[1].xyzz
> 36: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 37: ADD TEMP[3].x, IMM[0].xxxx, -TEMP[3].xxxx
> 38: POW TEMP[3].x, TEMP[3].xxxx, CONST[1].yyyy
> 39: MAD TEMP[3].x, TEMP[5].xxxx, TEMP[3].xxxx, CONST[1].xxxx
> 40: MAX TEMP[3].x, TEMP[3].xxxx, IMM[1].xxxx
> 41: MUL TEMP[3].xyz, TEMP[4].xyzz, TEMP[3].xxxx
> 42: MUL TEMP[3].xyz, TEMP[3].xyzz, CONST[8].xxxx
> 43: MAD TEMP[2].xyz, TEMP[3].xyzz, CONST[7].xxxx, TEMP[2].xyzz
> 44: DP3 TEMP[1].x, TEMP[1].xyzz, IMM[1].xxyy
> 45: ABS TEMP[1].x, TEMP[1].xxxx
> 46: ADD TEMP[1].x, IMM[0].xxxx, -TEMP[1].xxxx
> 47: POW TEMP[1].x, TEMP[1].xxxx, CONST[0].yyyy
> 48: MUL TEMP[1].x, TEMP[1].xxxx, CONST[0].xxxx
> 49: MUL TEMP[1].xyz, TEMP[4].xyzz, TEMP[1].xxxx
> 50: MUL TEMP[1].xyz, TEMP[1].xyzz, CONST[8].xxxx
> 51: MAD TEMP[2].xyz, TEMP[1].xyzz, CONST[7].xxxx, TEMP[2].xyzz
> 52: MOV OUT[0], TEMP[2]
> 53: END
>Fragment Program: before compilation
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'rewrite depth out'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform KILP'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform loops'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'emulate branches'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[1].xyz, input[1].xyyy, 2D[2];
> 12: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 13: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 14: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 15: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 16: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 17: RSQ temp[3].x, temp[3].xxxx;
> 18: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 19: RCP temp[3].x, input[2].zzzz;
> 20: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 21: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 22: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 23: TEX temp[3].xyz, temp[3].xyyy, RECT[1];
> 24: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 25: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 26: RSQ temp[3].x, temp[3].xxxx;
> 27: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 28: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 29: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 30: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 31: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 32: MUL temp[5], const[2], temp[4].xxxx;
> 33: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 34: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 35: ADD temp[4].xyz, temp[4], const[5];
> 36: TEX temp[4].xyz, temp[4].xyzz, CUBE[0];
> 37: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 38: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 39: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 40: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 41: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 42: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 45: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 46: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 47: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 48: ABS temp[1].x, temp[1].xxxx;
> 49: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 50: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 51: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 52: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 53: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 54: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 55: MOV output[0], temp[2];
>Fragment Program: after 'transform TEX'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[7], input[1].xyyy, 2D[2];
> 12: MOV temp[1].xyz, temp[7];
> 13: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 14: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 15: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 16: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 17: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 18: RSQ temp[3].x, temp[3].xxxx;
> 19: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 20: RCP temp[3].x, input[2].zzzz;
> 21: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 22: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 23: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 24: MUL temp[8], temp[3].xyyy, const[18];
> 25: TEX temp[9], temp[8], 2D[1];
> 26: MOV temp[3].xyz, temp[9];
> 27: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 28: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 29: RSQ temp[3].x, temp[3].xxxx;
> 30: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 31: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 32: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 33: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 34: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 35: MUL temp[5], const[2], temp[4].xxxx;
> 36: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 37: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 38: ADD temp[4].xyz, temp[4], const[5];
> 39: TEX temp[10], temp[4].xyzz, CUBE[0];
> 40: MOV temp[4].xyz, temp[10];
> 41: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 42: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 45: POW temp[3].x, temp[3].xxxx, const[1].yyyy;
> 46: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 47: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 48: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 49: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 50: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 51: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 52: ABS temp[1].x, temp[1].xxxx;
> 53: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 54: POW temp[1].x, temp[1].xxxx, const[0].yyyy;
> 55: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 56: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 57: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 58: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 59: MOV output[0], temp[2];
>Fragment Program: after 'native rewrite'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].wwww;
>  1: MUL temp[6].xyz, input[0], temp[6].wwww;
>  2: MAD temp[6].xyz, temp[6].xyz0, const[16].xyz0, const[17].xyz0;
>  3: MOV temp[0], temp[6];
>  4: MAD temp[0].y, temp[6], const[13].xxxx, const[13].yyyy;
>  5: TEX temp[1], input[1].xyyy, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].wwww;
>  8: ADD temp[3].xyz, const[14].xxxx, -temp[1].xyzz;
>  9: ADD temp[4].x, const[14].xxxx, -const[8].xxxx;
> 10: MAD temp[2].xyz, temp[3].xyzz, temp[4].xxxx, temp[1].xyzz;
> 11: TEX temp[7], input[1].xyyy, 2D[2];
> 12: MOV temp[1].xyz, temp[7];
> 13: MAD temp[1].xyz, temp[1].xyzz, const[14].yyyy, const[14].zzzz;
> 14: MUL temp[3].xyz, temp[1].xxxx, input[5].xyzz;
> 15: MAD temp[3].xyz, temp[1].yyyy, input[4].xyzz, temp[3].xyzz;
> 16: MAD temp[1].xyz, temp[1].zzzz, input[6].xyzz, temp[3].xyzz;
> 17: DP3 temp[3].x, temp[1].xyzz, temp[1].xyzz;
> 18: RSQ temp[3].x, |temp[3].xxxx|;
> 19: MUL temp[1].xyz, temp[1].xyzz, temp[3].xxxx;
> 20: RCP temp[3].x, input[2].zzzz;
> 21: MIN temp[3].x, temp[3].xxxx, const[14].wwww;
> 22: MUL temp[3].x, const[6].xxxx, temp[3].xxxx;
> 23: MAD temp[3].xy, temp[1].xyyy, temp[3].xxxx, temp[0].xyyy;
> 24: MUL temp[8], temp[3].xyyy, const[18];
> 25: TEX temp[9], temp[8], 2D[1];
> 26: MOV temp[3].xyz, temp[9];
> 27: MUL temp[2].xyz, temp[2].xyzz, temp[3].xyzz;
> 28: DP3 temp[3].x, input[2].xyzz, input[2].xyzz;
> 29: RSQ temp[3].x, |temp[3].xxxx|;
> 30: MUL temp[3].xyz, input[2].xyzz, temp[3].xxxx;
> 31: DP3 temp[5].x, temp[1].xyzz, temp[3].xyzz;
> 32: MUL temp[5].xyz, temp[5].xxxx, temp[1].xyzz;
> 33: MUL temp[5].xyz, const[14].yyyy, temp[5].xyzz;
> 34: ADD temp[4].xyz, temp[3].xyzz, -temp[5].xyzz;
> 35: MUL temp[5], const[2], temp[4].xxxx;
> 36: MAD temp[5], const[3], temp[4].yyyy, temp[5];
> 37: MAD temp[4], const[4], temp[4].zzzz, temp[5];
> 38: ADD temp[4].xyz, temp[4], const[5];
> 39: TEX temp[10], temp[4].xyzz, CUBE[0];
> 40: MOV temp[4].xyz, temp[10];
> 41: ADD temp[5].x, const[14].xxxx, -const[1].xxxx;
> 42: DP3 temp[3].x, -temp[3].xyzz, temp[1].xyzz;
> 43: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 44: ADD temp[3].x, const[14].xxxx, -temp[3].xxxx;
> 45: LG2 temp[11].w, temp[3].xxxx;
> 46: MUL temp[11].w, temp[11].wwww, const[1].yyyy;
> 47: EX2 temp[3].x, temp[11].wwww;
> 48: MAD temp[3].x, temp[5].xxxx, temp[3].xxxx, const[1].xxxx;
> 49: MAX temp[3].x, temp[3].xxxx, const[15].xxxx;
> 50: MUL temp[3].xyz, temp[4].xyzz, temp[3].xxxx;
> 51: MUL temp[3].xyz, temp[3].xyzz, const[8].xxxx;
> 52: MAD temp[2].xyz, temp[3].xyzz, const[7].xxxx, temp[2].xyzz;
> 53: DP3 temp[1].x, temp[1].xyzz, const[15].xxyy;
> 54: MOV temp[1].x, |temp[1].xxxx|;
> 55: ADD temp[1].x, const[14].xxxx, -temp[1].xxxx;
> 56: LG2 temp[12].w, temp[1].xxxx;
> 57: MUL temp[12].w, temp[12].wwww, const[0].yyyy;
> 58: EX2 temp[1].x, temp[12].wwww;
> 59: MUL temp[1].x, temp[1].xxxx, const[0].xxxx;
> 60: MUL temp[1].xyz, temp[4].xyzz, temp[1].xxxx;
> 61: MUL temp[1].xyz, temp[1].xyzz, const[8].xxxx;
> 62: MAD temp[2].xyz, temp[1].xyzz, const[7].xxxx, temp[2].xyzz;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'deadcode'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[6].ww__;
>  2: MAD temp[6].xy, temp[6].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[0].x, temp[6].x___;
>  4: MAD temp[0].y, temp[6]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].___w;
>  8: ADD temp[3].xyz, const[14].xxx_, -temp[1].xyz_;
>  9: ADD temp[4].x, const[14].x___, -const[8].x___;
> 10: MAD temp[2].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 11: TEX temp[7].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[1].xyz, temp[7].xyz_;
> 13: MAD temp[1].xyz, temp[1].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[3].xyz, temp[1].xxx_, input[5].xyz_;
> 15: MAD temp[3].xyz, temp[1].yyy_, input[4].xyz_, temp[3].xyz_;
> 16: MAD temp[1].xyz, temp[1].zzz_, input[6].xyz_, temp[3].xyz_;
> 17: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 18: RSQ temp[3].x, |temp[3].x___|;
> 19: MUL temp[1].xyz, temp[1].xyz_, temp[3].xxx_;
> 20: RCP temp[3].x, input[2].z___;
> 21: MIN temp[3].x, temp[3].x___, const[14].w___;
> 22: MUL temp[3].x, const[6].x___, temp[3].x___;
> 23: MAD temp[3].xy, temp[1].xy__, temp[3].xx__, temp[0].xy__;
> 24: MUL temp[8].xy, temp[3].xy__, const[18].xy__;
> 25: TEX temp[9].xyz, temp[8].xy__, 2D[1];
> 26: MOV temp[3].xyz, temp[9].xyz_;
> 27: MUL temp[2].xyz, temp[2].xyz_, temp[3].xyz_;
> 28: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[3].x, |temp[3].x___|;
> 30: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 31: DP3 temp[5].x, temp[1].xyz_, temp[3].xyz_;
> 32: MUL temp[5].xyz, temp[5].xxx_, temp[1].xyz_;
> 33: MUL temp[5].xyz, const[14].yyy_, temp[5].xyz_;
> 34: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 35: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 36: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 37: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 38: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 39: TEX temp[10].xyz, temp[4].xyz_, CUBE[0];
> 40: MOV temp[4].xyz, temp[10].xyz_;
> 41: ADD temp[5].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[3].x, -temp[3].xyz_, temp[1].xyz_;
> 43: MAX temp[3].x, temp[3].x___, const[15].x___;
> 44: ADD temp[3].x, const[14].x___, -temp[3].x___;
> 45: LG2 temp[11].w, temp[3].___x;
> 46: MUL temp[11].w, temp[11].___w, const[1].___y;
> 47: EX2 temp[3].x, temp[11].w___;
> 48: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 49: MAX temp[3].x, temp[3].x___, const[15].x___;
> 50: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 51: MUL temp[3].xyz, temp[3].xyz_, const[8].xxx_;
> 52: MAD temp[2].xyz, temp[3].xyz_, const[7].xxx_, temp[2].xyz_;
> 53: DP3 temp[1].x, temp[1].xyz_, const[15].xxy_;
> 54: MOV temp[1].x, |temp[1].x___|;
> 55: ADD temp[1].x, const[14].x___, -temp[1].x___;
> 56: LG2 temp[12].w, temp[1].___x;
> 57: MUL temp[12].w, temp[12].___w, const[0].___y;
> 58: EX2 temp[1].x, temp[12].w___;
> 59: MUL temp[1].x, temp[1].x___, const[0].x___;
> 60: MUL temp[1].xyz, temp[4].xyz_, temp[1].xxx_;
> 61: MUL temp[1].xyz, temp[1].xyz_, const[8].xxx_;
> 62: MAD temp[2].xyz, temp[1].xyz_, const[7].xxx_, temp[2].xyz_;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'emulate loops'
># Radeon Compiler Program
>  0: RCP temp[6].w, input[0].___w;
>  1: MUL temp[6].xy, input[0].xy__, temp[6].ww__;
>  2: MAD temp[6].xy, temp[6].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[0].x, temp[6].x___;
>  4: MAD temp[0].y, temp[6]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[1], input[1].xy__, 2D[3];
>  6: MUL temp[1], temp[1], input[3];
>  7: MOV temp[2].w, temp[1].___w;
>  8: ADD temp[3].xyz, const[14].xxx_, -temp[1].xyz_;
>  9: ADD temp[4].x, const[14].x___, -const[8].x___;
> 10: MAD temp[2].xyz, temp[3].xyz_, temp[4].xxx_, temp[1].xyz_;
> 11: TEX temp[7].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[1].xyz, temp[7].xyz_;
> 13: MAD temp[1].xyz, temp[1].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[3].xyz, temp[1].xxx_, input[5].xyz_;
> 15: MAD temp[3].xyz, temp[1].yyy_, input[4].xyz_, temp[3].xyz_;
> 16: MAD temp[1].xyz, temp[1].zzz_, input[6].xyz_, temp[3].xyz_;
> 17: DP3 temp[3].x, temp[1].xyz_, temp[1].xyz_;
> 18: RSQ temp[3].x, |temp[3].x___|;
> 19: MUL temp[1].xyz, temp[1].xyz_, temp[3].xxx_;
> 20: RCP temp[3].x, input[2].z___;
> 21: MIN temp[3].x, temp[3].x___, const[14].w___;
> 22: MUL temp[3].x, const[6].x___, temp[3].x___;
> 23: MAD temp[3].xy, temp[1].xy__, temp[3].xx__, temp[0].xy__;
> 24: MUL temp[8].xy, temp[3].xy__, const[18].xy__;
> 25: TEX temp[9].xyz, temp[8].xy__, 2D[1];
> 26: MOV temp[3].xyz, temp[9].xyz_;
> 27: MUL temp[2].xyz, temp[2].xyz_, temp[3].xyz_;
> 28: DP3 temp[3].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[3].x, |temp[3].x___|;
> 30: MUL temp[3].xyz, input[2].xyz_, temp[3].xxx_;
> 31: DP3 temp[5].x, temp[1].xyz_, temp[3].xyz_;
> 32: MUL temp[5].xyz, temp[5].xxx_, temp[1].xyz_;
> 33: MUL temp[5].xyz, const[14].yyy_, temp[5].xyz_;
> 34: ADD temp[4].xyz, temp[3].xyz_, -temp[5].xyz_;
> 35: MUL temp[5].xyz, const[2].xyz_, temp[4].xxx_;
> 36: MAD temp[5].xyz, const[3].xyz_, temp[4].yyy_, temp[5].xyz_;
> 37: MAD temp[4].xyz, const[4].xyz_, temp[4].zzz_, temp[5].xyz_;
> 38: ADD temp[4].xyz, temp[4].xyz_, const[5].xyz_;
> 39: TEX temp[10].xyz, temp[4].xyz_, CUBE[0];
> 40: MOV temp[4].xyz, temp[10].xyz_;
> 41: ADD temp[5].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[3].x, -temp[3].xyz_, temp[1].xyz_;
> 43: MAX temp[3].x, temp[3].x___, const[15].x___;
> 44: ADD temp[3].x, const[14].x___, -temp[3].x___;
> 45: LG2 temp[11].w, temp[3].___x;
> 46: MUL temp[11].w, temp[11].___w, const[1].___y;
> 47: EX2 temp[3].x, temp[11].w___;
> 48: MAD temp[3].x, temp[5].x___, temp[3].x___, const[1].x___;
> 49: MAX temp[3].x, temp[3].x___, const[15].x___;
> 50: MUL temp[3].xyz, temp[4].xyz_, temp[3].xxx_;
> 51: MUL temp[3].xyz, temp[3].xyz_, const[8].xxx_;
> 52: MAD temp[2].xyz, temp[3].xyz_, const[7].xxx_, temp[2].xyz_;
> 53: DP3 temp[1].x, temp[1].xyz_, const[15].xxy_;
> 54: MOV temp[1].x, |temp[1].x___|;
> 55: ADD temp[1].x, const[14].x___, -temp[1].x___;
> 56: LG2 temp[12].w, temp[1].___x;
> 57: MUL temp[12].w, temp[12].___w, const[0].___y;
> 58: EX2 temp[1].x, temp[12].w___;
> 59: MUL temp[1].x, temp[1].x___, const[0].x___;
> 60: MUL temp[1].xyz, temp[4].xyz_, temp[1].xxx_;
> 61: MUL temp[1].xyz, temp[1].xyz_, const[8].xxx_;
> 62: MAD temp[2].xyz, temp[1].xyz_, const[7].xxx_, temp[2].xyz_;
> 63: MOV output[0], temp[2];
>Fragment Program: after 'register rename'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, const[14].xxx_, -temp[18].xyz_;
>  9: ADD temp[21].x, const[14].x___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MOV temp[24].xyz, temp[23].xyz_;
> 13: MAD temp[25].xyz, temp[24].xyz_, const[14].yyy_, const[14].zzz_;
> 14: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 15: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 16: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 17: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 18: RSQ temp[30].x, |temp[29].x___|;
> 19: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 20: RCP temp[32].x, input[2].z___;
> 21: MIN temp[33].x, temp[32].x___, const[14].w___;
> 22: MUL temp[34].x, const[6].x___, temp[33].x___;
> 23: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 24: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 25: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 26: MOV temp[38].xyz, temp[37].xyz_;
> 27: MUL temp[39].xyz, temp[22].xyz_, temp[38].xyz_;
> 28: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 29: RSQ temp[41].x, |temp[40].x___|;
> 30: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 31: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 32: MUL temp[44].xyz, temp[43].xxx_, temp[31].xyz_;
> 33: MUL temp[45].xyz, const[14].yyy_, temp[44].xyz_;
> 34: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 35: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 36: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 37: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 38: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 39: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 40: MOV temp[52].xyz, temp[51].xyz_;
> 41: ADD temp[53].x, const[14].x___, -const[1].x___;
> 42: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 43: MAX temp[55].x, temp[54].x___, const[15].x___;
> 44: ADD temp[56].x, const[14].x___, -temp[55].x___;
> 45: LG2 temp[57].w, temp[56].___x;
> 46: MUL temp[58].w, temp[57].___w, const[1].___y;
> 47: EX2 temp[59].x, temp[58].w___;
> 48: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 49: MAX temp[61].x, temp[60].x___, const[15].x___;
> 50: MUL temp[62].xyz, temp[52].xyz_, temp[61].xxx_;
> 51: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 52: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 53: DP3 temp[65].x, temp[31].xyz_, const[15].xxy_;
> 54: MOV temp[66].x, |temp[65].x___|;
> 55: ADD temp[67].x, const[14].x___, -temp[66].x___;
> 56: LG2 temp[68].w, temp[67].___x;
> 57: MUL temp[69].w, temp[68].___w, const[0].___y;
> 58: EX2 temp[70].x, temp[69].w___;
> 59: MUL temp[71].x, temp[70].x___, const[0].x___;
> 60: MUL temp[72].xyz, temp[52].xyz_, temp[71].xxx_;
> 61: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 62: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 63: MOV output[0], temp[19];
>Fragment Program: after 'dataflow optimize'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[14].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[14].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, none.00-1_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>CONST[14] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[15] = {     0.0000    -1.0000     0.0000     0.0000 }
>CONST[19] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dataflow swizzles'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[16].xy__, const[17].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[13]._x__, const[13]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[14].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[14].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[18].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, const[19].xyz_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>CONST[10] = {     1.0000     2.0000    -1.0000    10.0000 }
>CONST[14] = {     0.0000     0.0000    -1.0000     0.0000 }
>Fragment Program: after 'dead constants'
># Radeon Compiler Program
>  0: RCP temp[13].w, input[0].___w;
>  1: MUL temp[14].xy, input[0].xy__, temp[13].ww__;
>  2: MAD temp[15].xy, temp[14].xy__, const[11].xy__, const[12].xy__;
>  3: MOV temp[16].x, temp[15].x___;
>  4: MAD temp[16].y, temp[15]._y__, const[9]._x__, const[9]._y__;
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: MUL temp[18], temp[17], input[3];
>  7: MOV temp[19].w, temp[18].___w;
>  8: ADD temp[20].xyz, none.111_, -temp[18].xyz_;
>  9: ADD temp[21].x, none.1___, -const[8].x___;
> 10: MAD temp[22].xyz, temp[20].xyz_, temp[21].xxx_, temp[18].xyz_;
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: MAD temp[25].xyz, temp[23].xyz_, const[10].yyy_, none.-1-1-1_;
> 13: MUL temp[26].xyz, temp[25].xxx_, input[5].xyz_;
> 14: MAD temp[27].xyz, temp[25].yyy_, input[4].xyz_, temp[26].xyz_;
> 15: MAD temp[28].xyz, temp[25].zzz_, input[6].xyz_, temp[27].xyz_;
> 16: DP3 temp[29].x, temp[28].xyz_, temp[28].xyz_;
> 17: RSQ temp[30].x, |temp[29].x___|;
> 18: MUL temp[31].xyz, temp[28].xyz_, temp[30].xxx_;
> 19: RCP temp[32].x, input[2].z___;
> 20: MIN temp[33].x, temp[32].x___, const[10].w___;
> 21: MUL temp[34].x, const[6].x___, temp[33].x___;
> 22: MAD temp[35].xy, temp[31].xy__, temp[34].xx__, temp[16].xy__;
> 23: MUL temp[36].xy, temp[35].xy__, const[13].xy__;
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: MUL temp[39].xyz, temp[22].xyz_, temp[37].xyz_;
> 26: DP3 temp[40].x, input[2].xyz_, input[2].xyz_;
> 27: RSQ temp[41].x, |temp[40].x___|;
> 28: MUL temp[42].xyz, input[2].xyz_, temp[41].xxx_;
> 29: DP3 temp[43].x, temp[31].xyz_, temp[42].xyz_;
> 30: MUL temp[45].xyz * 2, temp[43].xxx_, temp[31].xyz_;
> 31: ADD temp[46].xyz, temp[42].xyz_, -temp[45].xyz_;
> 32: MUL temp[47].xyz, const[2].xyz_, temp[46].xxx_;
> 33: MAD temp[48].xyz, const[3].xyz_, temp[46].yyy_, temp[47].xyz_;
> 34: MAD temp[49].xyz, const[4].xyz_, temp[46].zzz_, temp[48].xyz_;
> 35: ADD temp[50].xyz, temp[49].xyz_, const[5].xyz_;
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: ADD temp[53].x, none.1___, -const[1].x___;
> 38: DP3 temp[54].x, -temp[42].xyz_, temp[31].xyz_;
> 39: MAX temp[55].x, temp[54].x___, none.0___;
> 40: ADD temp[56].x, none.1___, -temp[55].x___;
> 41: LG2 temp[57].w, temp[56].___x;
> 42: MUL temp[58].w, temp[57].___w, const[1].___y;
> 43: EX2 temp[59].x, temp[58].w___;
> 44: MAD temp[60].x, temp[53].x___, temp[59].x___, const[1].x___;
> 45: MAX temp[61].x, temp[60].x___, none.0___;
> 46: MUL temp[62].xyz, temp[51].xyz_, temp[61].xxx_;
> 47: MUL temp[63].xyz, temp[62].xyz_, const[8].xxx_;
> 48: MAD temp[64].xyz, temp[63].xyz_, const[7].xxx_, temp[39].xyz_;
> 49: DP3 temp[65].x, temp[31].xyz_, const[14].xyz_;
> 50: ADD temp[67].x, none.1___, -|temp[65].x___|;
> 51: LG2 temp[68].w, temp[67].___x;
> 52: MUL temp[69].w, temp[68].___w, const[0].___y;
> 53: EX2 temp[70].x, temp[69].w___;
> 54: MUL temp[71].x, temp[70].x___, const[0].x___;
> 55: MUL temp[72].xyz, temp[51].xyz_, temp[71].xxx_;
> 56: MUL temp[73].xyz, temp[72].xyz_, const[8].xxx_;
> 57: MAD temp[19].xyz, temp[73].xyz_, const[7].xxx_, temp[64].xyz_;
> 58: MOV output[0], temp[19];
>Fragment Program: after 'pair translate'
># Radeon Compiler Program
>  0: src0.w = input[0]
>     RCP temp[13].w, src0.w
>  1: src0.xyz = input[0], src0.w = temp[13]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>  2: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  3: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  4: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  5: TEX temp[17], input[1].xy__, 2D[3];
>  6: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3]
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
>  7: src0.w = temp[18]
>     MAD temp[19].w, src0.w, src0.1, src0.0
>  8: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
>  9: src0.xyz = const[8]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
> 10: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 11: TEX temp[23].xyz, input[1].xy__, 2D[2];
> 12: src0.xyz = temp[23], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
> 13: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 14: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 15: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 16: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 17: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 18: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 19: src0.xyz = input[2]
>     REPL_ALPHA temp[32].x
>     RCP, src0.z
> 20: src0.xyz = temp[32], src0.w = const[10]
>     MIN temp[33].x, src0.x__, src0.w__
> 21: src0.xyz = const[6], src1.xyz = temp[33]
>     MAD temp[34].x, src0.x__, src1.x__, src0.000
> 22: src0.xyz = temp[31], src1.xyz = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src1.xx_, src2.xy_
> 23: src0.xyz = temp[35], src1.xyz = const[13]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
> 24: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 25: src0.xyz = temp[22], src1.xyz = temp[37]
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 26: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
> 27: src0.xyz = temp[40]
>     REPL_ALPHA temp[41].x
>     RSQ, |src0.x|
> 28: src0.xyz = input[2], src1.xyz = temp[41]
>     MAD temp[42].xyz, src0.xyz, src1.xxx, src0.000
> 29: src0.xyz = temp[31], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
> 30: src0.xyz = temp[43], src1.xyz = temp[31]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
> 31: src0.xyz = temp[42], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
> 32: src0.xyz = const[2], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
> 33: src0.xyz = const[3], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
> 34: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 35: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 36: TEX temp[51].xyz, temp[50].xyz_, CUBE[0];
> 37: src0.xyz = const[1]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
> 38: src0.xyz = temp[42], src1.xyz = temp[31]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
> 39: src0.xyz = temp[54]
>     MAX temp[55].x, src0.x__, src0.0__
> 40: src0.xyz = temp[55]
>     MAD temp[56].x, src0.1__, src0.111, -src0.x__
> 41: src0.xyz = temp[56]
>     LG2 temp[57].w, src0.x
> 42: src0.xyz = const[1], src0.w = temp[57]
>     MAD temp[58].w, src0.w, src0.y, src0.0
> 43: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 44: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].x, src0.x__, src1.x__, src2.x__
> 45: src0.xyz = temp[60]
>     MAX temp[61].x, src0.x__, src0.0__
> 46: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 47: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 48: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 49: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 50: src0.xyz = temp[65]
>     MAD temp[67].x, src0.1__, |src0.111|, -|src0.x__|
> 51: src0.xyz = temp[67]
>     LG2 temp[68].w, src0.x
> 52: src0.xyz = const[0], src0.w = temp[68]
>     MAD temp[69].w, src0.w, src0.y, src0.0
> 53: src0.w = temp[69]
>     REPL_ALPHA temp[70].x
>     EX2, src0.w
> 54: src0.xyz = temp[70], src1.xyz = const[0]
>     MAD temp[71].x, src0.x__, src1.x__, src0.000
> 55: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 56: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 57: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 58: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'pair scheduling'
># Radeon Compiler Program
>  0: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
>     RCP temp[32].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[40]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
>     RSQ temp[41].w, |src1.x|
>  2: src0.xyz = input[2], src0.w = temp[41], src1.xyz = temp[41], src1.w = temp[32], src2.w = const[10]
>     MAD temp[42].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[33].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[0]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
>     RCP temp[13].w, src0.w
>  4: src0.xyz = input[0], src0.w = temp[13], src1.xyz = const[6], src1.w = temp[33]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[34].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[17], input[1].xy__, 2D[3];
> 10: TEX temp[23].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[23], src0.w = temp[18], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[19].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 16: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[31], src0.w = temp[34], src1.xyz = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[42], src1.xyz = temp[31], src2.xyz = temp[65]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
>     MAD temp[67].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[31], src0.w = temp[67], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
>     LG2 temp[68].w, src0.w
> 25: src0.xyz = temp[35], src1.xyz = const[13], src2.xyz = temp[54]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
>     MAX temp[55].w, src2.x, src0.0
> 26: src0.xyz = temp[43], src0.w = temp[68], src1.xyz = temp[31], src2.xyz = const[0]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[69].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[42], src0.w = temp[55], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[56].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[56], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[57].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[69], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[70].w, src0.w
> 30: src0.xyz = temp[70], src0.w = temp[70], src1.xyz = const[0], src1.w = temp[57], src2.xyz = const[1]
>     MAD temp[71].x, src0.w__, src1.x__, src0.000
>     MAD temp[58].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 33: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].w, src0.x, src1.x, src2.x
> 35: src0.xyz = temp[60], src0.w = temp[60]
>     MAX temp[61].x, src0.w__, src0.0__
> 36: BEGIN_TEX;
> 37: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 38: TEX temp[51].xyz, temp[50].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[22], src1.xyz = temp[37] SEM_WAIT
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 42: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'dead sources'
># Radeon Compiler Program
>  0: src0.xyz = input[2]
>     DP3 temp[40].x, src0.xyz, src0.xyz
>     RCP temp[32].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[40]
>     MAD temp[53].x, src0.1__, src0.111, -src0.x__
>     RSQ temp[41].w, |src1.x|
>  2: src0.xyz = input[2], src0.w = temp[41], src1.w = temp[32], src2.w = const[10]
>     MAD temp[42].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[33].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[0]
>     MAD temp[21].x, src0.1__, src0.111, -src0.x__
>     RCP temp[13].w, src0.w
>  4: src0.xyz = input[0], src0.w = temp[13], src1.xyz = const[6], src1.w = temp[33]
>     MAD temp[14].xy, src0.xy_, src0.ww_, src0.000
>     MAD temp[34].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[14], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[15].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[15], src1.xyz = const[9]
>     MAD temp[16].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[15]
>     MAD temp[16].x, src0.x__, src0.111, src0.000
>  8: BEGIN_TEX;
>  9: TEX temp[17], input[1].xy__, 2D[3];
> 10: TEX temp[23].xyz, input[1].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[17], src0.w = temp[17], src1.xyz = input[3], src1.w = input[3] SEM_WAIT
>     MAD temp[18].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[18].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[23], src0.w = temp[18], src1.xyz = const[10]
>     MAD temp[25].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[19].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[18]
>     MAD temp[20].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[25], src1.xyz = input[5]
>     MAD temp[26].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[20], src1.xyz = temp[21], src2.xyz = temp[18]
>     MAD temp[22].xyz, src0.xyz, src1.xxx, src2.xyz
> 16: src0.xyz = temp[25], src1.xyz = input[4], src2.xyz = temp[26]
>     MAD temp[27].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[25], src1.xyz = input[6], src2.xyz = temp[27]
>     MAD temp[28].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[28]
>     DP3 temp[29].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[29]
>     REPL_ALPHA temp[30].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[28], src1.xyz = temp[30]
>     MAD temp[31].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[31], src0.w = temp[34], src2.xyz = temp[16]
>     MAD temp[35].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[31], src1.xyz = const[14]
>     DP3 temp[65].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[42], src1.xyz = temp[31], src2.xyz = temp[65]
>     DP3 temp[54].x, -src0.xyz, src1.xyz
>     MAD temp[67].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[31], src0.w = temp[67], src1.xyz = temp[42]
>     DP3 temp[43].x, src0.xyz, src1.xyz
>     LG2 temp[68].w, src0.w
> 25: src0.xyz = temp[35], src1.xyz = const[13], src2.xyz = temp[54]
>     MAD temp[36].xy, src0.xy_, src1.xy_, src0.000
>     MAX temp[55].w, src2.x, src0.0
> 26: src0.xyz = temp[43], src0.w = temp[68], src1.xyz = temp[31], src2.xyz = const[0]
>     MAD temp[45].xyz * 2, src0.xxx, src1.xyz, src0.000
>     MAD temp[69].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[42], src0.w = temp[55], src1.xyz = temp[45]
>     MAD temp[46].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[56].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[56], src1.xyz = temp[46]
>     MAD temp[47].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[57].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[69], src1.xyz = temp[46], src2.xyz = temp[47]
>     MAD temp[48].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[70].w, src0.w
> 30: src0.w = temp[70], src1.xyz = const[0], src1.w = temp[57], src2.xyz = const[1]
>     MAD temp[71].x, src0.w__, src1.x__, src0.000
>     MAD temp[58].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[46], src2.xyz = temp[48]
>     MAD temp[49].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[58]
>     REPL_ALPHA temp[59].x
>     EX2, src0.w
> 33: src0.xyz = temp[49], src1.xyz = const[5]
>     MAD temp[50].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[53], src1.xyz = temp[59], src2.xyz = const[1]
>     MAD temp[60].w, src0.x, src1.x, src2.x
> 35: src0.w = temp[60]
>     MAX temp[61].x, src0.w__, src0.0__
> 36: BEGIN_TEX;
> 37: TEX temp[37].xyz, temp[36].xy__, 2D[1];
> 38: TEX temp[51].xyz, temp[50].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[22], src1.xyz = temp[37] SEM_WAIT
>     MAD temp[39].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[51], src1.xyz = temp[71]
>     MAD temp[72].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[51], src1.xyz = temp[61]
>     MAD temp[62].xyz, src0.xyz, src1.xxx, src0.000
> 42: src0.xyz = temp[72], src1.xyz = const[8]
>     MAD temp[73].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[62], src1.xyz = const[8]
>     MAD temp[63].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[63], src1.xyz = const[7], src2.xyz = temp[39]
>     MAD temp[64].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[73], src1.xyz = const[7], src2.xyz = temp[64]
>     MAD temp[19].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[19], src0.w = temp[19]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>Fragment Program: after 'register allocation'
># Radeon Compiler Program
>  0: src0.xyz = input[1]
>     DP3 temp[0].z, src0.xyz, src0.xyz
>     RCP temp[0].w, src0.z
>  1: src0.xyz = const[1], src1.xyz = temp[0]
>     MAD temp[0].z, src0.__1, src0.__1, -src0.__x
>     RSQ temp[1].w, |src1.z|
>  2: src0.xyz = input[1], src0.w = temp[1], src1.w = temp[0], src2.w = const[10]
>     MAD temp[1].xyz, src0.xyz, src0.www, src0.000
>     MIN temp[0].w, src1.w, src2.w
>  3: src0.xyz = const[8], src0.w = input[6]
>     MAD temp[6].z, src0.__1, src0.__1, -src0.__x
>     RCP temp[1].w, src0.w
>  4: src0.xyz = input[6], src0.w = temp[1], src1.xyz = const[6], src1.w = temp[0]
>     MAD temp[6].xy, src0.xy_, src0.ww_, src0.00_
>     MAD temp[0].w, src1.x, src1.w, src0.0
>  5: src0.xyz = temp[6], src1.xyz = const[11], src2.xyz = const[12]
>     MAD temp[6].xy, src0.xy_, src1.xy_, src2.xy_
>  6: src0.xyz = temp[6], src1.xyz = const[9]
>     MAD temp[7].y, src0._y_, src1._x_, src1._y_
>  7: src0.xyz = temp[6]
>     MAD temp[7].x, src0.x__, src0.11_, src0.00_
>  8: BEGIN_TEX;
>  9: TEX temp[8], input[0].xy__, 2D[3];
> 10: TEX temp[9].xyz, input[0].xy__, 2D[2] SEM_WAIT SEM_ACQUIRE;
> 11: src0.xyz = temp[8], src0.w = temp[8], src1.xyz = input[2], src1.w = input[2] SEM_WAIT
>     MAD temp[2].xyz, src0.xyz, src1.xyz, src0.000
>     MAD temp[1].w, src0.w, src1.w, src0.0
> 12: src0.xyz = temp[9], src0.w = temp[1], src1.xyz = const[10]
>     MAD temp[8].xyz, src0.xyz, src1.yyy, -src0.111
>     MAD temp[1].w, src0.w, src0.1, src0.0
> 13: src0.xyz = temp[2]
>     MAD temp[9].xyz, src0.111, src0.111, -src0.xyz
> 14: src0.xyz = temp[8], src1.xyz = input[4]
>     MAD temp[4].xyz, src0.xxx, src1.xyz, src0.000
> 15: src0.xyz = temp[9], src1.xyz = temp[6], src2.xyz = temp[2]
>     MAD temp[2].xyz, src0.xyz, src1.zzz, src2.xyz
> 16: src0.xyz = temp[8], src1.xyz = input[3], src2.xyz = temp[4]
>     MAD temp[3].xyz, src0.yyy, src1.xyz, src2.xyz
> 17: src0.xyz = temp[8], src1.xyz = input[5], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.zzz, src1.xyz, src2.xyz
> 18: src0.xyz = temp[3]
>     DP3 temp[0].x, src0.xyz, src0.xyz
> 19: src0.xyz = temp[0]
>     REPL_ALPHA temp[0].x
>     RSQ, |src0.x|
> 20: src0.xyz = temp[3], src1.xyz = temp[0]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
> 21: src0.xyz = temp[3], src0.w = temp[0], src2.xyz = temp[7]
>     MAD temp[0].xy, src0.xy_, src0.ww_, src2.xy_
> 22: src0.xyz = temp[3], src1.xyz = const[14]
>     DP3 temp[4].x, src0.xyz, src1.xyz
> 23: src0.xyz = temp[1], src1.xyz = temp[3], src2.xyz = temp[4]
>     DP3 temp[4].x, -src0.xyz, src1.xyz
>     MAD temp[0].w, src0.1, |src0.1|, -|src2.x|
> 24: src0.xyz = temp[3], src0.w = temp[0], src1.xyz = temp[1]
>     DP3 temp[4].y, src0.xyz, src1.xyz
>     LG2 temp[0].w, src0.w
> 25: src0.xyz = temp[0], src1.xyz = const[13], src2.xyz = temp[4]
>     MAD temp[0].xy, src0.xy_, src1.xy_, src0.00_
>     MAX temp[2].w, src2.x, src0.0
> 26: src0.xyz = temp[4], src0.w = temp[0], src1.xyz = temp[3], src2.xyz = const[0]
>     MAD temp[3].xyz * 2, src0.yyy, src1.xyz, src0.000
>     MAD temp[0].w, src0.w, src2.y, src0.0
> 27: src0.xyz = temp[1], src0.w = temp[2], src1.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src0.111, -src1.xyz
>     MAD temp[2].w, src0.1, src0.1, -src0.w
> 28: src0.xyz = const[2], src0.w = temp[2], src1.xyz = temp[1]
>     MAD temp[3].xyz, src0.xyz, src1.xxx, src0.000
>     LG2 temp[2].w, src0.w
> 29: src0.xyz = const[3], src0.w = temp[0], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[3].xyz, src0.xyz, src1.yyy, src2.xyz
>     EX2 temp[0].w, src0.w
> 30: src0.w = temp[0], src1.xyz = const[0], src1.w = temp[2], src2.xyz = const[1]
>     MAD temp[4].x, src0.w__, src1.x__, src0.0__
>     MAD temp[0].w, src1.w, src2.y, src0.0
> 31: src0.xyz = const[4], src1.xyz = temp[1], src2.xyz = temp[3]
>     MAD temp[1].xyz, src0.xyz, src1.zzz, src2.xyz
> 32: src0.w = temp[0]
>     REPL_ALPHA temp[3].x
>     EX2, src0.w
> 33: src0.xyz = temp[1], src1.xyz = const[5]
>     MAD temp[1].xyz, src0.xyz, src0.111, src1.xyz
> 34: src0.xyz = temp[0], src1.xyz = temp[3], src2.xyz = const[1]
>     MAD temp[0].w, src0.z, src1.x, src2.x
> 35: src0.w = temp[0]
>     MAX temp[0].z, src0.__w, src0.__0
> 36: BEGIN_TEX;
> 37: TEX temp[3].xyz, temp[0].xy__, 2D[1];
> 38: TEX temp[5].xyz, temp[1].xyz_, CUBE[0] SEM_WAIT SEM_ACQUIRE;
> 39: src0.xyz = temp[2], src1.xyz = temp[3] SEM_WAIT
>     MAD temp[1].xyz, src0.xyz, src1.xyz, src0.000
> 40: src0.xyz = temp[5], src1.xyz = temp[4]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 41: src0.xyz = temp[5], src1.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.zzz, src0.000
> 42: src0.xyz = temp[2], src1.xyz = const[8]
>     MAD temp[2].xyz, src0.xyz, src1.xxx, src0.000
> 43: src0.xyz = temp[0], src1.xyz = const[8]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src0.000
> 44: src0.xyz = temp[0], src1.xyz = const[7], src2.xyz = temp[1]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 45: src0.xyz = temp[2], src1.xyz = const[7], src2.xyz = temp[0]
>     MAD temp[0].xyz, src0.xyz, src1.xxx, src2.xyz
> 46: src0.xyz = temp[0], src0.w = temp[1]
>     MAD color[0].xyz, src0.xyz, src0.111, src0.000
>     MAD color[0].w, src0.w, src0.1, src0.0
>pc=54*************************************
>Hardware program
>----------------
>NODE 0: alu_offset: 0, tex_offset: 0, alu_end: 7, tex_end: 0  (code_addr: 000001c0)
>  0: xyz:  t1  t0  t0  bias-> t0.z                 (02000001)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t1.xyz   t1.xyz   t1.xxx    op: 00804000 
>       w:     t1.z     t1.x     t1.x    op: 05000002
>  1: xyz:  c1  t0  t0  bias-> t0.z                 (02000021)
>       w:  t0  t0  t0  bias-> t1.w                 (00840000)
>     xyz:      1.0      1.0  -c1.xxx    op: 00084a95 
>       w:   |t0.z|     c1.x     c1.x    op: 05800045
>  2: xyz:  t1 t32  t0  bias-> t1.xyz               (03840001)
>       w:  t1  t0 c10  bias-> t0.w                 (0082a001)
>     xyz:   t1.xyz   t1.www      0.0    op: 00050600 
>       w:     t0.w    c10.w     t1.x    op: 0100058a
>  3: xyz:  c8  t0  t0  bias-> t6.z                 (02180028)
>       w:  t6  t0  t0  bias-> t1.w                 (00840006)
>     xyz:      1.0      1.0  -c8.xxx    op: 00084a95 
>       w:     t6.w     c8.x     c8.x    op: 05000009
>  4: xyz:  t6  c6  t0  bias-> t6.xy                (01980986)
>       w:  t1  t0  t0  bias-> t0.w                 (00800001)
>     xyz:   t6.xyz   t1.www      0.0    op: 00050600 
>       w:     c6.x     t0.w      0.0    op: 00040503
>  5: xyz:  t6 c11 c12  bias-> t6.xy                (019acac6)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz  c11.xyz  c12.xyz    op: 00020200 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  6: xyz:  t6  c9  t0  bias-> t7.y                 (011c0a46)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz   c9.xxx   c9.xyz    op: 00010280 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>  7: xyz:  t6  t0  t0  bias-> t7.x                 (009c0006)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t6.xyz      1.0      0.0    op: 00050a80 
>       w:     t6.x     t6.x     t6.x    op: 00000000
>NODE 1: alu_offset: 8, tex_offset: 0, alu_end: 24, tex_end: 1  (code_addr: 00020608)
>  TEX:
>    TEX t8, t0, texture[3]   (00009a00)
>    TEX t9, t0, texture[2]   (00009240)
>  8: xyz:  t8  t2  t0  bias-> t2.xyz               (03880088)
>       w:  t8  t2  t0  bias-> t1.w                 (00840088)
>     xyz:   t8.xyz   t2.xyz      0.0    op: 00050200 
>       w:     t8.w     t2.w      0.0    op: 00040509
>  9: xyz:  t9 c10  t0  bias-> t8.xyz               (03a00a89)
>       w:  t1  t0  t0  bias-> t1.w                 (00840001)
>     xyz:   t9.xyz  c10.yyy     -1.0    op: 000d4300 
>       w:     t1.w      1.0      0.0    op: 00040889
> 10: xyz:  t2  t0  t0  bias-> t9.xyz               (03a40002)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:      1.0      1.0  -t2.xyz    op: 00080a95 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 11: xyz:  t8  t4  t0  bias-> t4.xyz               (03900108)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.xxx   t4.xyz      0.0    op: 00050201 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 12: xyz:  t9  t6  t2  bias-> t2.xyz               (03882189)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t9.xyz   t6.zzz   t2.xyz    op: 00020380 
>       w:     t9.x     t9.x     t9.x    op: 00000000
> 13: xyz:  t8  t3  t4  bias-> t3.xyz               (038c40c8)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.yyy   t3.xyz   t4.xyz    op: 00020202 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 14: xyz:  t8  t5  t3  bias-> t3.xyz               (038c3148)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t8.zzz   t5.xyz   t3.xyz    op: 00020203 
>       w:     t8.x     t8.x     t8.x    op: 00000000
> 15: xyz:  t3  t0  t0  bias-> t0.x                 (00800003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t3.xyz   t3.xxx    op: 00804000 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 16: xyz:  t0  t0  t0  bias-> t0.x                 (00800000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:   |t0.x|     t0.x     t0.x    op: 05800040
> 17: xyz:  t3  t0  t0  bias-> t3.xyz               (038c0003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t0.xxx      0.0    op: 00050280 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 18: xyz:  t3 t32  t7  bias-> t0.xy                (01807003)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz   t0.www   t7.xyz    op: 00020600 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 19: xyz:  t3 c14  t0  bias-> t4.x                 (00900b83)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t3.xyz  c14.xyz   t3.xxx    op: 00804200 
>       w:     t3.x     t3.x     t3.x    op: 00000000
> 20: xyz:  t1  t3  t4  bias-> t4.x                 (009040c1)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:  -t1.xyz   t3.xyz   t1.xxx    op: 00804220 
>       w:      1.0    |1.0|  -|t4.x|    op: 0019a891
> 21: xyz:  t3  t1  t0  bias-> t4.y                 (01100043)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t3.xyz   t1.xyz   t3.xxx    op: 00804200 
>       w:     t0.w     t3.x     t3.x    op: 04800009
> 22: xyz:  t0 c13  t4  bias-> t0.xy                (01804b40)
>       w:  t0  t0  t0  bias-> t2.w                 (00880000)
>     xyz:   t0.xyz  c13.xyz      0.0    op: 00050200 
>       w:     t4.x      0.0     t0.x    op: 01800806
> 23: xyz:  t4  t3  c0  bias-> t3.xyz               (038e00c4)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t4.yyy   t3.xyz      0.0    op: 08050202 
>       w:     t0.w     c0.y      0.0    op: 00040389
> 24: xyz:  t1  t3  t0  bias-> t1.xyz               (038400c1)
>       w:  t2  t0  t0  bias-> t2.w                 (00880002)
>     xyz:   t1.xyz      1.0  -t3.xyz    op: 00090a80 
>       w:      1.0      1.0    -t2.w    op: 000a4891
> 25: xyz:  c2  t1  t0  bias-> t3.xyz               (038c0062)
>       w:  t2  t0  t0  bias-> t2.w                 (00880002)
>     xyz:   c2.xyz   t1.xxx      0.0    op: 00050280 
>       w:     t2.w     c2.x     c2.x    op: 04800009
> 26: xyz:  c3  t1  t3  bias-> t3.xyz               (038c3063)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   c3.xyz   t1.yyy   t3.xyz    op: 00020300 
>       w:     t0.w     c3.x     c3.x    op: 04000009
> 27: xyz: t32  c0  c1  bias-> t4.x                 (00921800)
>       w:  t0  t2  t0  bias-> t0.w                 (00800080)
>     xyz:   t0.www   c0.xyz      0.0    op: 0005020c 
>       w:     t2.w     c1.y      0.0    op: 0004038a
> 28: xyz:  c4  t1  t3  bias-> t1.xyz               (03843064)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   c4.xyz   t1.zzz   t3.xyz    op: 00020380 
>       w:     c4.x     c4.x     c4.x    op: 00000000
> 29: xyz:  t0  t0  t0  bias-> t3.x                 (008c0000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 05004081 
>       w:     t0.w     t0.x     t0.x    op: 04000009
> 30: xyz:  t1  c5  t0  bias-> t1.xyz               (03840941)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t1.xyz      1.0   c5.xyz    op: 00010a80 
>       w:     t1.x     t1.x     t1.x    op: 00000000
> 31: xyz:  t0  t3  c1  bias->                      (000210c0)
>       w:  t0  t0  t0  bias-> t0.w                 (00800000)
>     xyz:   t0.xxx   t0.xxx   t0.xxx    op: 00004081 
>       w:     t0.z     t3.x     c1.x    op: 00018182
> 32: xyz: t32  t0  t0  bias-> t0.z                 (02000000)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.www      0.0  t32.xxx    op: 02804a0c 
>       w:    t32.x    t32.x    t32.x    op: 00000000
>NODE 2: alu_offset: 33, tex_offset: 2, alu_end: 7, tex_end: 1  (code_addr: 004221e1)
>  TEX:
>    TEX t3, t0, texture[1]   (000088c0)
>    TEX t5, t1, texture[0]   (00008141)
> 33: xyz:  t2  t3  t0  bias-> t1.xyz               (038400c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   t3.xyz      0.0    op: 00050200 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 34: xyz:  t5  t4  t0  bias-> t2.xyz               (03880105)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t4.xxx      0.0    op: 00050280 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 35: xyz:  t5  t0  t0  bias-> t0.xyz               (03800005)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t5.xyz   t0.zzz      0.0    op: 00050380 
>       w:     t5.x     t5.x     t5.x    op: 00000000
> 36: xyz:  t2  c8  t0  bias-> t2.xyz               (03880a02)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c8.xxx      0.0    op: 00050280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 37: xyz:  t0  c8  t0  bias-> t0.xyz               (03800a00)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c8.xxx      0.0    op: 00050280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 38: xyz:  t0  c7  t1  bias-> t0.xyz               (038019c0)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t0.xyz   c7.xxx   t1.xyz    op: 00020280 
>       w:     t0.x     t0.x     t0.x    op: 00000000
> 39: xyz:  t2  c7  t0  bias-> t0.xyz               (038009c2)
>       w:  t0  t0  t0  bias->                      (00000000)
>     xyz:   t2.xyz   c7.xxx   t0.xyz    op: 00020280 
>       w:     t2.x     t2.x     t2.x    op: 00000000
> 40: xyz:  t0  t0  t0  bias-> o0.xyz               (1c000000)
>       w:  t1  t0  t0  bias-> o0.w                 (01000001)
>     xyz:   t0.xyz      1.0      0.0    op: 00050a80 
>       w:     t1.w      1.0      0.0    op: 00040889

Actions: View

Attachments on bug 57888: 71005 | 71006 | 71007 | 71019 | 71054 | 71059 | 71060

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.