System Environment: ---------------------------------------- Platform: HSW Libdrm: (master)libdrm-2.4.45-3- g3586337f3703ce4833a375f66b08df064a1cec28 Mesa: (master)f6beb4c6b647b0d5077c7abd6c99cdaba6e0c112 Xserver: (master)xorg-server-1.14.99.1-82- g7e97166c4fc38076b6224e4dbdd632b43444149f Xf86_video_intel: (master)2.21.6-47-gd6c30d1d4df6bcdfa075bd29da7c8aabee20774c Cairo: (master)631bf299256e11a17511977f357e0353fb5615f7 Libva: (staging)880c916092c6d68096c95e49405714d8a21f9322 Libva_intel_driver: (staging)0a5cfdab47c788569feb0a3e4293906687523f88 Kernel: (drm-intel-nightly) 675fa342aa3a54af0a81f23809fff9590dfb6fc1 Test result: ---------------------------------------- HSW-ult FBC disable Average_watt FBC enable Average_watt game 21 21 IVB FBC disable Average_watt FBC enable Average_watt game 34.8 34.7 Findings: ---------------------------------------- For game, FBC saved 0.1w on IVB, but didn’t saved power on HSW. FBC pass/fail criteria is 0.1-0.2w.
In the game workload, the criteria is for FBC not to consume more power. Since every frame requires recompression (costing power), and ideally the game is running at the vertical refresh, there is no power saving from used the same compressed data for more than one refresh. For this reason, FBC is turned off under this workload.
But I confused why game workload saved 0.1w power, and HSW didn't. (In reply to comment #1) > In the game workload, the criteria is for FBC not to consume more power. > Since every frame requires recompression (costing power), and ideally the > game is running at the vertical refresh, there is no power saving from used > the same compressed data for more than one refresh. For this reason, FBC is > turned off under this workload.
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.