Bug 4490 - finer-grain pointer ballistics
Summary: finer-grain pointer ballistics
Status: RESOLVED DUPLICATE of bug 2927
Alias: None
Product: xorg
Classification: Unclassified
Component: Input/Mouse (show other bugs)
Version: unspecified
Hardware: Other Linux (All)
: lowest enhancement
Assignee: Xorg Project Team
QA Contact:
URL: http://www.microsoft.com/whdc/device/...
Depends on:
Reported: 2005-09-17 14:49 UTC by Krishna Sethuraman
Modified: 2006-04-04 05:25 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Description Krishna Sethuraman 2005-09-17 14:49:37 UTC
I think the mouse motion under other window/graphics servers 'feels' (per my 
hand-eye perception) smoother.  I'd like the X server to have more 
customizability in its acceleration.  The only thing I could find about this on 
the web is described under the term 'pointer ballistics', from the URL above. 
Using mice under Mac, Windows, and XFree86 or xorg X servers produce different 
motion characteristics.  In particular, when using a Mac and a Linux system 
side-by-side, pointer motion from a mouse directly connected to the Mac 'feels' 
* Run osxvnc (http://osxvnc.sf.net) on the Mac 
* Run x2vnc (http://fredrik.hubbe.net/x2vnc.html) on the XFree86/xorg system, 
  and connect to the osxvnc server. 
* Move the mouse directly connected to the Mac and notice the pointer motion 
* Move the mouse directly connected to the XFree86/xorg system over to the 
  Mac's screen via the x2vnc mechanism, and notice the pointer motion 
* Notice the difference in 'feel' between the two 
I got a couple things from my quick read of the 'pointer ballistics' 
document in that URL: 
* Windows approximates a smooth curve by changing acceleration 
  multiple times (rather than X's once) during mouse travel. 
* Windows (and I suspect the Mac as well) implements 'subpixilation' 
  at slow speeds.  This takes multiple mouse hardware move events 
  in the same direction (?) to move the pointer by one pixel, improving 
  fine pointer control. 
I've no formal HCI training, so I don't have a real solid idea of how the 
motion should be made tweakable (i.e., if 'pointer ballistics' is the best or 
only way of describing the effect).  Let me know if you can't reproduce the 
effect I described above and I'll see if I can come up with a better example. 
Finally, I'm not certain that this kind of control belongs in the X server vs 
in an O/S driver (I'm just not sure where the abstraction boundary lies in this 
case).  If I misfiled this, I apologize. 
Wireless optical mouse (Logitech Cordless Optical Mouse, Model 
M-RM67A), receiver connected via USB. 
Debian Sarge, running xserver-xorg 6.8.2 from http://backports.org
Comment 1 Krishna Sethuraman 2005-09-17 23:13:20 UTC
See also bug 138 and bug 2927 .  Maybe this has already been taken care of and
I'm using it improperly.
Comment 2 Adam Jackson 2006-04-04 22:25:38 UTC

*** This bug has been marked as a duplicate of 2927 ***

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.