X Gets Predictable Pointer Acceleration
X.Org 7.4 / X Server 1.5 has experienced an incredibly long delay in getting out the door. It was originally supposed to ship in February, then May, and now its stagnate until Mesa 7.1 ships. It looks like it will be a late August or early September release, which is almost a year after X.Org 7.3 had shipped.
If the delays alone weren't bad enough, this X update will be shipping with a slimmer set of features than what was originally expected. Most recently DRI2 was dropped from X.Org 7.4 due to the TTM vs. GEM kernel memory manager situation, but it also lost out on Multi-Pointer X (MPX) and other features.
Permitting these features don't get further delayed, X.Org 7.5 / X Server 1.6 will hopefully end up being a nice release. We don't expect this next update to ship until the middle of next year or so, but it already contains the MPX integration, input improvements, and should contain a fixed up version of the Direct Rendering Infrastructure 2. In addition, a new acceleration method has been announced for the X Server.
Last week the UXA acceleration architecture was announced by Keith Packard as a GEM-ified architecture based upon EXA acceleration, but this isn't to be confused with this new announcement. In fact, this announcement by Simon Thum is for Predictable Pointer Acceleration.
The predictable part of this mouse pointer acceleration is to tell whether the pointer will move. Its features include user-selectable profiles control pointer acceleration, adaptive and constant deceleration, acceleration becomes predictable, and no overshoot when X blocks for a short time. Planned for the future is sub-pixel position and velocity and velocity estimation. However, most users likely won't notice this change when it goes into effect. This has been proposed for X Server 1.6.
The announcement for this new pointer acceleration can be read on the X.Org mailing list. Detailed documentation surrounding X's Predictable Pointer Acceleration can be found on the X.Org Wiki.
If the delays alone weren't bad enough, this X update will be shipping with a slimmer set of features than what was originally expected. Most recently DRI2 was dropped from X.Org 7.4 due to the TTM vs. GEM kernel memory manager situation, but it also lost out on Multi-Pointer X (MPX) and other features.
Permitting these features don't get further delayed, X.Org 7.5 / X Server 1.6 will hopefully end up being a nice release. We don't expect this next update to ship until the middle of next year or so, but it already contains the MPX integration, input improvements, and should contain a fixed up version of the Direct Rendering Infrastructure 2. In addition, a new acceleration method has been announced for the X Server.
Last week the UXA acceleration architecture was announced by Keith Packard as a GEM-ified architecture based upon EXA acceleration, but this isn't to be confused with this new announcement. In fact, this announcement by Simon Thum is for Predictable Pointer Acceleration.
The predictable part of this mouse pointer acceleration is to tell whether the pointer will move. Its features include user-selectable profiles control pointer acceleration, adaptive and constant deceleration, acceleration becomes predictable, and no overshoot when X blocks for a short time. Planned for the future is sub-pixel position and velocity and velocity estimation. However, most users likely won't notice this change when it goes into effect. This has been proposed for X Server 1.6.
The announcement for this new pointer acceleration can be read on the X.Org mailing list. Detailed documentation surrounding X's Predictable Pointer Acceleration can be found on the X.Org Wiki.
Add A Comment