Intel Tiger Lake VRR Support Being Worked On For Linux, More PCI IDs Introduced

[phoronix]

Phoronix: Intel Tiger Lake VRR Support Being Worked On For Linux, More PCI IDs Introduced

With the forthcoming Linux 5.7 kernel Intel Tiger Lake "Gen12" Xe graphics are considered stable as in enabled by default but that doesn't mean they are done working on features for the highly-anticipated next-gen Intel graphics...

http://www.phoronix.com/scan.php?page=news_item&px=Intel-Tiger-Lake-VRR-Linux

[shmerl]

Progress of adaptive sync on Wayland is very slow. Looks like there is still no agreement on how to implement it.

[Guest]

Progress of adaptive sync on Wayland is very slow. Looks like there is still no agreement on how to implement it.

IMO the communication between apps and compositor doesn't need to change - it is already VRR-like in the sense that apps dispatch a new frame to the compositor when they finish rendering one, and they can do it at whatever rate they want.

The VRR we are talking about here is the rate at which the GPU/display controller sends frames to the display. General window managers/compositors like Mutter and KWin don't care about VRR because they are not super graphically demanding, and drawing the desktop + app windows at a high and constant refresh rate is more important.

I think the best scenario may be to create a new compositor dedicated to playing games, which can use VRR when talking to the display. It can also avoid compositing when games are played fullscreen, thus addressing any performance concerns due to compositing.