The author discusses unresolved problems with Wayland, highlighting issues with surface suspension and the broken FIFO implementation that lead to reduced GPU-bound performance. They argue that Wayland is currently not preferable over X11 due to severe performance and usability regressions. The author insists on reverting the default to X11 until certain protocols are released and stable for major compositors. Controversially, they emphasize that games and applications do not benefit from choosing Wayland. The author also delves into technical details about FIFO, commit timing, and potential solutions. There is a debate about the best course of action, with arguments for and against defaulting to Wayland. The complexity of the issue is acknowledged, with XWayland also posing challenges compared to native Wayland. The discussion revolves around the impact on users, developers, and the need for prioritizing solutions to improve the overall experience.
https://github.com/libsdl-org/SDL/pull/9345