Announcement

Collapse
No announcement yet.

X.Org vs. Wayland Linux Gaming Performance For NVIDIA GeForce + AMD Radeon In Early 2023

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • X.Org vs. Wayland Linux Gaming Performance For NVIDIA GeForce + AMD Radeon In Early 2023

    Phoronix: X.Org vs. Wayland Linux Gaming Performance For NVIDIA GeForce + AMD Radeon In Early 2023

    With recent NVIDIA's proprietary driver updates continuing to refine their Wayland support, the open-source AMDGPU Linux graphics drivers continuing to be enhanced, and work on the GNOME desktop with Mutter compositor continuing to advance, today's benchmarking article is looking at how the GNOME session under X.Org and Wayland for (X)Wayland is performing across various Linux games. It's been a while since I last ran a X.Org vs. (X)Wayland Linux gaming comparison so today's article is a fresh look from Ubuntu 22.10 while moving to the very latest graphics drivers and newest Steam Play Experimental state.

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Power consumption would be interesting for the cases where the framerate is way lower: does the GPU have to work harder or does it have a hard time to get work?

    Comment


    • #3
      Title says Wayland, then results are often quoted as (X)Wayland. Which one is it?

      It completely changes the meaning of the results.

      If it's XWayland, then the title should say that.

      Comment


      • #4
        Originally posted by franglais125 View Post
        Title says Wayland, then results are often quoted as (X)Wayland. Which one is it?

        It completely changes the meaning of the results.

        If it's XWayland, then the title should say that.
        Does it?
        Isn't it DRI in all of the cases anyway, so, unless there is a regression, the performance should be identical in all cases (X11, Wayland, XWayland)?

        Comment


        • #5
          It seems AMD's side usually performs a bit better under Wayland, except for that Dirt game. I've had a 7900 XTX for a while and can get everything to run, but there are a few regressions on Linux vs my 5700 xt. I'm sure the kinks will get worked out, and most things still run way better on the new hardware. I use Wayland with XWayland.

          It took about 6 months after its launch that my 5700 xt was flawless on Linux. Before then, I had all sorts of issues getting fixed every few weeks. It was still usable on the whole.

          Hopefully, AMD launches can be smoother on Linux going forward. I'm guessing that firmwareless/driverless patch will help with that.

          Also, I wonder how Wine with Wayland will fare when released. Really exciting work.
          Last edited by Mitch; 02 March 2023, 02:57 PM.

          Comment


          • #6
            Nvidia really doesn't like wayland.
            Dirt 2 looks really strange, perhaps someone should make an issue in mesa?

            Comment


            • #7
              Originally posted by Berniyh View Post
              Does it?
              Isn't it DRI in all of the cases anyway, so, unless there is a regression, the performance should be identical in all cases (X11, Wayland, XWayland)?
              XWayland is still an additional layer that's not present in the X11 tests that can have it's own bugs and performance issues so it's relevant to mention and can only, in theory, be slower than the game using Wayland directly.

              Also since SDL2 supports Wayland, some games can use it natively so being more specific would communicate to people when they're seeing something using Wayland vs using X11 under a Wayland session.

              Comment


              • #8
                Originally posted by franglais125 View Post
                Title says Wayland, then results are often quoted as (X)Wayland. Which one is it?

                It completely changes the meaning of the results.

                If it's XWayland, then the title should say that.
                I wonder if there was a single game running in true Wayland mode. I guess there was none.

                I'm a lot more curious to see a comparison of Windows games running under wine/DXVK in: Xorg, Wayland + XWayland, Wayand + patched Wine which supports native Wayland output. That would be awesome.

                In 1080p that is. Higher resolutions, especially 4K are too GPU heavy and could hide away any differences.

                Originally posted by RejectModernity View Post
                Nvidia really doesn't like wayland.
                Dirt 2 looks really strange, perhaps someone should make an issue in mesa?
                NVIDIA doesn't really care about Wayland. FTFY. Their corporate primary customers are OK with Xorg for the time being.
                Last edited by avis; 30 January 2023, 01:43 PM.

                Comment


                • #9
                  Originally posted by gabber View Post
                  Power consumption would be interesting for the cases where the framerate is way lower: does the GPU have to work harder or does it have a hard time to get work?
                  This is a really good idea, it goes to the "efficiency" of the code.

                  Comment


                  • #10
                    If Wayland performance is so bad, what is the reason for its existence?

                    Wayland has given us desktop users so many problems, so many bugs, so many disasters. What is it all good for? What does Wayland offer that is worth the enormous cost for end users who see their systems not boot, have error messages, lower performance, and other terrible effects. We are now so many years later that most distro's have switched or are in the process of switching, but i see no clear benefits for ordinary users. If they would know that some games / applications drop more than half in performance, what would be the reason to use it? What was wrong with X.org in the first place? Its complicated design may lower performance in theory i guess, but with Wayland having so many years of development it still struggles against X.org in some cases big time. The performance is unacceptable for some titles.

                    It seems to me perhaps a tiny minority is happy with Wayland and others would have benefited more from a stable and working X.org system?

                    And is that perhaps the real reason, license stuff? X.org is not GPL, is that the real issue? Linux devs feeling offended that something so vital is not GPL? I've seen the hostility toward ZFS and BSD, so i guess it is possible license stuff is part of the motivation. I lack the knowledge to form a definitive opinion on this subject, but from my point of view Linux desktop is still a hit and miss. Boot up an old Windows and it just works. Boot up a Linux LTS release and even the live CD/USB would show so many errors in the logs. And that is in the most vanilla / pristine state possible before even having it installed! Linux desktop simply has not achieved the stability of Windows yet. The devs seem focused on new toys, instead of getting the existing toys to work and have all the bugs ironed out.

                    Some examples: vanilla installation of ubuntu cannot play videos properly because of missing Vertical Sync - why? Why is vsync not the default, why do not things just work as they should work? Who wants vsync off when watching movies? Tell me! No one wants that right?! So why is it the default then? The 150% scaling doesn't work properly it works 100% and 200% but fractional scaling is still a mess. Hardware video acceleration often does not work for Firefox and VLC media player. All very important issues to be resolved for ordinary users, in 2023! I expected them fixed in 2000 already - Windows did?!

                    Comment

                    Working...
                    X