While there were a few graphics benchmarks in yesterday’s Ubuntu 25.04 beta benchmarks, today’s article is looking more at the Ubuntu 25.04 Linux gaming performance for both the GNOME 48 and KDE Plasma 6.3 desktops that default to the Wayland-based session by default while also trying out the X11 session for both of these desktops.
Originally my plan was to run an X11 vs. Wayland desktop gaming comparison on Ubuntu 25.04. But at least for the current beta state of Ubuntu 25.04, it didn’t work out as planned… In particular, at least for the desktop hardware I’ve been testing so far, the GNOME X11 session is rather busted. The GNOME Wayland session has been working out great with the same hardware but when switching to the GNOME X11 session instead it’s been constant screen flashing. KDE Plasma 6.3 worked fine on the same hardware under both X11 and Wayland but with the GNOME X11 session it was a constant screen blanking/flashing experience:
The Ubuntu 25.04 GNOME on X11 experience I’ve been having on a few systems… Works great on GNOME Wayland and KDE X11/Wayland. For a new article up soon. pic.twitter.com/vumsMsmWwb
— Phoronix (@phoronix) March 28, 2025
So in the end for my Linux gaming tests I ran a number of benchmarks using the default GNOME 48 Wayland session, KDE Plasma 6.3 on Wayland, and then KDE Plasma 6.3 on X11.
The same AMD Ryzen 9 9900X3D + AMD Radeon RX 7900 XTX system was used for all of this testing with just switching out the desktop session in use between running a variety of Linux gaming/graphics benchmarks for looking at the KDE Plasma 6.3 vs. GNOME 48 performance and then under KDE for the X11 vs. Wayland difference as of this Ubuntu 25.04 beta state.
When starting things off with the GravityMark OpenGL and Vulkan benchmark, there was little difference between the GNOME 48 and KDE Plasma 6.3 desktops when using the Wayland session. But it was interesting to note that the KDE Wayland session was delivering slightly better performance than the KDE X11 session when running at 4K.
Firing up other games, it was again interesting to see that the KDE Wayland session was delivering slightly better performance than the KDE X11 desktop session with the likes of Counter-Strike 2.
DiRT Rally 2.0 was one of the few exceptions where the KDE Plasma Wayland session led to slightly lower performance than the X11 session or relative to the GNOME Wayland use on Ubuntu 25.04.
It was good to see both GNOME 48 and KDE Plasma 6.3 delivering similar performance under their Wayland sessions on Ubuntu 25.04 while using the open-source AMD Radeon graphics drivers.
With HITMAn 3 the trend continued of the Plasma 6.3 Wayland session delivering slightly better performance than with the X11 desktop.
Even for older OpenGL-rendered games, the KDE Wayland session continued showing slightly better performance than KDE X11.
With Strange Brigade running on Steam Play, the KDE X11 desktop session ran into big issues but ran fine with KDE Wayland.
Across a variety of graphics and gaming workloads tested, it was great seeing Plasma 6.3 and GNOME 48 delivering similar performance with the Radeon RX 7900 XTX graphics on Ubuntu 25.04 beta.
There was a clear trend too of the KDE Wayland desktop delivering slightly better performance than using the KDE X11 desktop session. Paired with the show-stopping experience trying to use the GNOME X11 session on Ubuntu 25.04 and then paired with the KDE performance results, Wayland certainly is shining on Ubuntu 25.04.
More Ubuntu 25.04 benchmarks to come with other hardware and other workloads/scenarios over the coming weeks leading up to the Ubuntu 25.04 stable release in mid-April.