Igor Ljubuncic, who until 2024 was responsible for high-performance computing products at Canonical, conducted performance and energy consumption testing of the KDE Plasma 6.4 sessions based on X11 and Wayland. At the end of June, Igor published a review of the capabilities of KDE Plasma 6.4, concluding that the Wayland-based session performs worse and puts a greater load on the CPU and GPU than the X11-based session. In a series of new articles, Igor attempted to delve deeper into the performance testing of both sessions. <\/p>\n
Initially, it was noted that during testing on the Lenovo IdeaPad 3 laptop with integrated AMD GPU in the Wayland session, the CPU was constantly loaded at about 8% regardless of activity, while spikes fully loading the GPU were observed every 2-3 seconds. In the X11 session, the load on the CPU and GPU during idle was zero. Further checking with the radeontop utility showed higher resource consumption in the Wayland session, especially when enabling the 'Color Accuracy' profile in the screen settings. The table below shows the average GPU load generated in percentage with a similar set of running applications, collecting data every second for 60 seconds.
Power consumption testing with the powertop utility also showed an advantage for the X11 session: the power consumption of the Wayland session during multiple samples was 6.09 watts, Wayland with the 'Color Accuracy' profile was 6.05-6.08 watts, and X11 was 5.67-5.87 watts. Overall, it was concluded that the X11 session consumes 3-7% less battery charge than Wayland. <\/p>\n
Checking CPU load with the vmstat utility showed that during idle, the X11 session consumed 1.83% CPU, while Wayland used 1.97% (2.1% with the Color Accuracy profile). The utility was run on freshly created sessions in the Konsole terminal, calculating the average value during one-second sampling for 60 seconds. The CPU load using Wayland was found to be 7.6% higher (14% in Color Accuracy mode) than when using the X11 session. In the Wayland session, there were 25% more interrupts and 48% more context switches.
Next, an analysis of calls during idle was conducted using the perf utility. KWin_wayland consumed about 1% (1.5%) of CPU time, while KWin_X11 used 0.44%. Accesses to amdgpu accounted for 0.77% (1%) for Wayland and 0.65% for X11. In idle mode, the X11 session executed about half the number of CPU instructions compared to the Wayland-based session.
In the next article, tests were repeated on a configuration with Kubuntu 24.04 using KDE Plasma 5.27 on a Lenovo Y50-70 laptop with an Intel CPU and an NVIDIA graphics card. The results were approximately the same \u2014 the Wayland session was less efficient than X11. Power consumption was 22.42 watts for Wayland and 21.86 for X11, with CPU load at idle being 0.067 compared to 0.050, the number of context switches being 43.835\/s against 34.133\/s, and the load while watching 4K video in VLC was 12.54% versus 4.26%, with WebGL performance being 16 FPS compared to 29 FPS. <\/p>\n
Similarly, tests were conducted on a Lenovo IdeaPad 3 laptop with an AMD CPU\/GPU using a fresh build from the KDE Neon project. In the test reproducing 4K video in VLC, the CPU load using X11 with compositing turned off was 3.72%, X11 with compositing on was 9.8%, Wayland with Color Accuracy (CA) measured 25.71%, and Wayland in power-efficient mode (PE) was 31.51%. Thus, the CPU load when using Wayland was found to be 8-10 times higher than with X11. <\/p>\n
Using Wayland, there was also a recorded generation of 5-16% more interrupts. The difference in GPU load was within 2%: Wayland (PE) \u2014 56.33%, Wayland (CA) \u2014 57.33%, X11 (Comp ON) \u2014 57.98%, X11 (Comp OFF) \u2014 56.81%. In terms of power consumption, Wayland consumed 8-49% more energy: Wayland (PE) \u2014 13.8-20.4 watts, Wayland (CA) \u2014 13.8-14.1 watts, X11 (Comp ON) \u2014 10.7-12.1 watts, X11 (Comp OFF) \u2014 11.4-14.9 watts. In the WebGL Aquarium test, the results for Wayland and X11 were roughly the same: Wayland (PE) \u2014 16-38 FPS, Wayland (CA) \u2014 18-37 FPS, X11 (Comp ON) \u2014 16-42 FPS, X11 (Comp OFF) \u2014 21-42 FPS, but energy consumption with Wayland was about 8% higher. <\/p>\n
Ultimately, another experiment was conducted, this time encompassing GNOME from Fedora 42 and KDE Plasma 6.4 or KDE Neon. The testing was performed on a Lenovo IdeaPad 3 laptop with AMD CPU\/GPU. <\/p>\n