Measuring Input Latency on Linux: X11 vs. Wayland, VRR, and DXVK (marco-nett.de)

Article summary

The article measures input latency on Linux, comparing X11 and Wayland display servers, Variable Refresh Rate (VRR), and a low-latency DXVK fork. The results show that X11 has slightly lower latency than Wayland, but the difference is small. VRR has a more significant impact on latency, and the low-latency DXVK fork provides a noticeable improvement. The article also highlights the importance of avoiding XWayland due to its high latency.

Main themes

• Linux display servers
• Input latency
• X11 vs Wayland
• VRR and DXVK
• Gaming performance

What commenters say

• The transition to Wayland was rough and forced upon users without providing feature parity with X11.
• Wayland is not ready for prime time and has been claiming to be ready for years despite substantial deficiencies.
• The effort spent on developing Wayland support could have been better spent on improving existing applications and features.
• The difference in latency between X11 and Wayland is negligible and may not be noticeable to most users.
• XWayland is a significant contributor to high latency and should be avoided.
• The low-latency DXVK fork provides a noticeable improvement in gaming performance.
• The Wayland protocol is not inherently flawed, but its implementation and adoption have been problematic.
• Users should have a choice between X11 and Wayland, and the decision to switch should be based on individual needs and preferences.