fosstodon.org is one of the many independent Mastodon servers you can use to participate in the fediverse.
Fosstodon is an invite only Mastodon instance that is open to those who are interested in technology; particularly free & open source software. If you wish to join, contact us for an invite.

Administered by:

Server stats:

10K
active users

mort

I measured mouse cursor latency between Wayland and X11: mort.coffee/home/wayland-input

Turns out there's roughly 1 screen refresh cycle more mouse cursor latency on GNOME Wayland than on GNOME under X11 on my system. That's interesting.

mort.coffeeHard numbers in the Wayland vs X11 input latency discussion - Mort's Ramblings

@mort Can you think of any reasons why the mouse cursor latency is higher on #Wayland compared to #X11?

@GerryT I'm far from an expert in the area, however some people have suggested that Mutter has input event rate limiting that could be affecting things. Others have suggested that it might be that GNOME misses a frame when the screen goes from being static to having motion. Some have brought up the forced vsync in GNOME Wayland.

Personally, my guess is that Mutter is somehow delaying cursor position updates to the next screen refresh, while X.Org does it whenever and "races the beam". But idk.

@GerryT I want to highlight this comment I just saw from Lina from the Asahi Linux project: lobste.rs/s/oxtwre/hard_number

It seems like my guess was more or less correct, Mutter (and other Wayland compositors) will sync cursor position updates to the rest of the screen in one atomic screen update, while X works similar to running the cursor without vsync which improves latency but might cause tearing and means that the cursor can be a frame ahead of the rest of the screen.

lobste.rsHard numbers in the Wayland vs X11 input latency discussion | Lobsters

@mort a boxplot and some stats for you (wilcox p = 0.00012).