Really cool research. Overall, X11 still wins compared to Wayland(kwin) but only by 0.14 - 0.22 ms. The really bad performance hit is from XWayland which added 3.13 ms.

  • fruitcantfly@programming.dev
    link
    fedilink
    arrow-up
    2
    ·
    1 hour ago

    But the results… calling something “bad” for 3ms of latency is a bit ridiculous. That’s only significant to a cat, to us larger meat sacks 100ms would be blazing fast reaction speed, making 3ms negligible. For an average person it means even less.

    You are conflating reaction time and response time. They are very different things.

    Yes, your reaction time is in the 100s of miliseconds, but it doesn’t take much of a delay in the response to your reaction before it becomes perceivable, and the response time is what was measured here.

    If you have ever played rhythm games, then you may be aware that some of them have settings that allow you to offset the delay, such that your pressing a button at the exact same time as the note plays is also registered as happening at that time, even though it took some additional time for the input to reach the game. Without such a setting, you intentionally have to press buttons early (how much depends on your setup) to ensure that your input registers at the right time. A delay of 100ms would be extremely obvious

    • MonkderVierte@lemmy.zip
      link
      fedilink
      arrow-up
      1
      ·
      53 minutes ago

      You underestimate how short 100ms really is. Animations of 200ms feel snappy, anything less and you can leave the animation away.

      • fruitcantfly@programming.dev
        link
        fedilink
        arrow-up
        1
        ·
        41 minutes ago

        I’m well aware how short 100ms really is and it is very obvious if your inputs are off by 100ms or more; rhythm games are rarely that strict, but I’ve also played games where some inputs had to be timed to within 1/60th of a second (~16.7ms)