Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Linux latency measurements and compositor tuning

farnoy.dev

Post reply on HN
Print view View original

31–40 of 47 posts

Re: Linux latency measurements and compositor tuning

#31

Post by 7e »

nickjj wrote: post #15

When I switched to Linux from Windows last year I noticed I had a lot of keyboard input latency only when gaming. It was like I had ~100ms of input latency and it felt exactly the same as playing old school Quake I on a dial up connection without client side prediction turned on. It's like you're skating on ice with a delay when seeing the output vs when you performed the input. This was despite having a solid 60 FPS…

This is endemic with open source. Nobody owns the hard bugs, nobody does the systems spelunking, and there is little power to make cross-cutting changes. This is why I use a Mac. Not perfect, but better than that no-accountability midden-heap that is Linux.

Maybe AI coding agents will make the situation better, but because open source maintainers are too dim to understand the complex changes the AI makes, and too poor to have their own AIs to help them, they won't take the changes. I make improvements to open source but am forced to keep them to myself.

Re: Linux latency measurements and compositor tuning

#32

Post by Joe_Cool »

fasterik wrote: post #27
nickjj wrote: post #15

When I switched to Linux from Windows last year I noticed I had a lot of keyboard input latency only when gaming. It was like I had ~100ms of input latency and it felt exactly the same as playing old school Quake I on a dial up connection without client side prediction turned on. It's like you're skating on ice with a delay when seeing the output vs when you performed the input. This was despite having a solid 60 FPS…

I abandoned Windows for a variety of reasons and while I find Linux better in many ways, the graphics and compositing situation is a bummer. To add another example, I was trying to do a screen recording the other day and it was dropping frames like crazy. I don't know who to blame, whether it's Gnome, or Wayland, or OBS, or Nvidia. But the point is my confidence in the entire ecosystem is low. I have plenty of bad th…

Still on X11. OBS just works.

Re: Linux latency measurements and compositor tuning

#33

Post by tapland »

Joe_Cool wrote: post #32
fasterik wrote: post #27

Earlier quoted context omitted.

I abandoned Windows for a variety of reasons and while I find Linux better in many ways, the graphics and compositing situation is a bummer. To add another example, I was trying to do a screen recording the other day and it was dropping frames like crazy. I don't know who to blame, whether it's Gnome, or Wayland, or OBS, or Nvidia. But the point is my confidence in the entire ecosystem is low. I have plenty of bad th…

Still on X11. OBS just works.

I've been using OBS on wayland, they got captures working really well a few years ago.

But for VR stuff I've been going back to X11, and I was just last night trying to finish a screen capture program on wayland (w/ kde plasma) and idk I just have to do repeated screenshots ana analyze those > horrible fps, but at least I think I got that working finally after many attempts last year.

Gaming is not really an issue now with DEs that help ppl disable compositions and wine/proton.

Re: Linux latency measurements and compositor tuning

#35

Post by ndriscoll »

7e wrote: post #31
nickjj wrote: post #15

When I switched to Linux from Windows last year I noticed I had a lot of keyboard input latency only when gaming. It was like I had ~100ms of input latency and it felt exactly the same as playing old school Quake I on a dial up connection without client side prediction turned on. It's like you're skating on ice with a delay when seeing the output vs when you performed the input. This was despite having a solid 60 FPS…

This is endemic with open source. Nobody owns the hard bugs, nobody does the systems spelunking, and there is little power to make cross-cutting changes. This is why I use a Mac. Not perfect, but better than that no-accountability midden-heap that is Linux. Maybe AI coding agents will make the situation better, but because open source maintainers are too dim to understand the complex changes the AI makes, and too poo…

Except GP said it works fine with all configurations with KDE, which is what most people should be using. I've been using Linux for 15+ years and have never heard of niri.

Re: Linux latency measurements and compositor tuning

#36

Post by nickjj »

fasterik wrote: post #27
nickjj wrote: post #15

When I switched to Linux from Windows last year I noticed I had a lot of keyboard input latency only when gaming. It was like I had ~100ms of input latency and it felt exactly the same as playing old school Quake I on a dial up connection without client side prediction turned on. It's like you're skating on ice with a delay when seeing the output vs when you performed the input. This was despite having a solid 60 FPS…

I abandoned Windows for a variety of reasons and while I find Linux better in many ways, the graphics and compositing situation is a bummer. To add another example, I was trying to do a screen recording the other day and it was dropping frames like crazy. I don't know who to blame, whether it's Gnome, or Wayland, or OBS, or Nvidia. But the point is my confidence in the entire ecosystem is low. I have plenty of bad th…

I know what you mean about "just works".

With OBS recordings (not streaming) on Linux I occasionally get situations where short but heavy disk or network I/O will cause my webcam's video to get out of sync with my microphone (separate mic from the webcam). This makes it look like my lips are way out of sync with the video. It's something I haven't been able to track down as there's no errors or side effects posted in journald logs.

This is using vanilla Arch with niri. It happens pretty often and I fix it by splitting the audio and video in my editing tool and then shifting the audio track over half a second or so.

I recorded over 1,000 videos on Windows with the same hardware and the above never happened once.

Re: Linux latency measurements and compositor tuning

#37

Post by goodmythical »

ndriscoll wrote: post #35
7e wrote: post #31

Earlier quoted context omitted.

This is endemic with open source. Nobody owns the hard bugs, nobody does the systems spelunking, and there is little power to make cross-cutting changes. This is why I use a Mac. Not perfect, but better than that no-accountability midden-heap that is Linux. Maybe AI coding agents will make the situation better, but because open source maintainers are too dim to understand the complex changes the AI makes, and too poo…

Except GP said it works fine with all configurations with KDE, which is what most people should be using. I've been using Linux for 15+ years and have never heard of niri.

>which is what most people should be using

Most people should be running hundreds of megabytes of bloat?

KDE is unuseable on the brand new i3/8gb I run purchased last year.

I mean, the desktop renders, but god forbid I'd like to run more than three firefox tabs...

Re: Linux latency measurements and compositor tuning

#38

Post by eqvinox »

LollipopYakuza wrote: post #30
functionmouse wrote: post #24

Earlier quoted context omitted.

this is literally the only thing that ever mattered imo

Would you mind elaborating why? What makes it more important than other inputs?

Executing other actions isn't a feedback loop like moving the mouse pointer; for the mouse pointer you need the feedback from the system before you even get to an actual action. You can press keys on a keyboard blindly, if it doesn't matter where you can even click blindly, but you can't position the pointer blindly. It's closed loop vs. open loop.

Re: Linux latency measurements and compositor tuning

#39

Post by raggi »

nickjj wrote: post #15

When I switched to Linux from Windows last year I noticed I had a lot of keyboard input latency only when gaming. It was like I had ~100ms of input latency and it felt exactly the same as playing old school Quake I on a dial up connection without client side prediction turned on. It's like you're skating on ice with a delay when seeing the output vs when you performed the input. This was despite having a solid 60 FPS…

fwiw I’ve been running lots of stuff under niri with proton Wayland mode for the last ~6 months on nvidia with a vrr display and it’s been fine.

Can’t diagnose here, and not meaning any descent to your experience, just a different user experience for me.

Re: Linux latency measurements and compositor tuning

#40

Post by xemdetia »

Joe_Cool wrote: post #19
Leonard_of_Q wrote: post #14

No X11 because 'KDE drops it' is a rather silly argument for not including it, especially given that X11 is likely to trump Wayland-based systems on this metric. I'm using X11 and will continue to do so until something better comes along. Thus far Wayland is not better, just different and in several ways less functional. Yes, I do use X11 network transparency, sometimes through X2go, other times directly.

Also still on X11 because of xorgxrdp and X11 forwarding. It would be cool to see this test with a CRT connected to a VGA port (adapters make it slower) and a PS/2 input device.

I have been having great luck with waypipe (gitlab) as an X11 forwarding alternative (OS X to Ubuntu 24.04). The setup was relatively painless and it 'just worked.' while this might not suit your needs for other reasons I was pleasantly surprised.
Post reply on HN