Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Enabling shadows on xrender slows down picom #1351

Closed
siduck opened this issue Oct 10, 2024 · 12 comments
Closed

Enabling shadows on xrender slows down picom #1351

siduck opened this issue Oct 10, 2024 · 12 comments
Labels
backend: xrender Issue related to the xrender backend performance Issue related to performance

Comments

@siduck
Copy link

siduck commented Oct 10, 2024

Hi, just got to know that picom finally supports animations, after trying it it asked me to specify backend in the config

I added glx and then the ram usage got high, upto 170 mb

image

I was using picom-ftlabs-git on aur before, animations work great there + dont have to use glx and it barely takes 10mb!

simplescreenrecorder-2024-10-10_18.28.36.mp4
@yshui
Copy link
Owner

yshui commented Oct 10, 2024

170mb memory usage can hardly be considered high (you have 30GB! 170mb is only 0.6%). and you don't have to use glx if you don't want to, animation works on xrender.

@yshui yshui closed this as not planned Won't fix, can't repro, duplicate, stale Oct 10, 2024
@siduck
Copy link
Author

siduck commented Oct 10, 2024

170mb memory usage can hardly be considered high (you have 30GB! 170mb is only 0.6%).

I get nice animations and all, on picom-ftlabs-git package previous picom forks with animations in just 10 mb! which is so cool but i feel like the the new picom has increased ram usage

Xrender feels so laggy when enabling shadows

simplescreenrecorder-2024-10-10_20.12.32.mp4

@yshui
Copy link
Owner

yshui commented Oct 10, 2024

in just 10 mb!

does picom-ftlabs still use 10 mb on glx? glx do use more memory but that's mostly from the gl driver. and i personally won't worry about a difference so small (again, 0.6%).

laggy when enabling shadows

now, that's the real problem here. shadows shouldn't be this slow, and in fact it is not on my machine.

what GPU do you have?

@yshui yshui changed the title Backend glx has 17x memory usage Enabling shadows slows down picom Oct 10, 2024
@yshui yshui reopened this Oct 10, 2024
@yshui yshui changed the title Enabling shadows slows down picom Enabling shadows on xrender slows down picom Oct 10, 2024
@yshui yshui added backend: xrender Issue related to the xrender backend performance Issue related to performance labels Oct 10, 2024
@siduck
Copy link
Author

siduck commented Oct 10, 2024

image

@siduck
Copy link
Author

siduck commented Oct 10, 2024

does picom-ftlabs still use 10 mb on glx?

No, it uses 170mb too. i think by default it uses xrender.

On Original picom , i get below 10 mb with xrender too! but the shadows make things slow, weird

@yshui
Copy link
Owner

yshui commented Oct 10, 2024

does cpu usage go up when picom lags?

@siduck
Copy link
Author

siduck commented Oct 10, 2024

not at all

@siduck
Copy link
Author

siduck commented Oct 10, 2024

also i didnt face the lag on the picom version 11.2, when picom didnt had animations support

simplescreenrecorder-2024-10-10_22.02.18.mp4

@yshui
Copy link
Owner

yshui commented Oct 10, 2024

not at all

does turning off vsync and/or frame-pacing change anything?

@yshui
Copy link
Owner

yshui commented Oct 10, 2024

this is a duplicate of #1349

@yshui yshui closed this as not planned Won't fix, can't repro, duplicate, stale Oct 10, 2024
@siduck
Copy link
Author

siduck commented Oct 11, 2024

not at all

does turning off vsync and/or frame-pacing change anything?

unfortunately no

@siduck
Copy link
Author

siduck commented Nov 3, 2024

@yshui i recently added a new gpu ( was using IGPU before ), now i no longer face this lag

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
backend: xrender Issue related to the xrender backend performance Issue related to performance
Projects
None yet
Development

No branches or pull requests

2 participants