You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I generate animations on textures with vertex/fragment shaders, determine colors for lamps with a compute shader and copy the data then to the cpu to send it out via artnet packages over ethernet.
When the window of my app is very small, I get around 600fps, when it is fullscreen, I get around 160fps. This would be enough, but I'm normally using a different laptop for light installations as I don't want to have my work laptop unattended at events. The slower laptop only gets around 100fps in fullscreen, but I want to have around 140fps to have smooth animations.
I profiled the cpu and gpu usage with puffin and wgpu_profiler. It takes around 0,5ms for all my shaders and around 1ms for all the cpu work. This does not change if the window is small or fullscreen. But still the overall performance is very much different. My first guess was, that it is some issue with vsync, but vsync is disabled by using the wgpu setting AutoNoVsync.
Sadly I could not profile it with AMD profilers, as those are not supported with current mesa drivers. My hardware is not supported in older drivers though :/
The cheaper laptop has a intel gpu and I didn't get the tooling running on it either.
Do you have an idea what the problem could be? Could someone help me with another idea how to find the problem or even run the app on your pc with working profiling to give me pointers?
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I'm working on an open source project to do stage/event lighting with rust, wgpu, egui:
https://gitlab.com/photonenkollektiv/gled2
I generate animations on textures with vertex/fragment shaders, determine colors for lamps with a compute shader and copy the data then to the cpu to send it out via artnet packages over ethernet.
When the window of my app is very small, I get around 600fps, when it is fullscreen, I get around 160fps. This would be enough, but I'm normally using a different laptop for light installations as I don't want to have my work laptop unattended at events. The slower laptop only gets around 100fps in fullscreen, but I want to have around 140fps to have smooth animations.
I profiled the cpu and gpu usage with puffin and wgpu_profiler. It takes around 0,5ms for all my shaders and around 1ms for all the cpu work. This does not change if the window is small or fullscreen. But still the overall performance is very much different. My first guess was, that it is some issue with vsync, but vsync is disabled by using the wgpu setting AutoNoVsync.
Sadly I could not profile it with AMD profilers, as those are not supported with current mesa drivers. My hardware is not supported in older drivers though :/
The cheaper laptop has a intel gpu and I didn't get the tooling running on it either.
Do you have an idea what the problem could be? Could someone help me with another idea how to find the problem or even run the app on your pc with working profiling to give me pointers?
Beta Was this translation helpful? Give feedback.
All reactions