-
-
Notifications
You must be signed in to change notification settings - Fork 3.5k
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
Pipelined-rendering performance #3052
Comments
Is size being 10^3 for the old renderer a typo? |
I changed it manually to test different sizes. |
I am seeing very similar results with MacOS 11.2.3 |
Is this metal-specific? On my other machine I see
And also these lines, which didn't show up on the macbook.
|
On my Mac, disabling vsync, example
|
Yeah so far we're only seeing repros from metal users. This is definitely worth investigating prior to the 0.6 release. Given that our sprite rendering pipeline (which performs pretty well) uses pretty much the same renderer abstractions as our pbr pipeline, I have a feeling that this is something specific to how the pbr pipeline is implemented, but I guess we'll find out! I'll start poking around. I have mac/linux/windows machines to test on. |
I now have access to a m1 Mac, both
|
Just to add my experience; possibly related - I'm seeing poor performance on an M1 Mac on Big Sur, running @cart 's
|
Is this still an issue? We've moved significantly in terms of the end-to-end engine performance and specifically the renderer in the releases since 0.6. |
Bevy version
c5af133
Operating system & version
macOS Catalina, 10.15.7
AdapterInfo { name: "AMD Radeon RX 5700 XT", vendor: 0, device: 0, device_type: DiscreteGpu, backend: Metal }
What you did
test file for old renderer:
test file for new renderer:
my results:
What you expected to happen
new renderer to be faster
What actually happened
new renderer is slower
The text was updated successfully, but these errors were encountered: