Support non-sRGB image formats for RenderTarget::Image
#22031
Merged
+16
−32
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
I believe it was a bug that ManualTextureView::format could be different from the format of the actual wgpu::Texture.
I believe it was a bug that OutputColorAttachment::format could have been different from the actual format of the wgpu::Texture. This also fixes bevyengine#15201
Contributor
Author
|
CI failure doesn't immediately seem relevant to my changes https://github.com/bevyengine/bevy/actions/runs/19949128748/job/57205300759?pr=22031 |
We know that main_texture_format cannot be anything but Rgba16Float or Rgba8UnormSrgb.
greeble-dev
added
C-Bug
mate-h
approved these changes
Dec 9, 2025
Contributor
mate-h
left a comment
mate-h
left a comment
There was a problem hiding this comment.
Nice overall cleanup! Code changes look good to me, don't really have nitpicks. Can you open a github issue for the future work on gamma correction?
| #[inline] | ||
| pub fn out_texture_format(&self) -> TextureFormat { | ||
| self.out_texture.format | ||
| self.out_texture.view.texture().format() |
Contributor
There was a problem hiding this comment.
This seems like an important distinction here
| .cloned() | ||
| .zip(target.get_texture_format(&windows, &images, &manual_texture_views)) | ||
| .map(|(view, format)| { | ||
| OutputColorAttachment::new(view.clone(), format.add_srgb_suffix()) |
Contributor
There was a problem hiding this comment.
This is the fix for the linked issue, no longer include the sRGB suffix.
IceSentry
approved these changes
Dec 9, 2025
IceSentry
added
S-Ready-For-Final-Review
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Objective
RenderTarget::Imagedoesn't supportRgba8UnormandBgra8Unormtexture formats (non-srgb formats) #15201Solution
OutputColorAttachment::formatcould be different from the actual underlying texture format (ofOutputColorAttachment::view). This would eventually lead to a texture format mismatch on theBlitPipeline. I removed theOutputColorAttachment::formatfield to resolve this conflict.ManualTextureView.Testing
RenderTarget::Imagedoesn't supportRgba8UnormandBgra8Unormtexture formats (non-srgb formats) #15201 no longer cause a WGPU validation error.pbrexample on native and WebGPU and it still works as before.Future Work
Although the WGPU validation error is fixed, when using a non-sRGB texture format for the GPU image in the
headless_rendererexample, the output image seems to have incorrect gamma. I think this is because the example does a byte-for-byte copy from the GPU image to the CPU image without any gamma correction; and theimagecrate saves the image assuming a nonlinear color space.I don't think we need to fix this as part of this PR, but it might be nice for
bevy_image::Image::try_into_dynamicto eventually handle the gamma correction when saving linear color images.Here you can see a comparison
With sRGBNo sRGB
