DX12 increase `max_storage_buffers_per_shader_stage` and `max_storage_textures_per_shader_stage` #3798

Elabajaba · 2023-05-23T03:16:41Z

Checklist

Run cargo clippy.
Run RUSTFLAGS=--cfg=web_sys_unstable_apis cargo clippy --target wasm32-unknown-unknown if applicable.
Add change to CHANGELOG.md. See simple instructions inside file.

Connections
#3333

Description
DX12's max_storage_buffers_per_shader_stage and max_storage_textures_per_shader_stage were set too low. This changes it to use the limits laid out in https://learn.microsoft.com/en-us/windows/win32/direct3d12/hardware-support

WebGPU minimum is FeatureLevel(FL) 11.0 + resource binding tier 2, or FL11.1 which is 64 UAVs (gpuweb/gpuweb#1069), so I'm not sure if this should error if we're FL11.0 + tier 1.

Testing
It didn't crash when I ran the tests or the water example with WGPU_BACKEND="dx12".

…_textures_per_shader_stage`

Elabajaba · 2023-05-23T03:41:23Z

~~Feature level detection seems to be broken, don't merge as is.~~

Edit: Fixed

wgpu-hal/src/dx12/adapter.rs

JMS55

The doc comment on max_storage_textures_per_shader_stage says the default is 8, but according to the spec and wgpu's code it should be 4.

teoxoy

LGTM

![image](https://github.com/bevyengine/bevy/assets/47158642/dbb62645-f639-4f2b-b84b-26fd915c186d) # Objective - Add Screen space ambient occlusion (SSAO). SSAO approximates small-scale, local occlusion of _indirect_ diffuse light between objects. SSAO does not apply to direct lighting, such as point or directional lights. - This darkens creases, e.g. on staircases, and gives nice contact shadows where objects meet, giving entities a more "grounded" feel. - Closes #3632. ## Solution - Implement the GTAO algorithm. - https://www.activision.com/cdn/research/Practical_Real_Time_Strategies_for_Accurate_Indirect_Occlusion_NEW%20VERSION_COLOR.pdf - https://blog.selfshadow.com/publications/s2016-shading-course/activision/s2016_pbs_activision_occlusion.pdf - Source code heavily based on [Intel's XeGTAO](https://github.com/GameTechDev/XeGTAO/blob/0d177ce06bfa642f64d8af4de1197ad1bcb862d4/Source/Rendering/Shaders/XeGTAO.hlsli). - Add an SSAO bevy example. ## Algorithm Overview * Run a depth and normal prepass * Create downscaled mips of the depth texture (preprocess_depths pass) * GTAO pass - for each pixel, take several random samples from the depth+normal buffers, reconstruct world position, raytrace in screen space to estimate occlusion. Rather then doing completely random samples on a hemisphere, you choose random _slices_ of the hemisphere, and then can analytically compute the full occlusion of that slice. Also compute edges based on depth differences here. * Spatial denoise pass - bilateral blur, using edge detection to not blur over edges. This is the final SSAO result. * Main pass - if SSAO exists, sample the SSAO texture, and set occlusion to be the minimum of ssao/material occlusion. This then feeds into the rest of the PBR shader as normal. --- ## Future Improvements - Maybe remove the low quality preset for now (too noisy) - WebGPU fallback (see below) - Faster depth->world position (see reverted code) - Bent normals - Try interleaved gradient noise or spatiotemporal blue noise - Replace the spatial denoiser with a combined spatial+temporal denoiser - Render at half resolution and use a bilateral upsample - Better multibounce approximation (https://drive.google.com/file/d/1SyagcEVplIm2KkRD3WQYSO9O0Iyi1hfy/view) ## Far-Future Performance Improvements - F16 math (missing naga-wgsl support https://github.com/gfx-rs/naga/issues/1884) - Faster coordinate space conversion for normals - Faster depth mipchain creation (https://github.com/GPUOpen-Effects/FidelityFX-SPD) (wgpu/naga does not currently support subgroup ops) - Deinterleaved SSAO for better cache efficiency (https://developer.nvidia.com/sites/default/files/akamai/gameworks/samples/DeinterleavedTexturing.pdf) ## Other Interesting Papers - Visibility bitmask (https://link.springer.com/article/10.1007/s00371-022-02703-y, https://cdrinmatane.github.io/posts/cgspotlight-slides/) - Screen space diffuse lighting (https://github.com/Patapom/GodComplex/blob/master/Tests/TestHBIL/2018%20Mayaux%20-%20Horizon-Based%20Indirect%20Lighting%20(HBIL).pdf) ## Platform Support * SSAO currently does not work on DirectX12 due to issues with wgpu and naga: * gfx-rs/wgpu#3798 * gfx-rs/naga#2353 * SSAO currently does not work on WebGPU because r16float is not a valid storage texture format https://gpuweb.github.io/gpuweb/wgsl/#storage-texel-formats. We can fix this with a fallback to r32float. --- ## Changelog - Added ScreenSpaceAmbientOcclusionSettings, ScreenSpaceAmbientOcclusionQualityLevel, and ScreenSpaceAmbientOcclusionBundle --------- Co-authored-by: IceSentry <c.giguere42@gmail.com> Co-authored-by: IceSentry <IceSentry@users.noreply.github.com> Co-authored-by: Daniel Chia <danstryder@gmail.com> Co-authored-by: Elabajaba <Elabajaba@users.noreply.github.com> Co-authored-by: Robert Swain <robert.swain@gmail.com> Co-authored-by: robtfm <50659922+robtfm@users.noreply.github.com> Co-authored-by: Brandon Dyer <brandondyer64@gmail.com> Co-authored-by: Edgar Geier <geieredgar@gmail.com> Co-authored-by: Nicola Papale <nicopap@users.noreply.github.com> Co-authored-by: Carter Anderson <mcanders1@gmail.com>

![image](https://github.com/bevyengine/bevy/assets/47158642/dbb62645-f639-4f2b-b84b-26fd915c186d) # Objective - Add Screen space ambient occlusion (SSAO). SSAO approximates small-scale, local occlusion of _indirect_ diffuse light between objects. SSAO does not apply to direct lighting, such as point or directional lights. - This darkens creases, e.g. on staircases, and gives nice contact shadows where objects meet, giving entities a more "grounded" feel. - Closes bevyengine#3632. ## Solution - Implement the GTAO algorithm. - https://www.activision.com/cdn/research/Practical_Real_Time_Strategies_for_Accurate_Indirect_Occlusion_NEW%20VERSION_COLOR.pdf - https://blog.selfshadow.com/publications/s2016-shading-course/activision/s2016_pbs_activision_occlusion.pdf - Source code heavily based on [Intel's XeGTAO](https://github.com/GameTechDev/XeGTAO/blob/0d177ce06bfa642f64d8af4de1197ad1bcb862d4/Source/Rendering/Shaders/XeGTAO.hlsli). - Add an SSAO bevy example. ## Algorithm Overview * Run a depth and normal prepass * Create downscaled mips of the depth texture (preprocess_depths pass) * GTAO pass - for each pixel, take several random samples from the depth+normal buffers, reconstruct world position, raytrace in screen space to estimate occlusion. Rather then doing completely random samples on a hemisphere, you choose random _slices_ of the hemisphere, and then can analytically compute the full occlusion of that slice. Also compute edges based on depth differences here. * Spatial denoise pass - bilateral blur, using edge detection to not blur over edges. This is the final SSAO result. * Main pass - if SSAO exists, sample the SSAO texture, and set occlusion to be the minimum of ssao/material occlusion. This then feeds into the rest of the PBR shader as normal. --- ## Future Improvements - Maybe remove the low quality preset for now (too noisy) - WebGPU fallback (see below) - Faster depth->world position (see reverted code) - Bent normals - Try interleaved gradient noise or spatiotemporal blue noise - Replace the spatial denoiser with a combined spatial+temporal denoiser - Render at half resolution and use a bilateral upsample - Better multibounce approximation (https://drive.google.com/file/d/1SyagcEVplIm2KkRD3WQYSO9O0Iyi1hfy/view) ## Far-Future Performance Improvements - F16 math (missing naga-wgsl support https://github.com/gfx-rs/naga/issues/1884) - Faster coordinate space conversion for normals - Faster depth mipchain creation (https://github.com/GPUOpen-Effects/FidelityFX-SPD) (wgpu/naga does not currently support subgroup ops) - Deinterleaved SSAO for better cache efficiency (https://developer.nvidia.com/sites/default/files/akamai/gameworks/samples/DeinterleavedTexturing.pdf) ## Other Interesting Papers - Visibility bitmask (https://link.springer.com/article/10.1007/s00371-022-02703-y, https://cdrinmatane.github.io/posts/cgspotlight-slides/) - Screen space diffuse lighting (https://github.com/Patapom/GodComplex/blob/master/Tests/TestHBIL/2018%20Mayaux%20-%20Horizon-Based%20Indirect%20Lighting%20(HBIL).pdf) ## Platform Support * SSAO currently does not work on DirectX12 due to issues with wgpu and naga: * gfx-rs/wgpu#3798 * gfx-rs/naga#2353 * SSAO currently does not work on WebGPU because r16float is not a valid storage texture format https://gpuweb.github.io/gpuweb/wgsl/#storage-texel-formats. We can fix this with a fallback to r32float. --- ## Changelog - Added ScreenSpaceAmbientOcclusionSettings, ScreenSpaceAmbientOcclusionQualityLevel, and ScreenSpaceAmbientOcclusionBundle --------- Co-authored-by: IceSentry <c.giguere42@gmail.com> Co-authored-by: IceSentry <IceSentry@users.noreply.github.com> Co-authored-by: Daniel Chia <danstryder@gmail.com> Co-authored-by: Elabajaba <Elabajaba@users.noreply.github.com> Co-authored-by: Robert Swain <robert.swain@gmail.com> Co-authored-by: robtfm <50659922+robtfm@users.noreply.github.com> Co-authored-by: Brandon Dyer <brandondyer64@gmail.com> Co-authored-by: Edgar Geier <geieredgar@gmail.com> Co-authored-by: Nicola Papale <nicopap@users.noreply.github.com> Co-authored-by: Carter Anderson <mcanders1@gmail.com>

…_textures_per_shader_stage` (gfx-rs#3798)

…er_shader_stage`) (#3891)

Elabajaba added 3 commits May 22, 2023 23:06

dx12 increase max_storage_buffers_per_shader_stage and `max_storage…

70966e4

…_textures_per_shader_stage`

clippy

664a68c

changelog

9acab0c

fix feature level detection

665acbc

teoxoy requested changes May 25, 2023

View reviewed changes

wgpu-hal/src/dx12/adapter.rs Outdated Show resolved Hide resolved

JMS55 suggested changes May 26, 2023

View reviewed changes

remove .min, fix max_storage_textures_per_shader_stage doc

e7c460b

JMS55 mentioned this pull request May 27, 2023

Screen Space Ambient Occlusion (SSAO) MVP bevyengine/bevy#7402

Merged

teoxoy approved these changes May 30, 2023

View reviewed changes

teoxoy merged commit e6314c4 into gfx-rs:trunk May 30, 2023

JMS55 mentioned this pull request Jun 18, 2023

SSAO/SSGI followup tracking issue bevyengine/bevy#8888

Open

13 tasks

cwfitzgerald added the PR: needs back-porting PR with a fix that needs to land on crates label Jun 20, 2023

Elabajaba added a commit to Elabajaba/wgpu that referenced this pull request Jun 27, 2023

DX12 increase max_storage_buffers_per_shader_stage and `max_storage…

9f4af02

…_textures_per_shader_stage` (gfx-rs#3798)

Elabajaba mentioned this pull request Jun 27, 2023

Backport #3798 to 0.16 (DX12 increase max_storage_buffers/textures_per_shader_stage) #3891

Merged

3 tasks

cwfitzgerald pushed a commit that referenced this pull request Jun 29, 2023

Backport #3798 to 0.16 (DX12 increase `max_storage_buffers/textures_p…

f7d4c50

…er_shader_stage`) (#3891)

cwfitzgerald removed the PR: needs back-porting PR with a fix that needs to land on crates label Jun 29, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

DX12 increase `max_storage_buffers_per_shader_stage` and `max_storage_textures_per_shader_stage` #3798

DX12 increase `max_storage_buffers_per_shader_stage` and `max_storage_textures_per_shader_stage` #3798

Elabajaba commented May 23, 2023 •

edited

Loading

Elabajaba commented May 23, 2023 •

edited

Loading

JMS55 left a comment •

edited

Loading

teoxoy left a comment

DX12 increase max_storage_buffers_per_shader_stage and max_storage_textures_per_shader_stage #3798

DX12 increase max_storage_buffers_per_shader_stage and max_storage_textures_per_shader_stage #3798

Conversation

Elabajaba commented May 23, 2023 • edited Loading

Elabajaba commented May 23, 2023 • edited Loading

JMS55 left a comment • edited Loading

Choose a reason for hiding this comment

teoxoy left a comment

Choose a reason for hiding this comment

DX12 increase `max_storage_buffers_per_shader_stage` and `max_storage_textures_per_shader_stage` #3798

DX12 increase `max_storage_buffers_per_shader_stage` and `max_storage_textures_per_shader_stage` #3798

Elabajaba commented May 23, 2023 •

edited

Loading

Elabajaba commented May 23, 2023 •

edited

Loading

JMS55 left a comment •

edited

Loading