immersive-web
Wide Review for WebGPU and WGSL
The GPU for the Web Working Group has made significant progress on first versions of the WebGPU and WebGPU Shading Language (WGSL) specifications. The group is now firming up the specifications with a view to requesting publication as Candidate Recommendations later this year.
The Working Group would like to invite you to review the latest versions of the specifications:
- WebGPU: https://www.w3.org/TR/webgpu/
- WGSL: https://www.w3.org/TR/WGSL/
Working drafts are updated on a continuous basis but there should be no more breaking changes at this stage. Feel free to review dated versions of the specifications to avoid looking at a moving target.
Brandon prepared an explainer some time ago to lay out a plan for integrating WebGPU in WebXR [This repo]. If further changes to WebGPU seem warranted, now would be a good time to raise needs.
The Working Group plans to request publication as Candidate Recommendations no sooner than 15 October 2022 and welcomes comments prior to that date, preferably through issues on the group's GitHub repository: https://github.com/gpuweb/gpuweb/issues
/agenda to bring to the attention of the group
Is that really needed? Authors can already use frameworks to do the same. With a framework they'll also have more control.
In addition, if we allow rendering to a canvas, we need to have consistent rendering across UAs
Authors can already use frameworks to do the same. With a framework they'll also have more control.
Sure, but the point is not to require a framework.
In addition, if we allow rendering to a canvas, we need to have consistent rendering across UAs
Depends on the definition of "consistent": it's not hard requirement, as it isn't a requirement with CSS, WebGL, or with <model>, as otherwise <model> wouldn't be viable either.
Authors can already use frameworks to do the same. With a framework they'll also have more control.
Sure, but the point is not to require a framework.
At least for us, the point is that a model will render in actual 3D. :-)
In addition, if we allow rendering to a canvas, we need to have consistent rendering across UAs
Depends on the definition of "consistent": it's not hard requirement, as it isn't a requirement with CSS, WebGL, or with
<model>, as otherwise<model>wouldn't be viable either.
Don't 2D Canvas and WebGL have conformance tests on rendered pixels? I suspect we will have to develop a similar test suite if we allow a model as a canvas image source.