← Back to context

Comment by troupo

8 months ago

> like WebGPU, are Khronos IP that Apple has no reason to object to except on

You do know that Apple is basically the original author of WebGPU, right (together with Mozilla)?

> I wonder why Apple would deliberately avoid an API that might obsolete the requirement for games to use the App Store

And your fantasy of Apple deliberately avoiding it is based on what exactly?

https://webkit.org/blog/9528/webgpu-and-wsl-in-safari/

https://webkit.org/blog/14879/webgpu-now-available-for-testi...

> You do know that Apple is basically the original author of WebGPU, right (together with Mozilla)?

Apple is the original author of a lot of tech they end up abandoning. Certainly a lot of Khronos IP, paging through their history.

> And your fantasy of Apple deliberately avoiding it is based on what exactly?

Based on a 4 year (!!!) porting time from MacOS Safari to iOS Safari. Basically textbook feet-dragging there.

  • > Apple is the original author of a lot of tech they end up abandoning.

    Doubtful

    > Certainly a lot of Khronos IP, paging through their history.

    Everyone abandons Khronos IP, or doesn't really supports it, paging through history in general. Because Khronos IP ends up a designed-by-committee crapfest. Meanwhile WebGPU is not and has never been a Khronos IP. It's developed within a w3c working group: https://www.w3.org/2020/gpu/

    > Based on a 4 year (!!!) porting time from MacOS Safari to iOS Safari.

    Based on a 4-year porting of what from MacOS Safari top iOS Safari?

    - WebGPU spec is literally in draft status, so things can still change. It's literally in stage 2 of 5 of spec development

    - Neither Safari nor Firefox have enabled WebGPU yet. The fact that Chrome rushed and enabled it by default does not make the spec or the standard finished and ready to be enabled everywhere

    - webgpu can be enabled with a toggle in advanced settings in Safari on iOS (as is the case with most new features for in all browsers)