←back to thread

Dissecting the Apple M1 GPU, the end

(rosenzweig.io)
838 points alsetmusic | 2 comments | 27 Aug 25 01:44 UTC | HN request time: 0.001s | source
Show context
ActorNightly ◴[27 Aug 25 02:25 UTC] No.45034798[source]▶
Im glad she stepped away from Asahi linux. Its absolutely great from a techincal perspective and the progress that team has made, but talented people like her shouldn't be trying to reverse engineer software/hardware from shitty anti-consumer company that can make the entire project work in a heartbeat by publishing documentation, in lieu of building better stuff from the ground up.
replies(4): >>45034914 #>>45035307 #>>45036782 #>>45050209 #
ronsor ◴[27 Aug 25 02:45 UTC] No.45034914[source]▶
> in lieu of building better stuff from the ground up

To be fair, even if you have the best CPU and GPU designers, it's not as if you can call up TSMC and have them do a run of your shiny new processor on their latest (or even older) process. You can't fab them at home either.

replies(1): >>45035396 #
overfeed ◴[27 Aug 25 04:14 UTC] No.45035396[source]▶
Fortunately for her, Intel - her new employer has "fabs at home". Though on older nodes, TBF.
replies(1): >>45036046 #
1. distances ◴[27 Aug 25 06:21 UTC] No.45036046[source]▶
Intel's GPUs are manufactured on TSMC though.
replies(1): >>45047501 #
2. overfeed ◴[28 Aug 25 02:01 UTC] No.45047501[source]▶
That is correct. In my previous comment, for the sake of brevity, I deleted what I had written about Intel also being able to "call TSMC and have them doing a run of it's latest design", but it felt like that would have been belaboring the point that Intel isn't a rinky-dink chip operation, despite losing their commanding lead.