It does not seem to cover the Neural Accelerators, Apple's equivalent of the Tensor Cores. They only got released on M5 platform. This is probably the most important part to cover.
Apple Neural Engine: Architecture, Programming, and Performance
11–20 of 30 posts
Re: Apple Neural Engine: Architecture, Programming, and Performance
#12This Neural Engine seems useless for LLMs. Trapped in the wrong architecture
Re: Apple Neural Engine: Architecture, Programming, and Performance
#13This scans very much as AI-written.
This is obvious Claude slop writing, the author would be advised to use vale [1] with samples of their own writing as a guide. > Performance begins with the roofline. On the M1 the engine holds about 12 fp16 TFLOP/s of compute against a DRAM-bandwidth ceiling. The roofline has a ridge point near 141 FLOP per byte, a 2 MB working-set threshold, a 0.23 ms floor under any single dispatch, and efficiency near 0.37 picojo…
Please no. The author would be advised to write their own original thoughts.
Re: Apple Neural Engine: Architecture, Programming, and Performance
#14Earlier quoted context omitted.
This is obvious Claude slop writing, the author would be advised to use vale [1] with samples of their own writing as a guide. > Performance begins with the roofline. On the M1 the engine holds about 12 fp16 TFLOP/s of compute against a DRAM-bandwidth ceiling. The roofline has a ridge point near 141 FLOP per byte, a 2 MB working-set threshold, a 0.23 ms floor under any single dispatch, and efficiency near 0.37 picojo…
Please no. The author would be advised to write their own original thoughts.
It was a joke, nothing could save this "paper". I don't think the author wrote anything. They pointed claude at a directory and said "write a paper"
Re: Apple Neural Engine: Architecture, Programming, and Performance
#15It does not seem to cover the Neural Accelerators, Apple's equivalent of the Tensor Cores. They only got released on M5 platform. This is probably the most important part to cover.
Those are part of the GPU not the Neural Engine.
Re: Apple Neural Engine: Architecture, Programming, and Performance
#16Is there a non-slop version of this information available? I am reading up on GPU / ML micro architecture and am looking for some good sources.
I skimmed through it, what makes you think it is slop?
Re: Apple Neural Engine: Architecture, Programming, and Performance
#17This Neural Engine seems useless for LLMs. Trapped in the wrong architecture
Apple is releasing CoreAI which is supposed to be optimized for LLMs and the transformer architecture.
Re: Apple Neural Engine: Architecture, Programming, and Performance
#18Earlier quoted context omitted.
The burden of proof should be with the beholder. Must be so easy to scream AI when you don’t want to read an article.
You obviously haven't read it, because it is clunky garbage. > 19.4 Pacing compiles after a failure > A failed compile is not free of side effects on the shared compile service. A compile that fails restarts the service, which takes a few seconds to come back, and failures that keep arriving faster than the service can restart between them keep it from making progress, so unrelated compiles slow down until the failur…
Personally I'm not in the habit of printing and eating articles I read, but in the unlikely event that I did I find it even less likely that I would be concerned with its' nutritional content. (/s)
Re: Apple Neural Engine: Architecture, Programming, and Performance
#19Re: Apple Neural Engine: Architecture, Programming, and Performance
#20Earlier quoted context omitted.
The burden of proof should be with the beholder. Must be so easy to scream AI when you don’t want to read an article.
You obviously haven't read it, because it is clunky garbage. > 19.4 Pacing compiles after a failure > A failed compile is not free of side effects on the shared compile service. A compile that fails restarts the service, which takes a few seconds to come back, and failures that keep arriving faster than the service can restart between them keep it from making progress, so unrelated compiles slow down until the failur…
you forgot the bologna and iceberg lettuce