Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPU advancements in M3 and A17 Pro [video]

developer.apple.com

Post reply on HN
Print view View original

31–40 of 148 posts

Re: GPU advancements in M3 and A17 Pro [video]

#31

Post by ytch »

aroman wrote: post #28
ytch wrote: post #14

Earlier quoted context omitted.

It's one of ̶W̶W̶D̶C̶ developer sessions. Some of them at WWDC are basic 101 introduction for rookies. There are advanced (for me) sessions like: https://developer.apple.com/videos/play/wwdc2023/10127/ https://developer.apple.com/videos/play/wwdc2023/10042/ Although it's true that they won't discuss at hardware level.

How can it be a WWDC session if it's referring to hardware only announced a few weeks ago?

Yes, It's my fault. The website uses the same layout with WWDC sessions, so I misunderstand it :/

Re: GPU advancements in M3 and A17 Pro [video]

#32

Post by jeffybefffy519 »

pavlov wrote: post #26
zmmmmm wrote: post #8

Does apple document exactly how many actual true cores there are inside their GPUs? It is always confusing they say "40 core GPU" but I assume these are shader cores which each inside them can execute (per the video) "many thousands" of parallel execution paths. So how does one translate to an equivalent in "CUDA cores" type terminology?

What Apple calls a GPU core seems to be roughly the same as what Nvidia calls a “stream multiprocessor”. For example a 1080 GTX GPU has 20 stream multiprocessors (SM), each containing 128 cores, each of which supports 16 threads. Meanwhile Apple describes the M1 GPU as having 8 cores, where “each core is split into 16 Execution Units, which each contain eight Arithmetic Logic Units (ALUs). In total, the M1 GPU contai…

Even moreso we shouldnt assume they have similar architectural layouts.

Re: GPU advancements in M3 and A17 Pro [video]

#33

Post by jeffybefffy519 »

w10-1 wrote: post #12

The video is just enough of a peek into the GPU's to encourage people to write using Metal API's (and by the way, use the new APIs and FP16).

They should just support directx. Devs will never support two graphics api’s. it costs too much especially to grab the marginal mac os share that has powerful enough gpu’s. Id bed in 4 years apple moves to directx.

Re: GPU advancements in M3 and A17 Pro [video]

#34

Post by cmovq »

zmmmmm wrote: post #8

Does apple document exactly how many actual true cores there are inside their GPUs? It is always confusing they say "40 core GPU" but I assume these are shader cores which each inside them can execute (per the video) "many thousands" of parallel execution paths. So how does one translate to an equivalent in "CUDA cores" type terminology?

I think a better comparison is to look at floating point performance. For example, the 10 core M2 GPU does 3.6 TFLOPS (FP32) while an RTX 4060 does 15 TFLOPS and an RTX 4090 82.58

Re: GPU advancements in M3 and A17 Pro [video]

#35

Post by kanwisher »

gary_0 wrote: post #2

I skimmed the video but a lot of it sounded more like advertising than technical information to me. On the other hand, I'm looking forward to watching the Asahi folks crack this stuff open.

it literally has code and diagrams of how to organize SIMD instructions to fill out all the shader cores

Re: GPU advancements in M3 and A17 Pro [video]

#36

Post by nemothekid »

jeffybefffy519 wrote: post #33
w10-1 wrote: post #12

The video is just enough of a peek into the GPU's to encourage people to write using Metal API's (and by the way, use the new APIs and FP16).

They should just support directx. Devs will never support two graphics api’s. it costs too much especially to grab the marginal mac os share that has powerful enough gpu’s. Id bed in 4 years apple moves to directx.

DirectX is exclusive to the Windows platform. At this point, it's probably deeply tied into Windows. I don't see how you can make that bet.

Re: GPU advancements in M3 and A17 Pro [video]

#37

Post by reroute22 »

wmf wrote: post #4
gary_0 wrote: post #2

I skimmed the video but a lot of it sounded more like advertising than technical information to me. On the other hand, I'm looking forward to watching the Asahi folks crack this stuff open.

The beginning does sound like marketing but eventually it gets into technical information.

I'd say the beginning sounds like an introduction to GPU architectures in general, not marketing.

Somewhere in the ballpark of 5:30-6:00 or so it describes prior hardware design of the Apple's shader core, and starting 7:00 it goes into hardware design of the new M3/A17Pro shader core. It's actually surprisingly detailed, e.g. Nvidia's whitepapers provide less detail on the actual organization of their SMs.

Re: GPU advancements in M3 and A17 Pro [video]

#38

Post by creshal »

alphanullmeric wrote: post #7

It’s amazing how bad the competition is. The A17 pro has 2 performance cores and 4 efficiency cores. The Google G3 has 9 cores of 3 different types, the fastest being slower than Apple’s performance cores, the most efficient being less efficient than apple’s efficiency cores. And it’s a phone so you don’t take advantage of the extra parallelism. You just get the worst of both worlds. no wonder these android phones ha…

A big part of Apple's "winning formula" is taking their giant piles of money and negotiating exclusive contracts for whatever is scheduled to be the most advanced semiconductor node next year.

Anyone else literally cannot compete, they don't have billions in pocket change they don't know how to spend otherwise, so they'll have to wait until the exclusivity agreement expires.

Re: GPU advancements in M3 and A17 Pro [video]

#39

Post by aurareturn »

jeffybefffy519 wrote: post #33
w10-1 wrote: post #12

The video is just enough of a peek into the GPU's to encourage people to write using Metal API's (and by the way, use the new APIs and FP16).

They should just support directx. Devs will never support two graphics api’s. it costs too much especially to grab the marginal mac os share that has powerful enough gpu’s. Id bed in 4 years apple moves to directx.

DirectX is closed source. Also, there are more games on Metal than DirectX.

Re: GPU advancements in M3 and A17 Pro [video]

#40

Post by pixelpoet »

wmf wrote: post #17
zmmmmm wrote: post #8

Does apple document exactly how many actual true cores there are inside their GPUs? It is always confusing they say "40 core GPU" but I assume these are shader cores which each inside them can execute (per the video) "many thousands" of parallel execution paths. So how does one translate to an equivalent in "CUDA cores" type terminology?

Each Apple core (heh) has 128 FPUs so 40 cores would be akin to 5120 CUDA "cores".

Not quite, because Nvidia counts dual issue as a flat doubling of "core" (which previously you could accurately call a vector lane) count.
Post reply on HN