Hacker News Reader

Comment by hocuspocus | Hacker News Reader

> I also await their on dGPU solution.

Genuine question from someone who doesn't know hardware architecture that well:

It seems clear that we'll see non-PoP variants of their SoC's, for instance to support multiple RAM configurations in the high-end models.

However, at this point, if Apple's own GPU scales well, should we except 3rd party dGPU at all? Can't Apple ship SoCs with multiple times the M1 core count and graphics units, package 8 or 16GB of fast memory (GDDR or HBM), and call it a day? They're going to push their own software stack anyway, why bother with AMD GPUs fabbed on a previous generation TSMC process.

View on HN

Dedicated/external GPUs are 10x as powerful as internal ones. Apple doesn't have a miracle technology that can make a fanless 2080 Ti. AMD and Nvidia arent stopping improving cards just because M1 exists

I haven't said anything about Apple GPU architecture being a miracle fanless technology. Right now they ship SoCs with 7 or 8 graphics units, what happens when they put 64 of them?

Look at the architecture of AMD SoCs in the PS5 and Xbox. Obviously Apple cannot get away with unified memory in the higher-end machines, but what would make them prefer an AMD dGPU when it seems they have everything they need internally. Again, assuming the GPU scales well.

If you want GPU performance to match a dedicated GPU, you can't fit it in a Macbook (or any Mac except the Mac Pro, for that matter). Even if the M1's graphics units only used 50% as much power as an equivalently-powerful Nvidia or AMD dedicated GPU, putting enough graphics power onboard to match even a midrange GPU would immediately overwhelm the cooling abilities of the computer.

Apple could absolutely choose to make better cooling, but they've demonstrated time and time again that they'd rather just let thermal throttling happen and trust that their users will continue to buy a product that doesn't live up to its spec sheet's promises.

Any cooling that could support dedicated-GPU-level performance would require making the chassis dramatically thicker - take a look at some of the gaming laptops out there today with desktop-class GPUs, they can get be like 2 inches thick in pursuit of cooling good enough to support a full-power GPU without massive throttling.

I don't understand your argument at all. You realize Apple does put dedicated AMD GPUs in their Macbook Pros?

The Radeon Pro 5600M with 8GB of HBM2 has a pretty significant power envelope, and is coupled to 6 or 8 core Intel H series that are equally power hungry.

The fact 16" MBPs are undercooled is beyond the point. Apple ships laptops with a ~95W TDP. Why would they not ship a SoC with 16/24/32 graphics units for high-end laptops and iMacs?

Unless I'm mistaken, this isn't about the GPUs inside the MBP but about the external ones you can connect separate power and a Thunderbolt cable to.

In that case, you're not talking about a Radeon Pro 5600M but rather a desktop-class RTX 3090 with 24GB of high-performance RAM and a TDP of 350W. A MBP couldn't hope to power or cool one of those with a built-in GPU, never mind come close to matching the performance or supporting CUDA without nVidia sueing Apple out of existence.

In theory there's nothing stopping Apple from making a large single die that contains a desktop-class CPU and desktop-class GPU. There are also rumors that Apple is designing their own "Lifuka" discrete GPU that would presumably be a larger version of the same design that they use in their SoCs.

Reply on news.ycombinator.com