Aye, people have mused about the possibility for different multi-chip solutions for the Mac Pro, including SOC-over-board. There are no rumors that I know of that Apple has anything like that planned, but you never know.
There’s economies of scale in it for Nvidia. Apple’s Mac Pro costs what it costs partially because it leans on the R&D efforts from the millions and millions of base processors that sell in iPads and laptops. They stand on those shoulders, and put a few more of those cores on the die. A completely separate parallel effort to make something that doesn’t build on that tech? The prices can only go up from here.
People are still “trained” to think that the capability of high end solutions shouldn’t be available across the entire line. The fact that your average user, with their single threaded tasks, wouldn’t perceive a significant difference, day-to-day, between a Mac Pro and a MacBook Air (due to the single threaded performance being so similar) still seems foreign and wrong to them. Apple has done with AMD/Intel will never be able to do and instead of marveling at that fact, some are pointing at Intel “Why can’t it be more like that?” It could, Apple could just do what AMD/Intel does and ship solutions with FAR more disabled cores than they do today.
To combine these posts together, this is one advantage Apple has over Nvidia in this space, at least when it comes to boutique-firm/prosumer hardware. Even Apple's base machines have decent if not good margins and they don't have "big iron" device profits to protect. Nvidia's gaming chips have much lower profit margins than their professional devices and they try to actively dissuade people from using the gaming chips for the professional purposes (e.g. in addition to lower VRAM, you'll note Blackwell 2 gaming chips have a lot lower matmul throughput than Blackwell 1 professional chips and of course no FP64 acceleration and this has been true for many Nvidia product lines even when the two share silicon). This isn't to say that Apple wouldn't want high margins on a hypothetical Mac Pro Hidra, of course they would, but Apple are in a position to build something that delivers many, though not all, of the performance features Nvidia offers at the high end (the most obvious example is VRAM capacity, but increase ray tracing/FP32/matmul compute performance and suddenly ...), more economically than Nvidia (currently, Nvidia and AMD likely see the writing on the wall here too and more SOCs are expected from both - even AMD is supposedly working on an ARM chip).
My guess is their internal PCC servers will indeed have fancier interconnects if the rumors are anything to go on, but what Apple can (economically) put on the Mac Pro/Studio is a different matter.
Last edited:
