Monday, December 7th 2020
Riding on the Success of the M1, Apple Readies 32-core Chip for High-end Macs
Apple's M1 SoC is possibly the year's biggest semiconductor success story, as the chip has helped Apple begin its transition away from Intel's x86 machine architecture, and create its own silicon that's optimized for its software and devices; much like its A-series SoCs powering iOS devices. The company now plans to scale up this silicon with a new 32-core version designed for high-performance Mac devices, such as the fastest MacBook Pro models; and possibly even iMac Pros and Mac Pros. The new silicon could debut in a new-generation Mac Pro in 2022. Bloomberg reports that the new silicon will allow this workstation to be half the size of the current-gen Mac Pro workstation in form, while letting Apple keep its generational performance growth trajectory.
In addition, Apple is reportedly developing a 16-core "big" + 4 "small" core version of the M1, which could power more middle-of-the-market Macs, such as the iMac desktop, and the bulk of the MacBook Pro lineup. The 16B+4s core chip could debut as early as Spring 2021. Elsewhere, the company is reportedly stepping up efforts to develop its own high-end professional-visualization GPU that it can use in its iMac Pro and Mac Pro workstations, replacing the AMD Radeon Pro solutions found in the current generation. This graphics architecture will be built from the ground-up for the Metal 3D graphics API, as well as a parallel compute accelerator. Perhaps the 2022 debut of the Arm-powered Mac Pro could feature this GPU.
Sources:
Bloomberg, The Verge
In addition, Apple is reportedly developing a 16-core "big" + 4 "small" core version of the M1, which could power more middle-of-the-market Macs, such as the iMac desktop, and the bulk of the MacBook Pro lineup. The 16B+4s core chip could debut as early as Spring 2021. Elsewhere, the company is reportedly stepping up efforts to develop its own high-end professional-visualization GPU that it can use in its iMac Pro and Mac Pro workstations, replacing the AMD Radeon Pro solutions found in the current generation. This graphics architecture will be built from the ground-up for the Metal 3D graphics API, as well as a parallel compute accelerator. Perhaps the 2022 debut of the Arm-powered Mac Pro could feature this GPU.
48 Comments on Riding on the Success of the M1, Apple Readies 32-core Chip for High-end Macs
Apple saying highend mac well looking at 20k.us here or what :-)
Splitting up into multiple dies makes production easier in many ways, like Epyc/Threadripper/Ryzen.
big.LITTLE doesn't seem to do much for high-end compute: you pretty much always want more big cores if you have a difficult task (like CPU-based Raytracing) running. big.LITTLE style has advantages in long-running compute however: maybe the LITTLE cores can feed a GPU in a GPU-heavy situation (GPU-based raytracing??).
-------
Hmmmm... I think I'm pretty open about my fanboyism towards SIMD-compute. The M1 is very disappointing from the SIMD-perspective: just 128-bit wide. Even if they extend up to 256-bits, those M1 cores are utterly huge compared to Zen. It seems unlikely that they can offer as much parallelism in a SIMD situation. But I'm willing to be proven wrong on this front.
Maybe that is what Intel / AMD / Nvidia need to come out with ever better products. Apple has the advantage of positive consumer perception (among its fans).
Unfortunately apple products are stupid expensive.
Apple M1 = Revolutionary
The M1 has made it obvious that ARM will dominate the desktop within the next 10 years. The average desktop user doesn't need an x86 CPU. x86 users are about to become a niche market. Efficiency has clearly won out over flexibility.
In terms of efficiency, the M1 is currently the best CPU and GPU on the market. Apple's more powerful chips will also be more efficient than any competing products.
A lot of PC users are already in denial over the M1's superiority, and they'll stay that way for a long time because they're stupid.
On a side note, Apple is currently the only one doing high performance mainstream ARM cpu, Qualcomm don't have the money to follow up on Apple, and windows on ARM is still a dodgy OS. You should really avoid to make a comparison between a closed and vertically integrated system, and an open system where every actor need to do his part for it to "work".
Right windows and ARM doesn't look good because nobody seems a to have a solid plan to make a smooth transition, both on the software and hardware side. Qualcomm, AMD, or whever else would need to invest hard on ARM research. Don't forget that Apple is the most profitable company in the world with a gigantic R&D budget.
Yeah not like apple is going to open the os so performance well as long as you can do what you want through the apple store you're all set lol
On the other hand, I think the larger chips (like the hypothetical 32 core) could be revolutionary if they scale linearly and best the existing HEDT CPUs at real tasks. In that situation I could see Apple workstations (and maybe servers too!) and ARM becoming mainstream instead inhabiting relative niches in computing.
Anyways, exciting possibilities and I’m looking forward to seeing what Apple delivers in the future. In the meantime, let’s all remember that Apple’s marketing is just that while wait.
I'm not sure my next laptop will be MacBook, but my MBP from late 2013 was the best decision at that time for my use. Got fed up with Windows, i lost a lot of data when it crashed. And yes, i had backups, but not all folders got marked manually. But that's another story...
Interesting to see where all this goes.
I think M-based Macs are going to be more efficient and better values than PCs for a long time.
Anyway, I wonder what the end cost will be. MAC's are usually way over priced. But I am enticed to buy a Mac Mini with one of these new M1 processors.
It's pretty clear that the M1 outcompetes any similar-tier Intel and likely Ryzen CPUs as the many reviews online note that most Intel apps run as fast or notably faster in bloody emulation on these M1s. That's both due to the SOC design and good programming in Rosetta. This can also be done on the PC side as MS also has the deep pockets to do it, but it'll take a long time as they lack the experience.
And let's be honest, they also need the motivation to do so. I doubt Apple will win many new customers with even notably superior hardware as how many people will switch from a low price laptop/small PC to a more expensive Mac, and that's even before considering the sunk cost fallacy.
The performance, Single thread mostly is impressive, but it's the first 5nm CPU on the market. We will see how it compare to 5nm AMD and 7 nm Intel CPU if they get out. Apple had to suffer Intel Fabs issue and now they got their way out of these and are able to use the top process in the world
Still the revolution is the image that arm can be top cpu (some other arm architecture already show that it can really be performant like the Fujistu A64FX). Even rumors now have AMD to resurect K12 to benefits from the trend.
There are many things that are impactful with the M1, but the most interesting one is they added in the core front end hardware decode of many x86 instruction to improve the speed they run x86 code in Rosetta 2. This can have a very huge impact on the future as we might see other manufacturer doing similar things and even become at some point able to natively run different ISA.
But this is something that is still unclear if Intel will let that go or if Apple got the license in their Deals with Intel. Other vendors tried to do similar things and they got lawsuit from Intel so time will tell. But AMD or what left of Via, could do that without too much trouble.