Apple M1 Pro and M1 Max
Apple (video, Hacker News, MacRumors):
The CPU in M1 Pro and M1 Max delivers up to 70 percent faster CPU performance than M1, so tasks like compiling projects in Xcode are faster than ever. The GPU in M1 Pro is up to 2x faster than M1, while M1 Max is up to an astonishing 4x faster than M1, allowing pro users to fly through the most demanding graphics workflows.
[…]
M1 Pro offers up to 200GB/s of memory bandwidth with support for up to 32GB of unified memory. M1 Max delivers up to 400GB/s of memory bandwidth — 2x that of M1 Pro and nearly 6x that of M1 — and support for up to 64GB of unified memory. And while the latest PC laptops top out at 16GB of graphics memory, having this huge amount of memory enables graphics-intensive workflows previously unimaginable on a notebook. The efficient architecture of M1 Pro and M1 Max means they deliver the same level of performance whether MacBook Pro is plugged in or using the battery. M1 Pro and M1 Max also feature enhanced media engines with dedicated ProRes accelerators specifically for pro video processing.
[…]
Utilizing the industry-leading 5-nanometer process technology, M1 Pro packs in 33.7 billion transistors, more than 2x the amount in M1. A new 10-core CPU, including eight high-performance cores and two high-efficiency cores[…]
The M1 Max’s DRAM is as fast as Intel’s on-die LLC circa 2016. Between this and the SSD performance (as fast as RAM was about 10 years ago), Apple is making a mockery of memory hierarchies.
As for the M1 Pro/Max, reminder that a single P-core can saturate the M1’s memory bandwidth, even significantly downclocked. And the M1 already has a lot of memory bandwidth. All that extra memory bandwidth in the new chips has to make a pretty big difference.
See also: Ken Shirriff.
Previously:
Update (2021-10-19): Andrei Frumusanu:
Today’s reveal of the new generation Apple Silicon has been something we’ve been expecting for over a year now, and I think Apple has managed to not only meet those expectations, but also vastly surpass them. Both the M1 Pro and M1 Max look like incredibly differentiated designs, much different than anything we’ve ever seen in the laptop space. If the M1 was any indication of Apple’s success in their silicon endeavors, then the two new chips should also have no issues in laying incredible foundations for Apple’s Mac products, going far beyond what we’ve seen from any competitor.
Apple’s M1 Pro/Max is the second step in a major change in computing. What might be seen as an evolution from iPhone/ARM is really part of an Apple story that began in 1991 with PowerPC.
[…]
When you look at M1 Pro/Max today it is tempting to think of this in terms of performance, but performance per watt AND integrated graphics AND integrated memory AND integrated application processors is innovation in an entirely different direction.
Update (2021-10-29): Andrei Frumusanu (Hacker News):
The M1 Pro and M1 Max change the narrative completely – these designs feel like truly SoCs that have been made with power users in mind, with Apple increasing the performance metrics in all vectors. We expected large performance jumps, but we didn’t expect the some of the monstrous increases that the new chips are able to achieve.
On the CPU side, doubling up on the performance cores is an evident way to increase performance – the competition also does so with some of their designs. How Apple does it differently, is that it not only scaled the CPU cores, but everything surrounding them. It’s not just 4 additional performance cores, it’s a whole new performance cluster with its own L2. On the memory side, Apple has scaled its memory subsystem to never before seen dimensions, and this allows the M1 Pro & Max to achieve performance figures that simply weren’t even considered possible in a laptop chip. The chips here aren’t only able to outclass any competitor laptop design, but also competes against the best desktop systems out there, you’d have to bring out server-class hardware to get ahead of the M1 Max – it’s just generally absurd.
Andy Somerfield (via John Gruber):
The #M1Max is the fastest GPU we have ever measured in the @affinitybyserif Photo benchmark. It outperforms the W6900X - a $6000, 300W desktop part - because it has immense compute performance, immense on-chip bandwidth and immediate transfer of data on and off the GPU (UMA).
Yining Karl Li (tweet, Hacker News):
The wider takeaway here though is that in order to give the M1 Max some real competition, one has to skip laptop chips entirely and reach for not just high end desktop chips, but for server-class workstation hardware to really beat the M1 Max. For workloads that push the CPU to maximum utilization for sustained periods of time, such as production-quality path traced rendering, the M1 Max represents a fundamental shift in what is possible in a laptop form factor.
This is interesting: only about 6% improvement form 8 core M1 Pro to 10 core M1 Max when compiling a large Xcode project
Jean-Louis Gassée (Hacker News):
The Intel side of our village has dismissed the M1 Pro and Max as impressive but hardly threatening: “Sure, Apple has a fleeting advantage due to their access to TSMC’s denser 5 nanometer process, but once Intel gets there, x86 chips will outperform Apple Silicon, especially with their access to the vast library of Windows software.”
Some things never change. Intel fans had the same reaction, eight years ago, when Apple introduced its first 64-bit processor, the A7 that powered the iPhone 5.
Almost all of us expected Intel to win on the single-threaded front because of high clock rates and some serious architectural improvements but what is surprising is that they even beat the Apple M1 Max on the multi-threaded front. The Alder Lake Core i9 12900HK mobility processor gets an astounding 13256 score which is followed by Apple at 12753 points. The Intel 11980HK (stock) is further into the horizon at 9149 points and AMD clocks in at 8217 points. This is a generation over generation increase of almost 45% in roughly the same TDP - although not surprising because even though the ADL-P CPU only has 8 “big cores” the small cores have proven to be quite powerful as well.
Now keep in mind, I have no qualms that Apple is still going to win on a power efficiency metric - they always have since the A11 - but Apple’s reign as the fastest mobility chip “period” seems like it is going to be short-lived (we expect ADL-P to land in early 2022).
Update (2021-11-16): Rene Ritchie:
Tom Boger, Vice President of iPad & Mac Product Marketing and Tim Millet, Vice President of Platform Architecture, join me to talk about what they thought when Apple first decided to switch the Mac to custom silicon, what it was like bringing their low/slow/wide approach to a thermal envelope as big as the new MacBook Pro, how scalable architecture really scales up this much (and more), how they think about transistor budget in an increasing post-big compute core world, gaming on Mac, and which MBPs we’re all rocking!
Update (2021-11-24): Timothy Liu (via Hacker News):
I still had questions, so here I am with some (casual) benchmarks that I hope add some additional perspective into interesting hardware capabilities on the M1 Max SOC, just for fun and out of my curiosity.
