Apple's M1 Pro, M1 Max SoCs Investigated: New Performance and Efficiency Heights
by Andrei Frumusanu on October 25, 2021 9:00 AM EST- Posted in
- Laptops
- Apple
- MacBook
- Apple M1 Pro
- Apple M1 Max
CPU ST Performance: Not Much Change from M1
Apple didn’t talk much about core performance of the new M1 Pro and Max, and this is likely because it hasn’t really changed all that much compared to the M1. We’re still seeing the same Firestrom performance cores, and they’re still clocked at 3.23GHz. The new chip has more caches, and more DRAM bandwidth, but under ST scenarios we’re not expecting large differences.
When we first tested the M1 last year, we had compiled SPEC under Apple’s Xcode compiler, and we lacked a Fortran compiler. We’ve moved onto a vanilla LLVM11 toolchain and making use of GFortran (GCC11) for the numbers published here, allowing us more apple-to-apples comparisons. The figures don’t change much for the C/C++ workloads, but we get a more complete set of figures for the suite due to the Fortran workloads. We keep flags very simple at just “-Ofast” and nothing else.
In SPECint2017, the differences to the M1 are small. 523.xalancbmk is showcasing a large performance improvement, however I don’t think this is due to changes on the chip, but rather a change in Apple’s memory allocator in macOS 12. Unfortunately, we no longer have an M1 device available to us, so these are still older figures from earlier in the year on macOS 11.
Against the competition, the M1 Max either has a significant performance lead, or is able to at least reach parity with the best AMD and Intel have to offer. The chip however doesn’t change the landscape all too much.
SPECfp2017 also doesn’t change dramatically, 549.fotonik3d does score quite a bit better than the M1, which could be tied to the more available DRAM bandwidth as this workloads puts extreme stress on the memory subsystem, but otherwise the scores change quite little compared to the M1, which is still on average quite ahead of the laptop competition.
The M1 Max lands as the top performing laptop chip in SPECint2017, just shy of being the best CPU overall which still goes to the 5950X, but is able to take and maintain the crown from the M1 in the FP suite.
Overall, the new M1 Max doesn’t deliver any large surprises on single-threaded performance metrics, which is also something we didn’t expect the chip to achieve.
Facebook
Twitter
RSS
493 Comments
View All Comments
id4andrei - Monday, October 25, 2021 - link
Hey Andrei, on gaming, how does the API difference weigh in? I assume that Apple expects all devs to design their new games with Metal in mind. On the Windows side you have DX, OpenGL, Vulkan and the GPUs themselves are tuned to various quirks of these APIs.Ryan Smith - Monday, October 25, 2021 - link
Metal 2 is close enough in functionality and overhead to DX12 that I don't lose any sleep. But it is an API like any other, so devs need to be familiar with it to get the best performance from Apple's platform. Especially as Apple's GPU is a TBDR.Silma - Monday, October 25, 2021 - link
I don't understand how Intel, AMD and Qualcomm did not respond more urgently to the threat after the marketing of the M1.If they don't hurry, some people like me, who never considered switching to Apple, will entertain the idea.
willis936 - Monday, October 25, 2021 - link
It's 57 Bn transistors in a consumer SoC. How can companies who only sell SoCs compete with a company that sells their SoC at a loss?Ppietra - Monday, October 25, 2021 - link
"company that sells their SoC at a loss"????Apple is making profits when selling these machines so it cannot be selling at a loss.
What you could say is that Apple is not bound by the same manufacturing cost constrains as Intel or AMD because it doesn’t have to convince any other company to buy its chips.
phr3dly - Monday, October 25, 2021 - link
It would be really interesting to see the BOM for a MBP. It'll never happen of course, and direct comparisons to even an x86 MBP would be impossible.Certainly the vertical design gives Apple the opportunity to pocket what would otherwise be fat margins for Intel.
Ppietra - Monday, October 25, 2021 - link
Without a doubt Apple is taking advantage of not having to pay the for Intel (CPU) or AMD (GPU) margins.I am sure there will be some analyst trying to predict its cost – all iPhone BOM that you read do they same thing. Considering its size and some of the estimates for the previous M1 maybe somewhere close to 200 dollars for the M1 Max.
sirmo - Monday, October 25, 2021 - link
This chip is more expensive to manufacture than the 3090. It even has the wider memory bus. No way could you sell this as a component alone. Apple is clearly subsidizing this run to convince customers to stay for the transition.Ppietra - Monday, October 25, 2021 - link
How do you know that it is more expensive to manufacture than the 3090? And do you know how much the 3090 actually costs in order to come to such conclusion about "subsidizing"?? Wider memory bus has no relevance on how much the chip costs.Apple doesn’t sell computers at a loss, actually Apple almost certainly pays less for the chip than if it had to buy from Intel.
sirmo - Monday, October 25, 2021 - link
3090 is 28B transistors, this behemoth is 57B. As you go larger the yields get worse, so there is a bell curve to how high the costs ramp up. Also M1 Max is made on the most advanced 5nm node. Which is more expensive than Samsung's 8nm 3090 is made on.Like it would definitely cost more than a 3090, maybe even 3-4 more times more expensive.