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.
493 Comments
View All Comments
darwinosx - Friday, November 5, 2021 - link
Everything you said is wrong.C@illou - Tuesday, October 26, 2021 - link
Slight correction, Vulkan works great on windows (and also works on Linux, but that counts the same as "SteamOS"), that makes it the most compatible API.xeridea - Tuesday, October 26, 2021 - link
Vulkan runs on everything.Qozmo - Tuesday, October 26, 2021 - link
Worth mentioning that MoltenVK exists officially from Khronos Group which layers Vulcan on top of the MetalAPI enabling Vulcan apps to run on MacOS/iOSWrs - Monday, October 25, 2021 - link
Is it just me or does that make no economic sense? When I’m AAA gaming (flashy visuals, complex scenes, high fps) I don’t feel as if I’m looking for light and cool or portable. I’d be on a desk flinging a mouse, or wielding a controller in front of a TV.michael2k - Monday, October 25, 2021 - link
Maybe it isn't clear, but 'light and cool' means there is lots of headroom for overclocking. From the third page:Power Behaviour: No Real TDP, but Wide Range
Apple doesn’t advertise any TDP for the chips of the devices – it’s our understanding that simply doesn’t exist, and the only limitation to the power draw of the chips and laptops are simply thermals. As long as temperature is kept in check, the silicon will not throttle or not limit itself in terms of power draw.
You can imagine that in a desktop, with far better cooling and far more available power, that the M1P/M1M might grow well beyond the 92W of observed package power. The Mac Pro with 28 cores and 2 GPUs today will allow the CPU to consume 902W, there is a lot of space for performance to grow!
So imagine 10x more performance from a desktop Mac with 10 M1P in some kind of fabric (100 cores and 320 GPUs!) or a much smaller number of M1P, maybe 4 (40 cores and 128 GPUs) with each allowed to consume 2.5x as much power
sean8102 - Tuesday, October 26, 2021 - link
Problem is developer support. It seems there are only 2 "AAA" macOS AND ARM native games.https://www.applegamingwiki.com/wiki/M1_native_com...
That has to improve A LOT for getting a ARM Mac for gaming to make any sense. Otherwise you're always taking the performance hit of Rosetta 2. Plus not many AAA games are releasing for macOS since they announced the switch to ARM.
The chips are amazing in terms of performance and efficiency, but getting a mac esp a ARM based one for gaming wouldn't make much sense. At least for now and not unless developer support improves A LOT.
AshlayW - Tuesday, October 26, 2021 - link
Clock speeds do not scale with power consumption and Firestorm cores are not designed to reach high clock speeds, these cores would likely not break 3.5 if overclocked (wide, dense design for perf/W). AMD / Intel / NVIDIA's 5nm-class processors will put Apple back in its place for people wanting to NOT be locked into a walled garden from a company adamant on crushing consumer rights. It's just a shame that Apple's silicon engineers are so freakin' good, they're working for the wrong company (and hurting human progress by putting the best wafers/chips in Apple products).valuearb - Tuesday, October 26, 2021 - link
Lol apple is responsible for more human progress than all the other PC makers combined.MooseNSquirrel - Friday, October 29, 2021 - link
Only if your metric is marketing based.