Is anyone else dissapointed in single core performance of the M1 Pro/Max?

[ctjack]

I don't find myself limited with single core performance. I am more limited with 8gb of ram on my Air. At the same time i don't feel like upgrading to 14, 16" just because single core stayed the same, which is 80% of my daily use - all of the taks are single core unless you really do some multicore job everyday like video editing/rendering.

 

[crazy dave]

Which I expected the bigger charger and thicker and heavier enclosure housing a new cooling system to accomplish, thus leading to better (sustained) single-core performance. I would have preferred a MacBook Air (a year ago) but since the SSD size was restricted to 2 TB I waited, and it would have been nice if the delay and added bulk meant better single-core performance. As I said though, I’m only mildly surprised, and only a little annoyed.




I don’t think the processors should be judged on clock speed, they should be judged on performance. It seems reasonable to me that the highest-end machine would have better sustained single-core performance than the most basic one. Like it or not, there are lots of common tasks that for whatever reason are not multithreaded, and where faster execution is beneficial.

I am sure I will be perfectly happy with what I have ordered, but the my gut feeling about paying $5000 for it would have been better if it had been able to (single-core) outperform an iPad.

The MBP does have a better sustained performance than an iPad Pro. They just have the same peak performance in single core (again actually the MBP should be better for memory intensive stuff but are otherwise identical peak). What you’re asking for is higher single core peak performance which as I said only comes through higher clocks and at the cost of much higher energy usage and a less efficient chip. Apple is designing their cores so that the same core can be used from the iPad Pro to (so far) the MBP and everyone gets the best latency with the best efficiency but differ in bandwidth.

Now I disagree with the other poster’s notion that bandwidth >>> latency for every application, in reality single core is indeed important and I agree often more so for day to day tasks. However! The new MBP are about bandwidth, not latency. They have the same great latency of course but their focus is on providing incredible bandwidth without sacrificing energy usage. Maintaining the clocks helps keep wattage down (which btw helps with sustained performance - especially in a mobile package which the MBP still is).

Further, the reality is that AMD and Intel differentiate by MHz in order to still sell their weaker silicon and justify high prices on higher silicon to create a massive product range that they can hawk to businesses. That is not Apple’s market. And while I understand your world view, after all that’s what we’ve been conditioned to expect for so long, overall it’s been a massive negative for consumers and the industry. It’s what in part led those companies, especially Intel, to the design space they’re currently in - why battery and plug-in performance are so different, why the various turbo modes have blossomed into a confusing array of wtf, why the actual power used is so very different for the same chip from different laptop makers which you won’t know unless you see a review, and why Apple’s new M1 chips are 6-7x better perf/W than an Intel TGL chip while also out performing it. It’s not a good design philosophy and on top of that even AMD and Intel hit limits like the 5900 having better gaming performance than the 5950. Or Intel having to offer a “new” top cpu tier that was actually lower than its old one because they couldn’t actually fabricate enough of the top chip to sell any. It was a pointless bracket.

That the lowly iPad Pro has the same burst performance as a MBP is a good thing. Now will Apple continue this design all the way up to the Mac Pro? Dunno. But I wouldn’t be surprised if they did. And even if they don’t it won’t be a big speed boost - like A14 to M1 probably 10% or less (if it’s still firestorm).

 

[yitwail]

I don’t think the processors should be judged on clock speed, they should be judged on performance. It seems reasonable to me that the highest-end machine would have better sustained single-core performance than the most basic one. Like it or not, there are lots of common tasks that for whatever reason are not multithreaded, and where faster execution is beneficial.

From what I read about Geekbench, it’s not a test of sustained performance, because it just runs for a few minutes, which in most cases wouldn’t be long enough to differentiate between devices with different thermal characteristics, so the similar single core scores between M1, M1 pro, and M1 max does not imply that a 13 inch MBP or MBA would have the same sustained single core performance as a 14 or 16 inch 2021 MBP.

 

[crazy dave]

From what I read about Geekbench, it’s not a test of sustained performance, because it just runs for a few minutes, which in most cases wouldn’t be long enough to differentiate between devices with different thermal characteristics, so the similar single core scores between M1, M1 pro, and M1 max does not imply that a 13 inch MBP or MBA would have the same sustained single core performance as a 14 or 16 inch 2021 MBP.

This is completely correct. GB measures burst performance as do most single core benchmarks. Only a few measure sustained performance - especially for single core. Now I dunno how much throttling an iPad Pro actually does on a single core, but whatever it is - it could even be zero ?‍♂️, it’ll definitely be zero on the MBP.

 

[Wando64]

and thats disapointing .

They are identical cores.
What were you expecting?

 

[piattj]

Any performance indicators re disk access speeds MBA M1 vs MBP M!Pro/Max? I think I read that Max outperformed Pro, but what about comaparison with MBA M1?

 

[darngooddesign]

No because the point of these Macs is multi-core, not single-core. If you want faster single-core, wait for the M2s.

 

[Kazgarth]

View attachment 1882566
ill share with you some mbp 14 bench if i can ,but im just surprised at the results this guy got .this is mba m1

Turn of any content blockers/add-ons for the page before running the test to get the full score.

 

[zarathu]

I use Affinity Photo. AP has their own benchmarks, and they say that the M1Pro, is the fastest machine they ever tested, bar none.

 

[rezwits]

Nah, man that's the whole point of why Moore's Law DIED. A Single Chip got shrunk as small as possible (and then a little bit more) and then the temps with speed, with power draw, got maxed out, scientifically speaking. Then you have RISC vs CISC, where single core RISC loses out a little, but now is smaller than the Intel CISC.

The true power of the whole future of Apple with the ARM line of SoC is that they are developing SERIOUS Concurrency APIs/Frameworks for BOTH CPU and GPU and ML units.

So they are gonna keep REALLY maxing out CPU cores and making sure people multi-thread their software as much as possible.

I have an App that I am working on, that when on my older Intel (15"-2019) with 8 core (16 core virtual), the render takes 5 minutes, with the MacBook Pro (Pro) 8 core the render takes 50 seconds! 6x IMPROVEMENT!!! and the Fans are all blaring and the memory is all nuts, on the Intel, on the M1P whisper quiet.

If I lower the settings a render that takes 60 seconds, takes 10 second. This is REALLY gonna make my damn DAY!!

Best of luck dudes!

(nb, unless something really kicks in and requests multiple cores, I am seeing the Efficiency Cores doing most of the work)

 

[Love-hate 🍏 relationship]

Turn of any content blockers/add-ons for the page before running the test to get the full score.

i did ,i even booted macOS in safe mode