Thursday, November 12th 2020

Apple's M1-Based MacBook Air Benchmarked

When Apple announced that they are going to switch their Mac lineup from Intel-based x86 processors to the custom "Apple Silicon," everyone was wondering how the new processors will look and perform. To everyone's luck, Apple has just a few days ago announced its first Apple Silicon custom processor for MacBook. The M1, as the company calls it, is their first processor designed for higher-power and performance tasks The M1 features eight CPU cores (four high-performance and four-high efficiency) paired with eight cores dedicated to the graphics. On the die, there is also a 16-core neural engine made to accelerate machine learning tasks found in the new applications.

Today, we are getting the first GeekBench 5 CPU benchmarks that showcase just how far Apple has come with its custom design. What we have is the M1 processor found in MacBook Air. This Mac model features a passive cooling system, cooling a CPU with a base frequency of 3.2 GHz. The system scored 1719 points in the single-core result, and 6967 points in the multi-core result. The single-threaded results measure itself with some of the highest-end offerings from Intel and AMD, while the multi-threaded results are very good given the mix and match of small and big cores.
Source: GeekBench 5
Add your own comment

115 Comments on Apple's M1-Based MacBook Air Benchmarked

#51
Valent117
windwhirl
That's starting to look like another Userbenchmark
just like cinebench R15 to R20, geekbench 4 was not scaling well in the high performance range
Posted on Reply
#52
Rahnak
Searing
wrong, go back to 2010 where you come from

Geekbench is great for different OS and ISA comparisons, and is also not memory dependent anymore

next week there will be a lot of surprised people who have had their heads in the sand for the last 5 years

trends matter, Apple has been improving more than Intel every year for 5 years, that is why they are ahead now
So you're telling me you believe the upcoming fanless, likely single digit TDP Macbook Air completely trashes your 10900 in single threaded use?

I'm not saying it's a bad chip, it seems very promising and Apple made some very bold claims. But c'mon.
Posted on Reply
#53
londiste
Rahnak
So you're telling me you believe the upcoming fanless, likely single digit TDP Macbook Air completely trashes your 10900 in single threaded use?
Why not? Single thread power usage is 10-15W for Intel/AMD CPUs. M1 seems to have 10W TDP as tested, at full node smaller manufacturing process (2 full nodes when comparing Intel and 10900).
Posted on Reply
#54
theoneandonlymrk
Rahnak
So you're telling me you believe the upcoming fanless, likely single digit TDP Macbook Air completely trashes your 10900 in single threaded use?

I'm not saying it's a bad chip, it seems very promising and Apple made some very bold claims. But c'mon.
It has more special purpose hardware than any ,I think other processor outside of FPGAs, in some workloads it is a very good and capable chip, possibly enough for most, never enough for some still.
Posted on Reply
#55
xorbe
SPECint is half compiler benchmark where they spend big man hours with specific flags for each test. Geekbench is useless. We have to wait.
Posted on Reply
#56
Rahnak
londiste
Why not? Single thread power usage is 10-15W for Intel/AMD CPUs. M1 seems to have 10W TDP as tested, at full node smaller manufacturing process (2 full nodes when comparing Intel and 10900).
I just find that very hard to believe, is all. I'll know soon enough, I guess. In the event that they do I think both Intel and AMD can just close up shop and go home.
theoneandonlymrk
It has more special purpose hardware than any ,I think other processor outside of FPGAs, in some workloads it is a very good and capable chip, possibly enough for most, never enough for some still.
For sure, but I mean for general purpose, across the board. This chip will shine in some specific tasks that Apple really optimizes for. Web browsing for instance, is something that the A family of chips are particularly good at, because it's one of their main uses, so I'm expecting the M1 to beat x86 chips in web benchmarks.
Posted on Reply
#57
seth1911
Geekbench is a pice of sh*t, sometimes i got MP 1k Points different with the same device (Geekbench 4.4).

But i like it, that Apple makes now a real golden cage.
Posted on Reply
#58
Minus Infinity
Can I get one woth a Quadro GPU, they do have Pro in the title afterall.

Give us some benchmarks worth shit, I couldn't care less how fast it is for opening browser I actually do real work. If this thing can run Matlab, Mathematica, Comsol, Ansys etc faster than any x86 or close at such low power and sustain that performance when getting hammered all day, then sign me up. But unless I can have a discrete GPU for CUDA it won't happen. How many years before we get discrete GPU's and drivers for ARM architecture in any new age Apple computer. It will only be a matter of time before the desktops also dump x86.
Posted on Reply
#59
windwhirl
Minus Infinity
Can I get one woth a Quadro GPU, they do have Pro in the title afterall.

Give us some benchmarks worth shit, I couldn't care less how fast it is for opening browser I actually do real work. If this thing can run Matlab, Mathematica, Comsol, Ansys etc faster than any x86 or close at such low power and sustain that performance when getting hammered all day, then sign me up. But unless I can have a discrete GPU for CUDA it won't happen. How many years before we get discrete GPU's and drivers for ARM architecture in any new age Apple computer. It will only be a matter of time before the desktops also dump x86.
Considering that Apple is promoting their own GPU and Neural stuff, they probably don't care?
Posted on Reply
#60
Searing
Rahnak
So you're telling me you believe the upcoming fanless, likely single digit TDP Macbook Air completely trashes your 10900 in single threaded use?

I'm not saying it's a bad chip, it seems very promising and Apple made some very bold claims. But c'mon.
At the same frequency? You bet. 15 billion transistors matters.
Vayra86
Can we get a community effort going to just straight up shitlist this bench from any news on TPU? Or alternatively, demand a neutral bench is run alongside it...
Maybe you should actually understand why? Geekbench 5 is much more intensive and is 64 bit only. There is absolutely nothing wrong with Geekbench 5, the old ones were too dependent on memory scores etc. and didn't give the CPUs a workout.
Posted on Reply
#61
Flanker
Minus Infinity
Can I get one woth a Quadro GPU, they do have Pro in the title afterall.

Give us some benchmarks worth shit, I couldn't care less how fast it is for opening browser I actually do real work. If this thing can run Matlab, Mathematica, Comsol, Ansys etc faster than any x86 or close at such low power and sustain that performance when getting hammered all day, then sign me up. But unless I can have a discrete GPU for CUDA it won't happen. How many years before we get discrete GPU's and drivers for ARM architecture in any new age Apple computer. It will only be a matter of time before the desktops also dump x86.
windwhirl
Considering that Apple is promoting their own GPU and Neural stuff, they probably don't care?
I think Apple wants to make all their users use Metal for GPGPU apps as well as rendering.
Posted on Reply
#62
timta2
illli
No, they are still overpriced. The $699 mac mini only comes with 8Gb ram (lol) and 256Gb ssd (lol again).
Oh, you might think you can upgrade this yourself? NOPE. Everything is soldered onto the mobo.
The $899 mini is STILL the same 8Gb ram (pathetic) and 512b ssd (so basically +$200 for 256Gb upgrade)
It's more than what most normal people need and when the day comes that they need more, they can just buy a new one.
Posted on Reply
#63
Vya Domus
Searing
There is absolutely nothing wrong with Geekbench 5, the old ones were too dependent on memory scores etc. and didn't give the CPUs a workout.
Lol what a joke, I've heard the same excuse every time, "the old one was just bad but the new one is better". Of course it has to be better, it gives Apple's chips higher scores after all, obviously. How come that never happens for other SoCs or chips, must be a coincidence.
Searing
15 billion transistors matters.
15 billion transistors for everything, not for just for the processors.
Posted on Reply
#64
Vayra86
Searing
At the same frequency? You bet. 15 billion transistors matters.



Maybe you should actually understand why? Geekbench 5 is much more intensive and is 64 bit only. There is absolutely nothing wrong with Geekbench 5, the old ones were too dependent on memory scores etc. and didn't give the CPUs a workout.
You realize this disqualifies it to begin with? If a new version is so radically different in how it scores, it immediately invalidates everything that came before it..... until Geekbench 6.
Posted on Reply
#65
Rahnak
Searing
At the same frequency? You bet. 15 billion transistors matters.
Nonono. Not at the same frequency. At stock. The fanless machine beats your desktop 10900 at single threaded use as is. That's the narrative you're signing under.

The image shows the M1 in the air at 3.2Ghz scores 1739.
The best score I could find for the 10900 (max boost single thread is 5.2Ghz) is 1535.
Posted on Reply
#66
Vya Domus
Rahnak
Nonono. Not at the same frequency. At stock. The fanless machine beats your desktop 10900 at single threaded use as is. That's the narrative you're signing under.

The image shows the M1 in the air at 3.2Ghz scores 1739.
The best score I could find for the 10900 (max boost single thread is 5.2Ghz) is 1535.
A wider core at lower frequencies will be more power efficient than a narrower core at higher frequency even if they generate similar performance, that's all. The problem Apple will face going forward is that their cores will scale horrendously with frequency. We can already see that, 3.2 Ghz is dismal, let's just say we wont be seeing one of their chips hit 4.5+ Ghz anytime soon and the problem is that clock speed does matter. There is code that simply cannot benefit from OoO execution and wide front/back ends.

In other words the performance gains are mostly over for them, they wont be able to make much wider designs or higher frequency. I guess that's why they decided to make the jump now, otherwise in the future their chips are going to look less and less impressive.
Posted on Reply
#67
JalleR
But can it run Crysis ?
Posted on Reply
#68
windwhirl
JalleR
But can it run Crysis ?
How about "NO"?
Posted on Reply
#69
dyonoctis
It's for the A12z, but unless the m1is at least x 2 times faster in multithread, the multicore score of the m1 in geekbench won't be representative for every workloads. The m1 is supposed to be as faster than the 9880h in mt...

Posted on Reply
#71
foedecide
Why? I appreciate your sentiment, lol, but if we are idiots, what is your counter argument. I mean you'd need to be an idiot not to provide one?
Vya Domus
Here is why Geekbench is dogshit :



Every time other SoCs inch closer to Apple's the app gets updated and a chasm appears again between their chips and everyone else's.

In 4.4 there was less than 10% difference between A13 and Exynos 990, in 5 that somehow became a colossal 50%. This happens every single time a new version appears, without exception. Only an idiot would take these numbers for granted and not realize that this benchmark is always optimized specifically for Apple's chips.
Unyet most people would look at day to day performance in their hands, majority of people arent benchmark wankers?
sweetsuicide
I totally agree with you. This is AArch64, not the day-to-day application performance.
Wow, progess is all I have to say, see you on the otherside when other benchmarks confirm the obvious....
thevoiceofreason
You mean like for example SPECint, here showing Apple's previous A14 chip powering Iphone 12?



I'm very curious about Cinebench R23 results of the M1.
timta2
It's more than what most normal people need and when the day comes that they need more, they can just buy a new one.
Research yourself, OP clearly dosent understand the differences or I'd argue the argument.
Rahnak
Nonono. Not at the same frequency. At stock. The fanless machine beats your desktop 10900 at single threaded use as is. That's the narrative you're signing under.

The image shows the M1 in the air at 3.2Ghz scores 1739.
The best score I could find for the 10900 (max boost single thread is 5.2Ghz) is 1535.
These people has a vested interest in years of anti-Apple sentiment, you cant expect facts to dissuaded that. This is the equivalent of Trump supporters backing Trump stance on Covid-19.
Vya Domus
let's just say we wont be seeing one of their chips hit 4.5+ Ghz anytime soon and the problem is that clock speed does matter
OK, so where is your proof on that q
Vya Domus
The problem Apple will face going forward is that their cores will scale horrendously with frequency. We can already see that, 3.2 Ghz is dismal, let's just say we wont be seeing one of their chips hit 4.5+ Ghz anytime soon and the problem is that clock speed does matter.
OK, so whats your authority on that statement? Care to expand with factual evidence? Anything other than what is clearly your own personal opinion?
Vayra86
Can we get a community effort going to just straight up shitlist this bench from any news on TPU? Or alternatively, demand a neutral bench is run alongside it...
Never considered Apple consistently, i.e. yearly unlike Intel improve hardware and performance noticeably. I'm guessing not as you seem to prefer some sort of conspiracy theory as to why .....
Posted on Reply
#72
Vya Domus
foedecide
Why? I appreciate your sentiment, lol, but if we are idiots, what is your counter argument. I mean you'd need to be an idiot not to provide one?
I'd be the third time I would explain it, I can't do this forever, if someone still thinks I haven't provided an argument or if this didn't at least rise some eyebrows then they deserve the label of an idiot as far as I'm concerned. Look, don't let me spoil your enjoyment of marveling at those pretty charts, I don't have any interest in convincing anyone.
foedecide
OK, so where is your proof on that
The proof is in the power characteristics of integrated circuits and microarchitecture design. Everyone well versed in those can see that. Narrow cores = highly scalable , wide cores = not so much.
foedecide
OK, so whats your authority on that statement?
The hell you on about. Want me to call Jim Keller or what ? Clearly I had enough authority to annoy you with my ideas, I am fine with that.
Posted on Reply
#73
foedecide
Vya Domus
I'd be the third time I would explain it, I can't do this forever, if someone still thinks I haven't provided an argument or if this didn't at least rise some eyebrows then they deserve the label of an idiot as far as I'm concerned. Look, don't let me spoil your enjoyment of marveling at those pretty charts, I don't have any interest in convincing anyone.



The proof is in the power characteristics of integrated circuits and microarchitecture design. Everyone well versed in those can see that. Narrow cores = highly scalable , wide cores = not so much.



The hell you on about. Want me to call Jim Keller or what ? Clearly I had enough authority to annoy you with my ideas, I am fine with that.
Honestly, you made me laugh with your idiotic personal opinion, so obviously you cant back your claims?
foedecide
Honestly, you made me laugh with your idiotic personal opinion, so obviously you cant back your claims?
I mean the reality (which you obviously wont admit too) is - in the real world, i.e. users - the M1 provides better performance verbatim, you then have to add, I may add, lol, completely separately better performance / battery life than any other laptop out there - not just comparable.
Vya Domus
Everyone well versed in those can see that. Narrow cores = highly scalable , wide cores = not so much.
Nice quote, but where actually is your proof on that? Clearly you are missing the point and making a whole host of presumptions both on me and Apple silicon?
Posted on Reply
#74
Vya Domus
foedecide
Honestly, you made me laugh with your idiotic personal opinion
I gave many detailed explanations on a whole bunch of things throughout this thread.

You ? Fuck all, at the moment, you just keep repeating "back up your claims" and nothing else like an actual idiot. I already wasted enough time with you, off to the ignore list you go.
Posted on Reply
#75
foedecide
Vya Domus
I guess that's why they
Oh wait, you guess, so you are basing this on your own personal opinion and no real knowledge???
Vya Domus
I gave many detailed explanations on a whole bunch of things throughout this thread.

You ? Fuck all, at the moment, you just keep repeating "back up your claims" and nothing else like an actual idiot. I already wasted enough time with you, off to the ignore list you go.
Sorry for asking you to provide actually evidence to your personal claims, I have learned not to challenge people who claim to know better, thank you.
Posted on Reply
Add your own comment