Apple Silicon M1 MacBook Pro Earns 7508 Multi-Core Score in Cinebench Benchmark

[Coconut Bean]

The 3 GHz Tiger Lake has slightly higher single-core performance (~1570). The multi-core performance is slower than the M1 at ~6200, but that's with only half the CPU cores. The M1 performance is impressive, but it doesn't really "blow away" Intel at least in this benchmark.

I was under the impression that the M1 Macbook Pro has active cooling, but I may be wrong.

True. Apple's offering on performance is little better than last gen Intel, with efficiency significantly better. But it's not 3x or whatever. Maybe in some edge cases.

People should be happy with this update. Let's hope many will buy and weed out all the bugs so that I can buy a better product next year.

 

[laurieballard]

Wow, lots of squabbling on here which is to be expected I suppose.

Few people buying a MBA know (or care) about the difference between an M1 chip and an i5/i7/i7 or various other chips that I have never even heard of.

Personally I think the MBP is too similar to the MBA at the moment to justify a price difference but once the 14" comes out, I suspect the spec difference will change things in that respect. Having bought a 2020 MBP (4 port) this year, I see no reason to upgrade at the moment and assuming the 14" ones run the same M1 chip, I probably won't bother with that either. Sure, its quicker than the one I have but other than rendering 4K footage which I do maybe once a week, I am really not going to need the extra power for 95% of the stuff I use it for.

 

[Fl0r!an]

Did you really just try to compare your 10900k? Intel's page says that's a 10-core processor with a TDP 125 W. Which means you're cheating. (And how much RAM have you got?)

Comparing to an Intel desktop CPU doesn't make any sense. Desktop CPUs (and the Intel K-Ones especially) are barely tuned for efficiency, and Intel currently can't compete there anyway.

Comparing to an AMD Ryzen CPU with 15W TDP is fair, where the M1 leads by ~10-20% in multicore performance. This makes the M1 a great CPU, but certainly not the miracle some people have hoped for, especially considering the more advanced process node on the M1.
I guess Apple's main advantage will be the efficiency in low-load situations due ti its big.LITTLE architecture and the load of integrated hardware functions for video, machine learning and the like.

It'll also be interesting to see if Apple can match the single core speed on its upcoming desktop parts, as this won't scale with TDP as much as the multi core speed. Apple's architecture is very wide, so it surely can't clock as high as AMD's or Intel's.

 

[JaySoul]

Sorry but captain marvel is fake.

the MacBook bench scores are real!

How DARE you!!

 

[RegularGuy09]

Yes, when you compare it with what intel has to offer, it's great. But Intel is currently far, far behind AMD in multicore performance on 15-25W laptop chips. So if just being better than intel was the goal, Apple could have gone with R7 4800U, which would mean lot less hassle with porting software and giving up Windows compatibility. For some people this transition is going to be very painful, so I'm really hoping for something that doesn't just beat Intel (that is a dead horse already), but also AMD, with a significant margin to make it worth the hassle.

R7 4800U is a 15-25W processor. Apple M1 is a 10W processor. The more apt comparison would be with the chip inside the 4 port 13inch MBP. Lets ignore that though for fun, and even then, this 10W processor will have twice the battery life, and will probably run cooler.

This is a monumental leap. People trying to downplay it are salty, and scared.

 

[xpaulso]

I did trade-in for my 16 MBP for $1400. I'm just paying $100 more to 'upgrade' to the latest MBA.

Yikes I would have bought this for more than $1400. What were the specs?

 

[deevey]

Too many coordinated benchmark leaks on M1. Let me wait till hit the hands of normal users and wait to hear from them after few weeks of practical use.

They are doing a ton of real world tests e.g Logic, Luma, Xcode etc.... along with a ton of benchmarks on the Three M1 Machines vs 16inch MBP live.

So far very impressive results on FCP- and the MPA and MBP have all just been running on battery the whole time with the machines topping out at a relatively cool 35º under load!

 

[ds2000]

Cool, how many watts does your rig run off of? And why are you comparing the entry level to your machine anyway? Can you link us to your entry level computer with 15 hour battery life?

Because discussion forum

 

[Yarravillain]

Lol. This is what you can buy, so this is what will be compared. It's like saying, remember the CPU compared to is on 10nm, if it was on 5nm it would be better.

That's not how it works.

Right, and I hadn’t stated a view on that one way or another. It’s out there, so go nuts and compare. All I’m saying is that there’s a reason the M1’s multi-core doesn’t beat the other processor.

Side note: I now believe upon further inspection that the comparison is a MBP with the 6-core i7. That’s still 6 evenly-performed cores being benchmarked vs. the M1’s 4:4. The single-core benchmark, by comparison, is very nice.

 

[nikaru]

True. Apple's offering on performance is little better than last gen Intel, with efficiency significantly better. But it's not 3x or whatever. Maybe in some edge cases.

People should be happy with this update. Let's hope many will buy and weed out all the bugs so that I can buy a better product next year.

Little better? The new MBA is significantly more powerful than its 999$ predecesor and it is on par or outperform curren gen 2600$ 16' MBP. Basically, for 999$ you can now buy a machine that would have cost you almost three times more just a couple weeks ago. If this is not a huge leap in performance, I dont know what is it.

MBA is now the fastest ultrabook on the market by a wide margin in both CPU and GPU power. Period. Not to mention the almost two days battery life which is unmatched by any other ultrabook there is.

 

[AussieStu]

I can't think of how this makes Intel look good in any light.
The M1 is also fanless, no?

I think only in the Air. The pro has a fan and so does the mini..

 

[deevey]

Little better? The new MBA is significantly more powerful than its 999$ predecesor and it is on par or outperform curren gen 2600$ 16' MBP. Basically, for 999$ you can now buy a machine that would have cost you almost three times more just a couple weeks ago. If this is not a huge leap in performance, I dont know what is it.

MBA is now the fastest ultrabook on the market by a wide margin in both CPU and GPU power. Period. Not to mention the almost two days battery life which is unmatched by any other ultrabook there is.

M1's are not outperforming the 16' in some heavy video work, but for many other tasks - so far the comparison is 8gb machines though, maybe with 16gb things might improve a bit.

The Air is really kicking butt in most every tasks and benchmarks though -just 15% under the M1 pro in FCP, but its FANLESS.

Seriously Amazing feat.

 

[aphotechre]

I won't join the Bandwagon until Apple can beat:

The world’s fastest supercomputer, Fugaku, boasts several architectural innovations that may pave the way for even greater performance.
Boasting nearly 7.3 million cores and a speed of 415.5 petaFLOPS

 

[cibonak]

If i just want to send email, will this info help me?

 

[ngextreme]

I'm holding out to see the future offerings when they're actually replacing i7 and i9.
The M1 is replacing i3, i5 offerings, as far as I can tell (i.e., M1's 4 perf cores replacing 4 core version of these chips). AMD comparisons should be limited to their lower offerings (i.e., ignore higher end R5 and all of R7), until apple releases their higher end chips.

Ryzen 9 and Threadripper are the high end chips. R7 is the mid tier offering, R5 and R3 are the low end chips. They match the intel numbering so its easy to know what to compare against.

 

[amartinez1660]

You're right in that the MacBook Pro doesn’t seem entry-priced compared to the other M1 Macs. But the base Mac mini, which should perform about the same, does have an entry-level price. And both the M1 MBP and Mac mini take the place of entry-level models in their respective lineups, so that’s likely an indication of where the M1 is aiming.

I pointed out earlier that multicore doesn’t match up with that of one of the high-performing Intel chips out there. It doesn’t have to. Single-core performance is amazing, but people who need strong multi-core performance can look forward to that. It’s pretty impressive Apple can even churn out fairly high-performance processors — exceeding the low-TDP i9 in single-core and being competitive with recent i7s in multi-core — while keeping to its low TDP and power envelope. With powerful single cores and the 4:4 big.LITTLE cores, to boot. I’d say it’s fair for others to see it as potential for Apple’s mid- and top-range configs.

I would also argue that it doesn’t have “entry level” hardware (well, fine, the face camera is entry level) construction quality either.

I find it very unfair to compare an entry level macbook with an entry level ~$600 HP/Del/etc squeaky bending sounding plastic frame, noise capturing mic, connection dropping Bluetooth, inaccurate trackpad computer and somewhat dubious screen.

I have had these tiers in the past, several of them in my student days. They do have the specs, might have 8GB and some tons of GHz cpu model and really have nothing against them, they work, they do the job... but come on, to expect a MacBook to be also priced at that price range?

Now, price expectations aside, another question would be, could Apple profitably price them that low? Disrupt the whole market wide? The price has to be able to also support the huge influx in customer support, issues, returns, etc
I have found Apple to be very very lenient compared to others on the returns front for example.

 

[matt_k]

R7 4800U is a 15-25W processor. Apple M1 is a 10W processor. The more apt comparison would be with the chip inside the 4 port 13inch MBP. Lets ignore that though for fun, and even then, this 10W processor will have twice the battery life, and will probably run cooler.

This is a monumental leap. People trying to downplay it are salty, and scared.

M1 definitely scales up with more headroom, otherwise MacBook Pro would have same result in cinebench as Air, which it doesn't (about 15% faster). Single fan MBP thermal envelope can definitely sustain more than 15W. The question is, how much of this the M1 actually uses and how much headroom is there. Considering first reviews where people say it's virtually silent during Cinebench, I am cautiously optimistic.

 

[matt_k]

Actually, OP is right. A short benchmark like Geekbench is better for comparing the raw performance of CPUs.

The throttling effects you mention are dependent on the thermals of the system the CPU is installed in. A great CPU installed in a system that has insufficient thermals, thus getting a lower Cinebench score, is still a great CPU. It’s just been let down by a lack of cooling.

An example: compare the Cinebench scores of the M1 MacBook Pro to the fanless M1 MacBook Air.

By that logic ryzen 9 5800x is the best CPU, it's just let down by insufficient cooling in laptops. We exist in real world with thermal and power constraints and GeekBench likes to pretend that's not true. For workload such as mine, which involves sustained load, cinebench is much closer to real-world performance than geekbench.

 

[amartinez1660]

You are comparing an 8-core, 16-thread 4800U to a 8-core, 8-thread M1 (actually 4-cores+4-little cores (25% SPEC2006 of a big core each) that with very likely has half the TDP. If you don't think about TDP that would be like comparing a 3800X's 13000 score to your 4800U's 10000 score and saying the 4800U is nothing worth writing home about because it doesn't score as well as the 3800X with its 4-fold higher TDP. The fact that the 4800U does so well within a 25W TDP, even with a built-in GPU, is what makes the 4800U so amazing. That the M1 is only 25% slower while running half the number of threads with a quarter of the resources on half of them, while essentially incorporating the chipset as well, and doing it within half the TDP of the 4800U is at least as amazing to me.

This is spot on, really liked the available pool of resources (TDP, cores, performance per core) comparison to what it traded off in performance comprare to that (complete other class) CPU. And we haven’t mentioned that it won 10hrs more of battery life.

If we think on a “personal available resources level”, saying ‘I don’t care about TDP’ is similar to saying I don’t care about battery life, I don’t care about size, I don’t care about price. In that no constraints case, well, fine, go buy the 1hr lasting 6kg “laptop” monster, yes it will score better, or just go all in, get a super computer to fill a room with...

 

[Coconut Bean]

Little better? The new MBA is significantly more powerful than its 999$ predecesor and it is on par or outperform curren gen 2600$ 16' MBP. Basically, for 999$ you can now buy a machine that would have cost you almost three times more just a couple weeks ago. If this is not a huge leap in performance, I dont know what is it.

MBA is now the fastest ultrabook on the market by a wide margin in both CPU and GPU power. Period. Not to mention the almost two days battery life which is unmatched by any other ultrabook there is.

Here you go. A 15W laptop chip getting a significantly higher score (cb r23) compared to the MBP (not the MBA where the impact of thermal throttling is still a question).

AMD Ryzen 7 4800U - Benchmark, Test and Specs

AMD Ryzen 7 4800U - Benchmark, Geekbench 5, Cinebench R20, Cinebench R23, Cinebench R15 and FP32 iGPU (GFLOPS) benchmark results

www.cpu-monkey.com

Apple's offering is very competitive no doubt about that and comparing to their own thermally handicapped previous gen very very good.

A question to you, if the M1 is equivalent to whatever is in the 16" why wasn't it launched in the 16" as well? Why did they only launch entry-level products w M1? (Answer, it's not equivalent)

 

[matt_k]

Here you go. A 15W laptop chip getting a significantly higher score (cb r23) compared to the MBP (not the MBA where the impact of thermal throttling is still a question).

AMD Ryzen 7 4800U - Benchmark, Test and Specs

AMD Ryzen 7 4800U - Benchmark, Geekbench 5, Cinebench R20, Cinebench R23, Cinebench R15 and FP32 iGPU (GFLOPS) benchmark results

www.cpu-monkey.com

Apple's offering is very competitive no doubt about that and comparing to their own thermally handicapped previous gen very very good.

The big question here is what would 4800u result be if ran with exact same power constraints as M1. So far I've only seen one video where they run whole 10 minutes of cinebench R23 and it seemed that MBP was completely silent the whole time (this needs more corroboration), which would suggest that it might not be using the whole thermal headroom. Also passively cooled Air has score only 15% lower. This leaves me optimistic for something like M1X.

Generally those over 10000 points renoir scores are achieved while bursting way beyond the nominal TDP.

 

[jmgregory1]

Here you go. A 15W laptop chip getting a significantly higher score (cb r23) compared to the MBP (not the MBA where the impact of thermal throttling is still a question).

AMD Ryzen 7 4800U - Benchmark, Test and Specs

AMD Ryzen 7 4800U - Benchmark, Geekbench 5, Cinebench R20, Cinebench R23, Cinebench R15 and FP32 iGPU (GFLOPS) benchmark results

www.cpu-monkey.com

Apple's offering is very competitive no doubt about that and comparing to their own thermally handicapped previous gen very very good.

A question to you, if the M1 is equivalent to whatever is in the 16" why wasn't it launched in the 16" as well? Why did they only launch entry-level products w M1? (Answer, it's not equivalent)

Apple didn’t launch the M1 in the 16” MBP for the same reason Chevy doesn’t put a 1.3 liter 3 cylinder engine in a Corvette. Sure, they could turbocharge the 1.3l and generate big power numbers, but they’ve got an even more powerful option to include, which is true for Apple too.

That is the crazy thing with their chips. Their base model performs at levels above their class, and their next line of mid level chips will perform even higher.

People who upgrade now to get this new level of performance and efficiency will want to upgrade again because the gains will be so big yet again. And the doubters will deny and complain and suggest that Apple chips may be powerful but they can’t compete with some super computer, because that’s all they’ll have left to compare them against.

At least that is my guess for how things will continue to play out

 

[PortoMavericks]

Apple reportedly has a perpetual ARM license.

I’m not saying it doesn’t.

Apple reportedly has a perpetual ARM license.

Good news for Apple because now Nvidia owns it.

 

[PickUrPoison]

By that logic ryzen 9 5800x is the best CPU, it's just let down by insufficient cooling in laptops. We exist in real world with thermal and power constraints and GeekBench likes to pretend that's not true. For workload such as mine, which involves sustained load, cinebench is much closer to real-world performance than geekbench.

That’s right: if you put a 100W CPU in a case that supports 50W of cooling, it would be tremendously let down by that insufficient cooling.

Now compare that to a 5800X in a machine that can easily remove 100W of heat.

Geekbench might show the same score, while Cinebench would show a drastic difference. Does that prove the 5800X is drastically better than the 5800X?

Like I said, a short benchmark like Geekbench is better than a longer benchmark like Cinebench for comparing raw performance of CPUs.

If you really want to see what the M1 can do, better to benchmark the mini (which enclosure we know can handle 65W) than the 10W thermals of the fanless MacBook Air.

 

[matt_k]

Geekbench might show the same score, while Cinebench would show a drastic difference. Does that prove the 5800X is drastically better than the 5800X?

It proves that 5800X is absolutely useless in a thin and light laptop. Which is something you couldn't tell by just looking at the geekbench score.

Like I said, a short benchmark like Geekbench is better than a longer benchmark like Cinebench for comparing raw performance of CPUs.

But raw performance is meaningless if you can't sustain it.

If you really want to see what the M1 can do, better to benchmark the mini (which enclosure we know can handle 65W) than the 10W thermals of the fanless MacBook Air.

I wouldn't expect it to be significantly different. Looking at the 15% difference between pro and air, it doesn't seem to be using all the available headroom of the pro (IIRC the single fan 13" has no problem sustaining 20-25W). So this may be pretty much what M1 can do, and considering what it does in a passively cooled in Air, I'm actually quite impressed and hopeful for something like M1X with more cores.