News First Apple M1 Ultra Benchmark Posted, Nearly Matches Threadripper 3990X

whatisupthere

Distinguished
Apr 5, 2010
40
1
18,535
Impressive result especially for the power usage.
Not completely unexpected when considering the M1 ultra uses 114 billion 5nm transistors while the treadripper uses 39 billion 7nm transistors. The M1 ultra chip is huge.
 

Spanky Deluxe

Distinguished
Mar 24, 2009
515
7
18,985
I've been waiting for a machine just like this for literally years. Placed an order right after the keynote. I can finally replace my over decade old Mac Pro!
 

Makaveli

Splendid
I just looked up 3990x geekbench scores and there are numerous above 38000.... The M1 isn't even close. Why would you report that the 3990x only got 25000? Is this supposed to be a paid Apple article or something?

not sure what happened here but even wcctech got it right and that is a low bar to pass...

 
  • Like
Reactions: artk2219 and hasten
Is this really that impactful? Or just a benchmark.
No one credible uses Geekbench, at least as a multiplatform benchmark. It was created by the owner of a Mac review website, so any multiplatform comparisons may be skewed in favor of Apple. I can't think of any legitimate PC hardware review sites that use Geekbench as part of their test suite. It's also a synthetic workload, not directly based on real-world software, so it's not necessarily representative of real-world workloads either. It's best to wait until the hardware is out and people have tested it in real-world software before drawing any conclusions.
 
May 21, 2021
126
66
170
I love it how everyone compares x86 CPUs built to accept a wide array of user determined memory configurations to an entire system-on-chip design tuned to get maximum performance out of the pre-defined secondary components. I wonder how much faster zen 3 or alderlake would be compared to the m1 ultra if they too had dedicated 800Gbps memory or if the m1 ultra was limited to 3200mhz DDR4
 
A new Geekbench 5 result shows Apple's new M1 Ultra flagship going toe-to-toe with AMD's monster 64 core Threadripper 3990X, while consuming just a fifth of the power.

First Apple M1 Ultra Benchmark Posted, Nearly Matches Threadripper 3990X : Read more

Pffft. If you say so. It doesn't even eclipse the 3970X, and the 3990X is further above that. It's not to say the scores aren't impressive and all, but it's clear you're being a bit disingenuous here. As someone else has said and posted a link to, the Geekbench scores are contrary to what you've published.

Funny that you also made no mention of the price: $3999. That's a HELL of a lot of coin, specs be damned. And the boast of supposed 3090 level graphics performance? Guess we'll believe that when we see it actually tested in the wild given as that sounds an awful lot like Tarzan level chest thumping from Apple with absolutely no actual substance to back it. Then again, this is coming from the manufacturer so those kind of things are pretty par for the course.
 
Last edited:

PiranhaTech

Reputable
Mar 20, 2021
134
86
4,660
Is this really that impactful? Or just a benchmark.
Yes and no. Apple has done picking and choosing of benchmarks in the past.

Apple M1 has great IPC per watt. If it just matched something like the Core i5 8000 series, I would have said it's a good CPU.

I just checked out Geekbench, and a lot of the benchmarks actually do things that Apple probably does as part of the SoC instead of the main CPU, or has done main CPU optimization for iOS. Image compression, ML, PDF rendering, Camera processing, etc. It was hard to tell if this was using the main CPU, or other parts of the M1 SoC. Don't get me wrong: it's VERY clever of Apple of leveraging the SoC in a way that others can't.

You can get some information out of it, but I would be careful trusting synthetic benchmarks. The system will be coming out at $4000, so there will probably be loads that will be good for it, but I wouldn't trust a few benchmark as the M1 Ultra being powered by the Ghost of Steve Jobs until we get more data

On the other hand, the Ryzen 3990X is going to be replaced soon. This also means that the 3990X did well vs a CPU that might have been specifically optimized for those particular benchmarks, it says a lot about the Threadripper
 
Last edited:
  • Like
Reactions: artk2219 and gg83

watzupken

Reputable
Mar 16, 2020
1,007
507
6,070
Yes and no. Apple has done picking and choosing of benchmarks in the past.

Apple M1 has great IPC per watt. If it just matched something like the Core i5 8000 series, I would have said it's a good CPU.

I just checked out Geekbench, and a lot of the benchmarks actually do things that Apple probably does as part of the SoC instead of the main CPU, or has done main CPU optimization for iOS. Image compression, ML, PDF rendering, Camera processing, etc. It was hard to tell if this was using the main CPU, or other parts of the M1 SoC. Don't get me wrong: it's VERY clever of Apple of leveraging the SoC in a way that others can't.

You can get some information out of it, but I would be careful trusting synthetic benchmarks. The system will be coming out at $4000, so there will probably be loads that will be good for it, but I wouldn't trust a few benchmark as the M1 Ultra being powered by the Ghost of Steve Jobs until we get more data

On the other hand, the Ryzen 3990X is going to be replaced soon. This also means that the 3990X did well vs a CPU that might have been specifically optimized for those particular benchmarks, it says a lot about the Threadripper
I agree that we should always take benchmark results with a bunch of salt. It may or may not apply to someone, but we should know what sorts of testing the benchmark is running. Having said that, I feel the demand for this SOC will still be high, especially if it runs very well for your use case while sipping power.
 
  • Like
Reactions: artk2219
D

Deleted member 14196

Guest
Don’t count on any of it being true they haven’t even released what details they were using when they tested everything so it’s all smoke and mirrors and a bunch of BS

If They really wanted to prove it could beat a real GPU then they would play games with it and show you the actual numbers from the games which they haven’t and they won’t do probably ever
 

salgado18

Distinguished
Feb 12, 2007
928
373
19,370
I seriously doubt it does all that under 60W. Is 60W the limit, or the typical load? Intel cpus have 120W as spec, but on turbo they can go upwards of 200W. Does it do the same?
 
  • Like
Reactions: artk2219

clsmithj

Distinguished
Nov 30, 2011
40
6
18,535
I think the author meant the M1 Ultra nearly matches a Threadripper 3970X from 2019.

But that doesn't have the same appeal.

Much like the 12900K nearly matches a Threadripper 2990WX from 2018.

The point to actually gleam from all of this is that AMD is so far ahead that the competition is just now catching up to their dust.
 

Heat_Fan89

Reputable
Jul 13, 2020
440
211
5,090
Pffft. If you say so. It doesn't even eclipse the 3970X, and the 3990X is further above that. It's not to say the scores aren't impressive and all, but it's clear you're being a bit disingenuous here. As someone else has said and posted a link to, the Geekbench scores are contrary to what you've published.

Funny that you also made no mention of the price: $3999. That's a HELL of a lot of coin, specs be damned. And the boast of supposed 3090 level graphics performance? Guess we'll believe that when we see it actually tested in the wild given as that sounds an awful lot like Tarzan level chest thumping from Apple with absolutely no actual substance to back it. Then again, this is coming from the manufacturer so those kind of things are pretty par for the course.
That's the thing that always gets me about the chest thumping. Oh the M1 is sooooooo much faster than xyz CPU and GPU. So then where are the developers creating AAA titles for that most awesome hardware package? So, where's Elden Ring for the Mac? Yeah I know, crickets.

But can it play Crysis?
That's my problem with the Apple marketing and Press spin. After every Apple/Mac Event it turns into my peter is bigger than your peter. I'm still waiting for for all those AAA titles on Steam to show up in my Mac library. I'm not holding my breath.
 

spongiemaster

Admirable
Dec 12, 2019
2,273
1,277
7,560
If They really wanted to prove it could beat a real GPU then they would play games with it and show you the actual numbers from the games which they haven’t and they won’t do probably ever
No one is going to buy one of these systems to game on, so gaming benchmarks would be as worthless as the performance comparison charts Apple released with this announcement. I don't think Apple is trying to say this GPU is faster than a 3090 playing Cyberpunk 2077. It won't be, not even remotely close. They're talking about compute and rendering performance, but without any real world application benchmarks with actual numbers, the claims are pretty dubious.
 
  • Like
Reactions: artk2219

Dantte

Distinguished
Jul 15, 2011
161
58
18,760
No one is going to buy one of these systems to game on, so gaming benchmarks would be as worthless as the performance comparison charts Apple released with this announcement. I don't think Apple is trying to say this GPU is faster than a 3090 playing Cyberpunk 2077. It won't be, not even remotely close. They're talking about compute and rendering performance, but without any real world application benchmarks with actual numbers, the claims are pretty dubious.
Chicken and the egg:

Are there no gaming benchmarks because no one buying these will game on them, or...
is no one buying them for gaming because there are no benchmarks?

The 3rd option is: no one is buying them for gaming, and, no one is doing gaming benchmarks, because its a foregone conclusion that it sucks for gaming and cant hold a flame to its PC competitors?
 
  • Like
Reactions: artk2219

salgado18

Distinguished
Feb 12, 2007
928
373
19,370
Chicken and the egg:

Are there no gaming benchmarks because no one buying these will game on them, or...
is no one buying them for gaming because there are no benchmarks?

The 3rd option is: no one is buying them for gaming, and, no one is doing gaming benchmarks, because its a foregone conclusion that it sucks for gaming and cant hold a flame to its PC competitors?
I think it's the 4th option: no one is buying them for gaming because there are few Mac native games, and these machines are a terrible value for gaming compared to PCs. But now there's a 5th too: no native Windows x86 install. Buying a Mac to game is a waste of money.

Also, the GPU is dependent on the CPU, so we can't max the GPU power while choosing a reasonable CPU and adequate RAM. These come in closed packages, you want the best GPU you get 128GB of RAM that no game uses.
 

jthill

Reputable
Apr 14, 2020
7
9
4,515
I just looked up 3990x geekbench scores and there are numerous above 38000.... The M1 isn't even close. Why would you report that the 3990x only got 25000? Is this supposed to be a paid Apple article or something?

It's because there's a nearly 10,000-point difference between Linux and Windows scores. If you're wondering why all of the fastest supercomputers on the planet run Linux, this is your first clue. It also explains why the long knives are out for Geekbench.
 
Apple doesn't view their machines as gaming machines. They view them as machines to make creative content, which I don't think they view gaming completely as. A clear example in my eyes is the lack of Vulkan support. Also no indication has been given towards any GPU drivers for Apple Silicon chips. I suppose either of these things could change, and I'm sure Apple has answers as to why they aren't doing either of those, but it's clear that Apple does not view gaming as a priority.

And honestly, gamers don't view Mac as a priority either. A happy relationship
 
  • Like
Reactions: artk2219