Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

EugW

macrumors G5
Original poster
Jun 18, 2017
14,882
12,857

Somebody elsewhere posted Apple A12Z (not M1) scores too. I can't vouch for them though, so take them with a grain of salt. Look for M1 scores soon though.

CinebenchR23M1.jpg


CinebenchR23i7-9750H.jpg


EDIT:

M1 Geekbench scores here:

MacBook Air (1687/7433):


Mac mini:

 
Last edited:

EugW

macrumors G5
Original poster
Jun 18, 2017
14,882
12,857
Why the Windows screenshot?
For single-core score comparison. A12Z is getting about the same score as i7-9750H.

Is that the DTK?
Seems so, but like I said, I can't vouch for this. (I guess the guy running the dev kit didn't want to be outed, because Apple doesn't allow benchmarking the dev kit.)
 

ssgbryan

macrumors 65816
Jul 18, 2002
1,488
1,420
Why the Windows screenshot?
To show multi core performance.

Looking at that - faster than a 4th generation quad core i7 (7 years old) mobile chip, but slower than a current (11 gen) quad core mobile chip.
 

ssgbryan

macrumors 65816
Jul 18, 2002
1,488
1,420
If it isn't an M1, then it is irrelevant.

Although for the iPad pro, that is actually pretty good.
 

senna95

macrumors newbie
Jun 29, 2017
11
5
So what is the R23 M1 multicore? I get 7920 on my i9 16 inch running macOS.

edit: nevermind I just saw it. 4530 is really impressive for the TDP.
 

NewUsername

macrumors 6502a
Aug 20, 2019
589
1,322
Well, the Geekbench score of the i7-9750H is similar to the one of the A12Z in iOS so that makes sense I guess.
 

Serban55

Suspended
Oct 18, 2020
2,153
4,344
Again apple compared and show the gains of the Macbook air with
"
  1. Testing conducted by Apple in October 2020 using preproduction MacBook Air systems with Apple M1 chip and 8-core GPU, as well as production 1.2GHz quad-core Intel Core i7-based MacBook Air systems, all configured with 16GB RAM and 2TB SSD. Tested with prerelease Final Cut Pro 10.5 using a 55-second clip with 4K Apple ProRes RAW media, at 4096x2160 resolution and 59.94 frames per second, transcoded to Apple ProRes 422. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Air
And for the macbook pro the comparasion was with

Testing conducted by Apple in October 2020 using preproduction 13-inch MacBook Pro systems with Apple M1 chip, as well as production 1.7GHz quad-core Intel Core i7-based 13-inch MacBook Pro systems, all configured with 16GB RAM and 2TB SSD
 
  • Like
Reactions: KPOM

senna95

macrumors newbie
Jun 29, 2017
11
5
Sure, but I was interested to see what is possible with these chips compared to the best mobile/desktop chips available right now and Cinebench is one of the only 'fair' benchmarks to compare them.
 

leman

macrumors Core
Oct 14, 2008
19,520
19,670
So what is the R23 M1 multicore? I get 7920 on my i9 16 inch running macOS.

edit: nevermind I just saw it. 4530 is really impressive for the TDP.

We don't know the M1 scores. The above scores are presumably for the A12Z in the DTK.
 

IvanKaramazov

macrumors member
Jul 23, 2020
32
49
This is actually fascinating to me, as Cinebench is one of those benchmarks that people are always championing at the same time they complain about Geekbench as being unreliable cross-platform. If these numbers are accurate for the A12Z DTK (which is far from proven I suppose), then it suggests that Geekbench and Cinebench track pretty closely on Arm as well. The single core score of the A12Z is right between the ST for the two x86 processors on either side, and same for the MT.
 

leman

macrumors Core
Oct 14, 2008
19,520
19,670
Those are pretty much the same as the M1 right? At least within 10-25%.

Nope. M1 is built on the A14 architecture which is full two generations and one node shrink ahead of A12Z. In Geekbench 5, A14 is about 40-50% faster than the A12. Frankly, I will be very disappointed if M1 ends up slower than the Tiger Lake CPUs. I expect single-core score of at least 1600
 

senna95

macrumors newbie
Jun 29, 2017
11
5
Nope. M1 is built on the A14 architecture which is full two generations and one node shrink ahead of A12Z. In Geekbench 5, A14 is about 40-50% faster than the A12. Frankly, I will be very disappointed if M1 ends up slower than the Tiger Lake CPUs. I expect single-core score of at least 1600
Ah yes you're completely right. I thought it was an A14Z for some reason. Interested to see if it, for example, can get close to a Ryzen 2700 or something similar.
 

MyopicPaideia

macrumors 68020
Mar 19, 2011
2,155
980
Sweden
Those are pretty much the same as the M1 right? At least within 10-25%.
No.

The A14 in the iPad Air is getting similar scores to the older A12Z in the iPad Pro’s and the Apple Silicon Developer Transition Kit Mac mini.

The M1 is most likely a very similar SoC to what the A14X will be, so the M1 will presumably significantly outperform the results of these benchmarks.
 
  • Like
Reactions: onfire23

joshallegro

macrumors member
Oct 19, 2008
42
2
I have a late 2018 MacMini with Intel "Core i7" 3.2GHz Six Core - 8th Gen. I can’t seem to see that processor on this list in order to compare it to the A1?
 

joshallegro

macrumors member
Oct 19, 2008
42
2
Does this (or any other) benchmark show the performance of the "Unified Memory" in comparison to eg. the Mac Mini's DDR RAM per GB? Im trying to figure out what the 16GB Unified Memory option will be the equivalent to in DDR RAM
 

senna95

macrumors newbie
Jun 29, 2017
11
5
Oh, thanks. Never run a benchmark before. Whats the process - download Cinebench and then what?
Yeah you can run it and select on the top left whether you want to run the multi-threaded benchmark or a single-threaded one and then just wait for the result. Be aware, the single-threaded one will likely take a while.
 
  • Like
Reactions: joshallegro

joshallegro

macrumors member
Oct 19, 2008
42
2
Yeah you can run it and select on the top left whether you want to run the multi-threaded benchmark or a single-threaded one and then just wait for the result. Be aware, the single-threaded one will likely take a while.

Should WiFi be turned off when running the benchmark? And do I need to delete background processes from Activity Monitor etc?
 

leman

macrumors Core
Oct 14, 2008
19,520
19,670
Does this (or any other) benchmark show the performance of the "Unified Memory" in comparison to eg. the Mac Mini's DDR RAM per GB? Im trying to figure out what the 16GB Unified Memory option will be the equivalent to in DDR RAM

16GB ;)

But of corse, it's different hardware. The balance between RAM and computational resources is going to be different.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.