Primate Labs Launches Geekbench 6 Benchmarking Suite

[Love-hate 🍏 relationship]

It shouldn't be complicated because these benchmarks are not meant to be compared to different versions of the program.

True but what I mean is that mentally it feels more complicated

And also you can’t always find new benchmarks for older devices

 

[Realityck]

So if an Intel i7-12700 is baseline (2500), then most of the M-series scores I'm seeing here are around the speed of an Intel i7? I thought M1/M2 killed Intel, performance-wise?

Thats a 12th Generation i7 offering 12 cores (8 performance/4 efficiency) Max turbo 4.90Ghz processor

Intel® Core™ i7-12700 Processor (25M Cache, up to 4.90 GHz) - Product Specifications | Intel

Intel® Core™ i7-12700 Processor (25M Cache, up to 4.90 GHz) quick reference with specifications, features, and technologies.

www.intel.com

Comparably a M1 base frequency is 3.2GHZ as a comparison, so yes the M1 speeds are close without having to run at 4.90Ghz.

 

[obviouslogic]

So if an Intel i7-12700 is baseline (2500), then most of the M-series scores I'm seeing here are around the speed of an Intel i7? I thought M1/M2 killed Intel, performance-wise?

No one ever said that. The claims have always been about efficiency. That Intel score is achievable by boosting the clock up to 5Ghz.

 

[Nick_P]

Interesting. Can someone please explain how the new benchmarks can still serve as a valid comparison with older hardware that has only run the previous version(s), given that the tests have changed with the new version? Or can they? Doesn’t changing the test defeat the very purpose of a benchmark? Do we need to run the new benchmarks on old hardware before it can be compared with new hardware?

Not comparable at all. Every 3 or so years Geekbench releases a new version and all the benchmarks start over.

 

[DeepIn2U]

M2 Pro Mac mini (16/512 10C/16C)
CPU Single: 2657
CPU Multi: 12,182.
GPU Metal:72,779
OpenCL: 44,136.

M2 Pro Mini 10C/16GB
CPU Single: 2651
CPU Multi: 12085
GPU Metal: 71993

I break for Mac Mini M2 Pro - dats what I haz come here for!
@opuscroakus could you please update your original post above with the OpenCL score result from your GB6 test please? Thanks.

Mac Studio M1 Ultra (macOS 13.2.1)
Single-Core: 2393
Multi-Core: 17227
Metal: 167494
OpenCL: 102795

Those MC and Metal scores is Sweet Christmas!

 

[Nick_P]

Edit: Updated results with nothing running in the background, and added 2nd machine:

M2 Air, 24GB Ram, 512 ssd, 10 core GPU:
CPU: 2621/9837
GPU: 45039 Metal, 27775 OpenCL

M1 Ultra Studio, 64GB Ram, 1TB ssd, 48 core GPU
CPU: 2407/17485
GPU: 147045 Metal, 89670 OpenCL

 

[Homme]

With the results of even A15X on the new iPad Pro can you guys even imagine the single and multi core performance on the A17, the A17X and it’s line of families from that 3nm ( apparently 3nm+) wafer. Mind blowing indeed

 

[1BadManVan]

With the results of even A15X on the new iPad Pro can you guys even imagine the single and multi core performance on the A17, the A17X and it’s line of families from that 3nm ( apparently 3nm+) wafer. Mind blowing indeed

I like how you’ve just decided you want to make up their own names, you that offended by them going by the M series or something? Seems pretty ridiculous, for attention I guess? Worked

 

[1BadManVan]

M2 Air, 24GB Ram, 512 ssd, 10 core GPU:
CPU: 2397/9590
GPU: 45166 Metal, 27778 OpenCL

A little low for your m2 air for the cpu. Mine came back at 2603/9662

 

[Jack Neill]

M1/16/1TB Air

2346

8478

 

[Homme]

I like how you’ve just decided you want to make up their own names, you that offended by them going by the M series or something? Seems pretty ridiculous, for attention I guess? Worked

No I’m not offended by them being called the M series, I’m just fantasizing 😊

 

[Nick_P]

A little low for your m2 air for the cpu. Mine came back at 2603/9662

I’ll try it again. I was web browsing at the time

Updated:
M2 Air, 24GB Ram, 512 ssd, 10 core GPU:
CPU: 2621/9837
GPU: 45039 Metal, 27775 OpenCL

 

[Homy]

The Stockfish people

Sounds like straight out of the X-files. 😄

 

[KPOM]

I’ll try it again. I was web browsing at the time

Yes, the first time I ran Geekbench on my M2 Pro mini it also came back below 2500. I usually run benchmarks 2 or 3 times just in case some random process decided to run in the background during one of the runs.

 

[1BadManVan]

I’ll try it again. I was web browsing at the time

Updated:
M2 Air, 24GB Ram, 512 ssd, 10 core GPU:
CPU: 2621/9837
GPU: 45039 Metal, 27775 OpenCL

Much better! I ran it again as well

M2 Air 16Gb Ram, 1tb ssd with the 10 core gpu

Cpu: 2621/9843
GPU: 45399 Metal and 27726 OpenCL

Definitely can't complain about the results on these

 

[Homme]

I love how the Air 2 and Mini 4 are just the few examples getting Geekbench 6. Will be interesting to see a slight improvement on the iPad Mini 4

 

[_Spinn_]

Glad to see this benchmark suite keeping up with hardware changes.

 

[WilliApple]

This feel likes Geekbench Vista, the iOS version sucks compared to 5. Im going to use 5 for my benchmarking needs.

 

[Nick_P]

Does anyone know why on GB5 comparison charts, the M2 Pro and M2 Max from the MacBook 14/16" Pro line is not there? At this point, there must be hundreds or even thousands of these results, but they are not in the chart. The description says only "qty 5" machines are needed to be included. I notice this every time there are new machines, that the chart does not update for weeks or longer. Thanks.

 

[svish]

Glad to hear that it continues to be free.

 

[Bustycat]

No I’m not offended by them being called the M series, I’m just fantasizing 😊

You just invented something that nobody knows what you were talking about.

 

[Bustycat]

Everyone is feeling their devices receive an upgrade, but that’s not the case for old machines. The CPU scores of my MacBook Pro (13-inch Mid 2012) dropped from 624 and 1320… to 236 and 778.

 

[DeepIn2U]

iPhone 14 Pro 256GB full charge, untethered.

Single-Core: 2512
Multi-Core: 6420
Metal Score: 22444

To coin Tubbs “You beast you!” Damn Apple wasn’t lying bout the A series performance. Wow.

 

[ikir]

IMO the value of this benchmark for creative users has become even more dubious than it already was. I have a Mac Pro 2013 12 core that, in most tasks (especially 4K video editing & colour grading), is comparable with my Mac mini M1, and in some tasks outperforms it by a decent margin. The exception is for something like hardware accelerated H.265 encoding where the Mini does exceptionally well. But this benchmark skews the results drastically in favour of the M1. In fact, even my iPhone 12 mini gets multi-core results that are almost as good as the 12 core Mac Pro, which strikes me as absurd. The Metal scores have also skewed in favour of the new hardware. I think Geekbench 4 reflected reality for purposes like mine much better. Maybe not for machine learning or H.265 encoding, but for run of the mill creative work.

My scores for reference (single-core/multi-core/Metal):

iPhone 12 mini: 1999/4203/16149
MBP 15" 2017 3.1GHz w Radeon Pro 560: 1200/3996/20002
Mac Pro 12 core 2.7 w FirePro D700: 638/4682/29152
Mac mini M1: 2320/8296/32313

No Apple Silicon destroy old intel, this new version just works better on apple GPU, we all know Geekbench 5 was not able to use full potential

 

[litvcom]

iPhone 13 Pro (iOS 15.7, iOS 16.1.2, iOS 16.3)