Apple Silicon M1: benchmarks and review for Mac Mini, MacBook Air and MBP 13 M1 and comparison with other brands

[Henk van Ess]

I found M1-benchmarks, bit.ly/3krh1Ap, by just typing in "Apple Silicon".

At the moment of writing, I get almost 100 Geekbench results. Can they be true?


Regarding comparisons and especially Rosetta performance, you can find another small, usefull write-up in this thread.


MAC MINI
Search Geekbench for "Macmini9,1"

So that’s a 1682 for single-core and 7097 for multi-core for Mac Mini M1. That spec would be better then iMac (27-inch Retina Mid 2020), see https://browser.geekbench.com/mac-benchmarks.

That 1682 tops the highest single-core score till now:

Multicore of 7097 is still better then a MBP 16

MACBOOK PRO
Search Geekbench for "MacBookPro17,1"

The highest single-core I found for the MBP 13 M1 was 1717 and the multi-core score was 7423

That is amazing, again better than Macbook Pro with i9-9980HK 2.4 Ghz and also better than:

MACBOOK AIR
Search Geekbench for "MacBookAir10,1"

AIR M1 versus Macbook Pro 13 M1
MacBookAir10,1 vs MacBookPro17,1 - Geekbench Browser

MacBookAir10,1 vs MacBookPro17,1 - Geekbench

Stats show: no big differences

Mac Mini M1 versus MacBook Pro 13 M1
Macmini9,1 vs MacBookPro17,1 - Geekbench Browser

Macmini9,1 vs MacBookPro17,1 - Geekbench

Stats show: no big differences

But do read what Apple said:

Watch minute 11.53 of this interview with Tim Millet (Apple)

“ I think you see across the different array of machines that we announced yesterday , you gonna see M1 at its best in everyone of those enclosures , it has the ability to fit down into the MacBook Air , you put a fan on it , you put it in a bigger enclosure , like the MacBook PRO, or the MacMini , you gonna see a different level of performance and capability.”

Also : when asked about the differentiation between the MacBook Air and the MacBook Pro, both of which are powered by the M1 processor, Apple’s Craig Federighi was “quick to point out that the latter of the two has an active cooling system“. According to The Independent, Federighi then started sketching a graph to illustrate how heat can play a role in performance.

Apple tells us how it made its new chip and MacBooks

Exclusive: Apple’s hardware, software and marketing chiefs explain how the company’s new processors came to exist

www.independent.co.uk

Compared to other machines :

Apple M1 chip benchmark vs. 6-core 3.7ghz 2019 iMac with AMD 580X

https://twitter.com/i/web/status/1326866126635143169

GB5 scores in a graph compared to NON-Mac's: (thanks @leman)

@leman compared the available GB5 scores for the best mobile Intel Tiger Lake (i7-1165G7, quad-core), the best available mobile Ryzen CPU (Ryzen 7 4800U, octa-core) and the M1 MacBook Air (quad core + four low-power cores) .

Few notes:

- the x axis is the GB5 score, the y axis is the relative count of registered results with this score
- right of the x axis means faster
- this is based on all the scores currently available in GB5 browser
- the sample sizes are very uneven, so M1 results should be taken with a grain of salt (we have hundreds of scores for other CPUs but only 8 for M1)
- please don't forget that this is the passively cooled MacBook Air being compared to actively cooled 15W and 30W CPUs, some of which have double the number of main cores

The current top 5 single core Mac was on Sunday 15th:

What do you all think? How do we have to read this?

 

[Xack]

Yes they’re fake. M1 Macs are still in the factory being produced and if an Apple employee posted a benchmark they would be in trouble.

 

[iMi]

If there was one or two results, maybe I'd question it. But with 40+ results it seems to be valid. We'll not know for sure until we see someone benchmark an actual consumer unit next week. Either way, these numbers are very encouraging.

 

[Henk van Ess]

How do you fake these results?

 

[iMi]

Yes they’re fake. M1 Macs are still in the factory being produced and if an Apple employee posted a benchmark they would be in trouble.

Not necessarily. Apple themselves could have run these benchmarks, test units, developer kits, review units given ahead of the public launch. There are plenty of ways for some folks to get their hands on a Mac before it's released to the public.

 

[Henk van Ess]

In the developer blurb promo, plenty of them said they had one.

 

[Henk van Ess]

But if this would be true... omg.. what does it mean to the highest specced MBP in 1-2 years when the lowest specced MBP with M1 (13 inch) has better specs that the current MBP 16. Can't believe my eyes. Will monitor the link and try to investigate who the people are behind the reports.

 

[leman]

They are definitely real. The performance is as expected.

P.S. ADP is probably "Apple development kit" or something like that. It's not the MacBook Air

 

[Henk van Ess]

I am crying

They are definitely real. The performance is as expected.

By you? On what base do you say this? The Anandtech article?

 

[MooffooM]

You made a serious error though...

The ADP3,2 score is not the Macbook Air, it is the A12 based DTK.

The Macbook Air benchmarks the same as the Macbook Pro, but will probably be subject to throttling for prolonged workloads in the real world.

 

[Henk van Ess]

Yes they’re fake

Please explain why. Not with "logic dictates" but proof

 

[Henk van Ess]

You made a serious error though...

The ADP3,2 score is not the Macbook Air, it is the A12 based DTK.

The Macbook Air benchmarks the same as the Macbook Pro, but will probably be subject to throttling for prolonged workloads in the real world.

I removed MBA specs

 

[leman]

I am crying

By you? On what base do you say this? The Anandtech article?

They are consistent with expected A14 performance when you add more cores and higher thermal envelope. Also, model numbers, specs. etc, everything makes perfect sense. And there is no other hardware with matching performance that could be used to spoof these results.

The results themselves are most likely from Apple employees and other people who got early access.

 

[Henk van Ess]

The ADP3,2 score is not the Macbook Air, it is the A12 based DTK.

Can you have a look at https://t.co/x9nPdl3E6l?amp=1 to see if you can find the MBA with M1 in these 40+ benchmarks? What would be a likely candidate?

 

[leman]

Can you have a look at https://t.co/x9nPdl3E6l?amp=1 to see if you can find the MBA with M1 in these 40+ benchmarks? What would be a likely candidate?

Search Geekbench for "MacBookAir10,1"

 

[MooffooM]

Can you have a look at https://t.co/x9nPdl3E6l?amp=1 to see if you can find the MBA with M1 in these 40+ benchmarks? What would be a likely candidate?

I edited my original post and added the image above... you want to search for Macbookair10,1

 

[Henk van Ess]

Search Geekbench for "MacBookAir10,1"

Just did that. Even more impressed now. It tops the wrong specs I posted about the Macbook Air (now deleted) in my very first message. And it also seems that M1 has no extra powers when "soldered" into a MBP 13, probably it will just hold longer because of the integrated fan

 

[Henk van Ess]

Macbookair10,1

got it

 

[Schmedly]

Yes they’re fake. M1 Macs are still in the factory being produced and if an Apple employee posted a benchmark they would be in trouble.

Well, its worth noting that Appleinsider obtained a benchmark for the “A14x“ prior to M1 branding, and it was roughly the same as the M1 benchmarks recently released. (https://appleinsider.com/articles/2...nchmarked-days-before-apple-silicon-mac-event)

 

[Henk van Ess]

Well I am lost for words. This could be an amazing breakthrough. With the MBP 13, for half the money of a fully specced MBP 16, you get better performance and better battery life. Blurb? Burp!

 

[deeddawg]

Yes they’re fake. M1 Macs are still in the factory being produced and if an Apple employee posted a benchmark they would be in trouble.

Do you honestly think there aren't production systems already in the hands of the various major reviewers/sites so they can do their reviews and have the ready for publication on/before the release day? Something that's been common practice for years?

 

[iMi]

I ordered the Mini with 16GB or RAM and 1TB storage just to see how it goes... If the numbers look good in real life, I'll keep the order. If not, I'll cancel it. Now I have to decide if I want to keep the new iMac that just got here or return it. Either way you slice it, these numbers look very promising.

 

[leman]

Just did that. Even more impressed now. It tops the wrong specs I posted about the Macbook Air (now deleted) in my very first message. And it also seems that M1 has no extra powers when "soldered" into a MBP 13, probably it will just hold longer because of the integrated fan

Pretty much this. Geekbench runs are not long enough to estimate the sustained performance. I'd expect the MBA to dip 20-30% in multicore performance over time (which is still exceptionally good).

The reality here is that we have an $1000 Apple passively cooled laptop that will outperform — in every key aspect — a Dell XPX 13" that costs at least $500 more.

 

[Henk van Ess]

Do you honestly think there aren't systems already in the hands of the various major reviewers/sites?

Yes, didn’t you watch the promo event? Panic-developer, a NGO person, and many more, bragging about the machine. See also

https://twitter.com/i/web/status/1326240889610571778

 

[Henk van Ess]

The reality here is that we have an $1000 Apple passively cooled laptop that will outperform — in every key aspect — a Dell XPX 13" that costs at least $500 more.

most of the first responses to the event were: where are the specs, what a marketing blurb of apple... Seems they even undersold it