22 ms·

> The M1 chip, which belongs to a MacBook Air with 8GB RAM, features a single-core score of 1687 and a multi-core score of 7433. According to the benchmark, t

by mcintyre1994 6y ago

> The M1 chip, which belongs to a MacBook Air with 8GB RAM, features a single-core score of 1687 and a multi-core score of 7433. According to the benchmark, the M1 has a 3.2GHz base frequency. > The Mac mini with M1 chip that was benchmarked earned a single-core score of 1682 and a multi-core score of 7067. > Update: There's also a benchmark for the 13-inch MacBook Pro with M1 chip and 16GB RAM that has a single-core score of 1714 and a multi-core score of 6802. Like the MacBook Air , it has a 3.2GHz base frequency. So single core we have: Air 1687, Mini 1682, Pro 1714 And multi core we have: Air 7433, Mini 7067, Pro 6802 I’m not sure what to make of these scores, but it seems wrong that the Mini and Pro significantly underperform the Air in multi core. I find it hard to imagine this benchmark is going to be representative of actual usage given the way the products are positioned, which makes it hard to know how seriously to take the comparisons to other products too. > When compared to existing devices, the M1 chip in the MacBook Air outperforms all iOS devices. For comparison's sake, the iPhone 12 Pro earned a single-core score of 1584 and a multi-core score of 3898, while the highest ranked iOS device on Geekbench's charts, the A14 iPad Air, earned a single-core score of 1585 and a multi-core score of 4647. This seems a bit odd too - the A14 iPad Air outperforms all iPad Pro devices?

chandler5555 6y ago
the latest ipad pro is still on the A12Z bionic. the air just got refreshed with the A14 bionic
pvg 6y ago
The A14 Air just came out and has a brand new CPU. The Pros have much fancier displays, lower pen latency, etc. Subjectively, in most typical use, the Pros already feel like they have more available cycles than iOS apps have got around to soaking up.
- mzl 6y ago
  This is about the new line of Macs and the M1 chip in them, not iPads
  - mcintyre1994 6y ago
    My comment also mentioned the part of the article that mentioned the A14 iPad Air being the best performing iOS device - I wasn't sure why that was the case.
- mcintyre1994 6y ago
  Thanks, that makes sense - I didn't realise there was no Pro line on the newest chips yet. It'll be interesting to see how the next iPad Pros compare to these M1 Macbooks.
  - BillinghamJ 6y ago
    The iPad Pros never even used the A13 series - they're still back on A12 (though with some variants & binning), so understandable that it could be a fairly big jump
simonh 6y ago
It may be possible the variations are due to differences in the thermal environment when the tests were conducted. I would expect the pro and mini to beat the air as they should have better thermals, but that may only show up over longer term tests and environmental factors could win out in shorter tests. Just a theory.
- joking 6y ago
  if I recall correctly, the geekbench score does run on small bursts and it's designed to find the peak performance without taking the thermal limitations in account.
  - starfallg 6y ago
    Hence why Cinebench is often used these days when evaluating real-world performance with sustained workloads.
    - reaperducer 6y ago
      Different tests. Different purposes. GB deliberately avoids running up the heat because it is focused on testing the chip, not the machine's cooling ability. Cinebench, as you say, tests "real-world" conditions, meaning the entire machine, not just the chip.
      - starfallg 6y ago
        The chip's ability to run at sustained load is a part of its design also. Precisely because modern chips has to throttle in order to meet power and thermal envelopes, we should be looking at sustained performance as a more accurate measure. In a majority of cases, burst performance only affects things like responsiveness, and those things should be measured instead for a better reflection of the benefits.
        adolph 6y ago
        If you perform an integrated test, would you not perform unit tests? An unit test may show areas for easy improvement if other aspects of the total package are changed. For example, if someone thought M1 was thermally constrained, they might decide to rip mini out of the case and attach a different cooling method.
dharma1 6y ago
it's probably <20s short tasks, if you run the CPU/GPU at full load for extended periods the thermals kick in and the M1 Macbook Air without fan will reduce clock speed. iPad pro - the current 2020 gen iPad pro has A12Z (essentially the same chip as 2018 A12X with extra GPU cores) - significantly older chip than A14. I think there will be an A14 iPad Pro refresh with mini led display in early 2021.
- mpweiher 6y ago
  > thermals ... reduce clock speed. I see that statement a lot, and yes, at some point that is going to happen. But the analysis seems to fail to take into account what utterly amazingly low power devices these chips are. So while it will happen, it might take a long time.
andy_threos_io 6y ago
I checked this geekbenchmark with our several different computers on hand, and I can confirm that it's total useless measurements for real world applications or performance.
- noir_lord 6y ago
  Torvalds ripped it apart nearly a decade ago. It's a useless benchmark, what I want to see is things like time to compile a bunch of different software, things that take long enough for the processor/cooling to reach thermal equilibrium etc. I.e. stuff that more closely matches the real world
  - AgloeDreams 6y ago
    He ripped apart a very different benchmark for what it was worth, that was GB3 at the time I believe. 5 was a rewrite to make it cross platform-equal. In real world use it actually is far more relevant than thermally limited benchmarks. It just measures max peak burst performance...which is important because 90% of all users use their computer to do only bursty tasks rather than long term processing. See exporting a 10 second clip on an iPhone or loading a heavy SPA webpage on a Mac. These are 5 second burst tasks where real world use would not be thermally limited but would see real change consistent with Geekbench. It's really only intended to be one of many benchmarks to tell the whole story; of course Linus would attack that because it doesn't make any real sense in his use and isn't the full story for him. If Geekbench was not tested, it would not cover the majority of computing uses and it would weigh cpus that had poor turbo or burst performance unfairly high for most uses. Geekbench is kinda like 0-60MPH times and other tests (like SPEC2006) are like top speed I guess? The whole story is between them.
    - easton 6y ago
      I believe this is the discussion OP is talking about: https://yarchive.net/comp/linux/benchmarks.html https://yarchive.net/comp/linux/benchmarks.html
  - GeekyBear 6y ago
throwaway4good 6y ago
The results seem a little weird but if remotely true then these machines are going to sell like cup cakes. Why would anyone (who is not forced) buy an Intel PC laptop when these are available and priced as competitive as they are?
- dbspin 6y ago
  Professional use in creative industry - film, TV, videography, audio production, and a million other things. 16GB ram and a souped up integrated GPU won't cut it for many 'Pro' applications Mac has traditionally excelled at.
- seunosewa 6y ago
  Many apps optimized for the x64 platform won't run as well as the benchmarks.
  - Cthulhu_ 6y ago
    I think this is an important one to keep in mind. I'm sure most native Mac apps will be compiled to ARM, but a lot of existing apps won't. Plus there's the brouhaha about Electron apps. I for one really wouldn't mind if Apple would build a native app to replace Electron apps, e.g. a chat app that works as a good client for (what I have open right now in Rambox) Discord, FB Messenger, Whatsapp and multiple Slack channels. Or their own Spotify client. Or a big update to XCode so it can use language servers for VS Code and it's viable to use for (what I do right now) Typescript, PHP and Go development. They have more than enough money to invest in dozens of development teams or startups to push out new native apps. One day I'll switch away from Chrome in favor of Safari as well. Maybe. (I am taking recommendations for native alternatives to apps)
    - asdfaoeu 6y ago
      Electron seems to have support in recent beta releases.
      - 6y ago
gregoriol 6y ago
One has to remember that Apple is still selling an Intel mac mini at the top of the range: it likely means something about the performance to expect from M1 vs Intel.
- miskin 6y ago
  They probably have a lot of customers still demanding Intel CPUs. Mac minis are often used in server farms as build servers and there are many companies that would require Intel CPU there for some time.
  - arrrg 6y ago
    All of that is true, however it is notable that the models they replaced (are not selling anymore) are all the lower end models. The two port MacBook Pro, the Air, only the lower end mini. Seems pretty obvious to me that there will be another more higher end variant of the M1, though maybe the only difference will be the amount of RAM, the number of GPU cores, the number of supported USB4 ports or something like that, not raw CPU performance. Either way, it seems obvious to me that the M1 is their low end Mac chip. That will be interesting to watch.
    - _ph_ 6y ago
      Yes, it looks like the M1 was designed mostly for the MacBook Air. The specs are a perfect fit and it makes a lot of sense, as the Air is their most popular laptop. Having the perfect Air - and the new one is truly impressive - will make for a lot of Mac sales. Also, they also put it in the bottom end MB Pro and Mini. But indeed with the next variant of Apple Silicon, the higher end variations, and other devices are probably going to be supported.
  - valuearb 6y ago
    The M1 Mini is far better than Intel Mini on cost/power, power/watt, power/heat measures. Server farms are going to switch rapidly, one leading Mini server farm just announced a 600 unit starter order, and the CEO noted that Big Sur also made significant changes to licensing to make its use in server farms easier.
gigatexal 6y ago
> This seems a bit odd too - the A14 iPad Air outperforms all iPad Pro devices? Well yeah, every year for the last bunch or years the A series of chips have had sizeable IPC improvements such that the A12 based iPad Pros are slower than the new Air. Apple's chip division is just industry leading here.
agloeregrets 6y ago
A14 Air outperforms all iPad Pros in single core, multicore is still faster on the A12X. Keep in mind the fastest iPad Pro is using a two generation old cpu.
- blunte 6y ago
  I have watched a number of reviews comparing the new Air to the Pro. The CPU performance increase isn't noticeable in most cases, and the Pro still offers better pencil latency and display refresh rate (plus camera/lidar, but probably most don't care). I wouldn't buy a Pro now because I would wait for the next version, but I wouldn't trade a current Pro for a new Air just for the CPU bump...
Reason077 6y ago
> “the A14 iPad Air outperforms all iPad Pro devices?” iPad Pro is still on an older generation of SoC (A12Z), while the iPad Air just got the new A14.
m12k 6y ago
AFAIK it's pretty common for new macs to spend a while creating an index of its hard drive. For that reason, if you want to run benchmarks, you should generally wait until it's done with that (e.g. an hour or probably less with these speedybois). It might be that the people running the Pro benchmarks didn't wait for that, in their rush to publish the first benchmark. This would be consistent with what we're seeing - the Pro has faster single core performance, but slightly lower multicore, because some of its "background" cores were busy creating the index, while the Air was done with that task.
- Joeri 6y ago
  Quite likely what happened, a second geekbench score has shown up for the pro and it matches the air: https://browser.geekbench.com/v5/cpu/search?utf8=&q=MacBookPro17%2C1 https://browser.geekbench.com/v5/cpu/search?utf8=&q=MacBookP... My guess: in geekbench air and pro score the same, because geekbench is shortlived and not thermally constrained. In cinebench you'll see the pro pulling ahead.
  - kzrdude 6y ago
    Who uploads these results, are they unverified?
    - simonh 6y ago
      The geekbench software uploads them automatically.
    - gowld 6y ago
      Anyone on the Internet can upload results. They are not verified.
Shivetya 6y ago
https://twitter.com/tldtoday/status/1326610187529023488 https://twitter.com/tldtoday/status/1326610187529023488 Has an interesting comparison of an iPhone 12 mini doing similar work to an i9 iMac Now I haven't dug into the details to verify both produced the same results. I believe most of the difference is from software encoding versus hardware encoding. the follow up tweets suggest similar output. it does show how workloads can cause people to jump to conclusions on simply one test and not having all the details to support the conclusion they desire to arrive at
Cabal 6y ago
I think more results are needed to smooth out the curves. If you check all the results by model so far, https://browser.geekbench.com/v5/cpu/search?q=Macmini9%2C1 https://browser.geekbench.com/v5/cpu/search?q=Macmini9%2C1 https://browser.geekbench.com/v5/cpu/search?q=MacBookPro17%2C1 https://browser.geekbench.com/v5/cpu/search?q=MacBookPro17%2... https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q=MacBookAir10%2C1 https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q... it looks like they're all in the same ballpark (i.e. the Air is not leading others, just comparable).
Tuna-Fish 6y ago
Geekbench is a series of extremely short and bursty benchmarks. Because of this, it doesn't really test the steady state speed a system is capable of, it's more testing the peak performance for short periods. In this view, it's entirely possible that the Air simply did not have time to throttle before the benchmark ran out.
- millzlane 6y ago
  Yep, try doing that test back to back 25 times and see who comes out on top.

chandler5555 6y ago

the latest ipad pro is still on the A12Z bionic. the air just got refreshed with the A14 bionic

pvg 6y ago

The A14 Air just came out and has a brand new CPU. The Pros have much fancier displays, lower pen latency, etc. Subjectively, in most typical use, the Pros already feel like they have more available cycles than iOS apps have got around to soaking up.

mzl 6y ago
This is about the new line of Macs and the M1 chip in them, not iPads
- mcintyre1994 6y ago
  My comment also mentioned the part of the article that mentioned the A14 iPad Air being the best performing iOS device - I wasn't sure why that was the case.
mcintyre1994 6y ago
Thanks, that makes sense - I didn't realise there was no Pro line on the newest chips yet. It'll be interesting to see how the next iPad Pros compare to these M1 Macbooks.
- BillinghamJ 6y ago
  The iPad Pros never even used the A13 series - they're still back on A12 (though with some variants & binning), so understandable that it could be a fairly big jump

simonh 6y ago

It may be possible the variations are due to differences in the thermal environment when the tests were conducted. I would expect the pro and mini to beat the air as they should have better thermals, but that may only show up over longer term tests and environmental factors could win out in shorter tests. Just a theory.

joking 6y ago
if I recall correctly, the geekbench score does run on small bursts and it's designed to find the peak performance without taking the thermal limitations in account.
- starfallg 6y ago
  Hence why Cinebench is often used these days when evaluating real-world performance with sustained workloads.
  - reaperducer 6y ago
    Different tests. Different purposes. GB deliberately avoids running up the heat because it is focused on testing the chip, not the machine's cooling ability. Cinebench, as you say, tests "real-world" conditions, meaning the entire machine, not just the chip.
    - starfallg 6y ago
      The chip's ability to run at sustained load is a part of its design also. Precisely because modern chips has to throttle in order to meet power and thermal envelopes, we should be looking at sustained performance as a more accurate measure. In a majority of cases, burst performance only affects things like responsiveness, and those things should be measured instead for a better reflection of the benefits.
      - adolph 6y ago
        If you perform an integrated test, would you not perform unit tests? An unit test may show areas for easy improvement if other aspects of the total package are changed. For example, if someone thought M1 was thermally constrained, they might decide to rip mini out of the case and attach a different cooling method.

dharma1 6y ago

it's probably <20s short tasks, if you run the CPU/GPU at full load for extended periods the thermals kick in and the M1 Macbook Air without fan will reduce clock speed. iPad pro - the current 2020 gen iPad pro has A12Z (essentially the same chip as 2018 A12X with extra GPU cores) - significantly older chip than A14. I think there will be an A14 iPad Pro refresh with mini led display in early 2021.

mpweiher 6y ago
> thermals ... reduce clock speed. I see that statement a lot, and yes, at some point that is going to happen. But the analysis seems to fail to take into account what utterly amazingly low power devices these chips are. So while it will happen, it might take a long time.

andy_threos_io 6y ago

I checked this geekbenchmark with our several different computers on hand, and I can confirm that it's total useless measurements for real world applications or performance.

noir_lord 6y ago
Torvalds ripped it apart nearly a decade ago. It's a useless benchmark, what I want to see is things like time to compile a bunch of different software, things that take long enough for the processor/cooling to reach thermal equilibrium etc. I.e. stuff that more closely matches the real world
- AgloeDreams 6y ago
  He ripped apart a very different benchmark for what it was worth, that was GB3 at the time I believe. 5 was a rewrite to make it cross platform-equal. In real world use it actually is far more relevant than thermally limited benchmarks. It just measures max peak burst performance...which is important because 90% of all users use their computer to do only bursty tasks rather than long term processing. See exporting a 10 second clip on an iPhone or loading a heavy SPA webpage on a Mac. These are 5 second burst tasks where real world use would not be thermally limited but would see real change consistent with Geekbench. It's really only intended to be one of many benchmarks to tell the whole story; of course Linus would attack that because it doesn't make any real sense in his use and isn't the full story for him. If Geekbench was not tested, it would not cover the majority of computing uses and it would weigh cpus that had poor turbo or burst performance unfairly high for most uses. Geekbench is kinda like 0-60MPH times and other tests (like SPEC2006) are like top speed I guess? The whole story is between them.
  - easton 6y ago
    I believe this is the discussion OP is talking about: https://yarchive.net/comp/linux/benchmarks.html https://yarchive.net/comp/linux/benchmarks.html
- GeekyBear 6y ago

throwaway4good 6y ago

The results seem a little weird but if remotely true then these machines are going to sell like cup cakes. Why would anyone (who is not forced) buy an Intel PC laptop when these are available and priced as competitive as they are?

dbspin 6y ago
Professional use in creative industry - film, TV, videography, audio production, and a million other things. 16GB ram and a souped up integrated GPU won't cut it for many 'Pro' applications Mac has traditionally excelled at.
seunosewa 6y ago
Many apps optimized for the x64 platform won't run as well as the benchmarks.
- Cthulhu_ 6y ago
  I think this is an important one to keep in mind. I'm sure most native Mac apps will be compiled to ARM, but a lot of existing apps won't. Plus there's the brouhaha about Electron apps. I for one really wouldn't mind if Apple would build a native app to replace Electron apps, e.g. a chat app that works as a good client for (what I have open right now in Rambox) Discord, FB Messenger, Whatsapp and multiple Slack channels. Or their own Spotify client. Or a big update to XCode so it can use language servers for VS Code and it's viable to use for (what I do right now) Typescript, PHP and Go development. They have more than enough money to invest in dozens of development teams or startups to push out new native apps. One day I'll switch away from Chrome in favor of Safari as well. Maybe. (I am taking recommendations for native alternatives to apps)
  - asdfaoeu 6y ago
    Electron seems to have support in recent beta releases.
    - 6y ago

gregoriol 6y ago

One has to remember that Apple is still selling an Intel mac mini at the top of the range: it likely means something about the performance to expect from M1 vs Intel.

miskin 6y ago
They probably have a lot of customers still demanding Intel CPUs. Mac minis are often used in server farms as build servers and there are many companies that would require Intel CPU there for some time.
- arrrg 6y ago
  All of that is true, however it is notable that the models they replaced (are not selling anymore) are all the lower end models. The two port MacBook Pro, the Air, only the lower end mini. Seems pretty obvious to me that there will be another more higher end variant of the M1, though maybe the only difference will be the amount of RAM, the number of GPU cores, the number of supported USB4 ports or something like that, not raw CPU performance. Either way, it seems obvious to me that the M1 is their low end Mac chip. That will be interesting to watch.
  - _ph_ 6y ago
    Yes, it looks like the M1 was designed mostly for the MacBook Air. The specs are a perfect fit and it makes a lot of sense, as the Air is their most popular laptop. Having the perfect Air - and the new one is truly impressive - will make for a lot of Mac sales. Also, they also put it in the bottom end MB Pro and Mini. But indeed with the next variant of Apple Silicon, the higher end variations, and other devices are probably going to be supported.
- valuearb 6y ago
  The M1 Mini is far better than Intel Mini on cost/power, power/watt, power/heat measures. Server farms are going to switch rapidly, one leading Mini server farm just announced a 600 unit starter order, and the CEO noted that Big Sur also made significant changes to licensing to make its use in server farms easier.

gigatexal 6y ago

> This seems a bit odd too - the A14 iPad Air outperforms all iPad Pro devices? Well yeah, every year for the last bunch or years the A series of chips have had sizeable IPC improvements such that the A12 based iPad Pros are slower than the new Air. Apple's chip division is just industry leading here.

agloeregrets 6y ago

A14 Air outperforms all iPad Pros in single core, multicore is still faster on the A12X. Keep in mind the fastest iPad Pro is using a two generation old cpu.

blunte 6y ago
I have watched a number of reviews comparing the new Air to the Pro. The CPU performance increase isn't noticeable in most cases, and the Pro still offers better pencil latency and display refresh rate (plus camera/lidar, but probably most don't care). I wouldn't buy a Pro now because I would wait for the next version, but I wouldn't trade a current Pro for a new Air just for the CPU bump...

Reason077 6y ago

> “the A14 iPad Air outperforms all iPad Pro devices?” iPad Pro is still on an older generation of SoC (A12Z), while the iPad Air just got the new A14.

m12k 6y ago

AFAIK it's pretty common for new macs to spend a while creating an index of its hard drive. For that reason, if you want to run benchmarks, you should generally wait until it's done with that (e.g. an hour or probably less with these speedybois). It might be that the people running the Pro benchmarks didn't wait for that, in their rush to publish the first benchmark. This would be consistent with what we're seeing - the Pro has faster single core performance, but slightly lower multicore, because some of its "background" cores were busy creating the index, while the Air was done with that task.

Joeri 6y ago
Quite likely what happened, a second geekbench score has shown up for the pro and it matches the air: https://browser.geekbench.com/v5/cpu/search?utf8=&q=MacBookPro17%2C1 https://browser.geekbench.com/v5/cpu/search?utf8=&q=MacBookP... My guess: in geekbench air and pro score the same, because geekbench is shortlived and not thermally constrained. In cinebench you'll see the pro pulling ahead.
- kzrdude 6y ago
  Who uploads these results, are they unverified?
  - simonh 6y ago
    The geekbench software uploads them automatically.
  - gowld 6y ago
    Anyone on the Internet can upload results. They are not verified.

Shivetya 6y ago

https://twitter.com/tldtoday/status/1326610187529023488 https://twitter.com/tldtoday/status/1326610187529023488 Has an interesting comparison of an iPhone 12 mini doing similar work to an i9 iMac Now I haven't dug into the details to verify both produced the same results. I believe most of the difference is from software encoding versus hardware encoding. the follow up tweets suggest similar output. it does show how workloads can cause people to jump to conclusions on simply one test and not having all the details to support the conclusion they desire to arrive at

Cabal 6y ago

I think more results are needed to smooth out the curves. If you check all the results by model so far, https://browser.geekbench.com/v5/cpu/search?q=Macmini9%2C1 https://browser.geekbench.com/v5/cpu/search?q=Macmini9%2C1 https://browser.geekbench.com/v5/cpu/search?q=MacBookPro17%2C1 https://browser.geekbench.com/v5/cpu/search?q=MacBookPro17%2... https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q=MacBookAir10%2C1 https://browser.geekbench.com/v5/cpu/search?utf8=%E2%9C%93&q... it looks like they're all in the same ballpark (i.e. the Air is not leading others, just comparable).

Tuna-Fish 6y ago

Geekbench is a series of extremely short and bursty benchmarks. Because of this, it doesn't really test the steady state speed a system is capable of, it's more testing the peak performance for short periods. In this view, it's entirely possible that the Air simply did not have time to throttle before the benchmark ran out.

millzlane 6y ago
Yep, try doing that test back to back 25 times and see who comes out on top.