The MacBook Pro's slight advantage over the Mac Mini tells me that Apple could probably tweak a lot more performance from the Mini if they were willing to sacrifice a bit on its aesthetics to have a more aggressive thermal management subsystem. Maybe a mini cheese grater design rather than the current cigar box design?
What are you basing that on? Based on the numbers DuhSesame (re)posted, the pro and the mini are identical--even with the 30-minute test. Calling 7763 versus 7716 a "slight advantage" is like saying that something selling for $99.39 is "slightly cheaper" than something selling for $100.00. While technically true, a 0.6% difference just statistical noise.
What these results tell me is that the M1 chip inside all the new Macs are identical (except in terms of memory and that sadly disabled 8th GPU core in the entry-level air), and that (unlike the very first MacBook Air, way back when) the lack of a fan not a huge deal (9% difference at 10 minutes, 18% different at 30 minutes).
The MacBook Pro's slight advantage over the Mac Mini tells me that Apple could probably tweak a lot more performance from the Mini if they were willing to sacrifice a bit on its aesthetics to have a more aggressive thermal management subsystem. Maybe a mini cheese grater design rather than the current cigar box design?
What are you basing that on? Based on the numbers DuhSesame (re)posted, the pro and the mini are identical--even with the 30-minute test. Calling 7763 versus 7716 a "slight advantage" is like saying that something selling for $99.39 is "slightly cheaper" than something selling for $100.00. While technically true, a 0.6% difference just statistical noise.
What these results tell me is that the M1 chip inside all the new Macs are identical (except in terms of memory and that sadly disabled 8th GPUcore in the entry-level air), and that (unlike the very first MacBook Air, way back when) the lack of a fan not a huge deal (9% difference at 10 minutes, 18% different at 30 minutes).
That 8th GPU core is probably not "disabled". One of the 8 x cores is probably defective, and this is Apple's way of using the whole system on a chip, instead of scrapping it. This is common in both CPU & GPU manufacturing.
There was one recent test that demonstrated the new Mac mini is an extremely capable developer machine, besting even a 10-core iMac in Xcode compile times. So I’m seriously thinking about getting an 8GB Mac mini just for developing.
This article could be a lot more concise if you left out all the restated stuff from the articles and news of the last several days.
If Apple sold a first-party Retina display that wasn’t $5000, and if the RAM wasn’t capped at 16GB, I might’ve seriously struggled to NOT buy the new mini. But...
So I’m still waiting to see what comes of the “smaller Mac Pro” rumors, and what they release without these constraints...
In the mean time, since boot camp is nearing the end, I’m also looking at buying a PC so I can finally move beyond 2011 with my gaming... sigh.
The MacBook Pro's slight advantage over the Mac Mini tells me that Apple could probably tweak a lot more performance from the Mini if they were willing to sacrifice a bit on its aesthetics to have a more aggressive thermal management subsystem. Maybe a mini cheese grater design rather than the current cigar box design?
The M1 is single chip - there is no variation. (Although some have a GPU core physically disabled.) The SoC is designed to throttle back when necessary. This has been demonstrated in the Air... it went from 3.2GHz (10W) and throttled down to 2.3GHz (7.5W) after running the CineBench test for 30 minutes. The MacBook Pro has active cooling so it can sustain higher performance for much longer before its cramped interior hits a thermal ceiling and the Mac mini has a lot of empty space inside allowing for an even higher thermal ceiling, it is basically capable of running indefinitely without needing to throttle. Variations in benchmark results should be expected - operating systems have a lot going on in the background and never is there a point when any computer is doing the exact same thing at the exact same performance level. So I wouldn’t worry about the mini’s performance in any these charts.
Furthermore, I believe this model of Mac mini is a stop gap version. Put out simply to supersede the DTK. All the advantages of the this first Mac specific SoC is intended for laptops (LP DRAM, low wattage), so I tend to think that Apple has also developed more performant desktop variants that we will see after all laptop models make the transistion.
Most people are looking at these first Apple Silicon Macs wrong - these aren't Apple's powerhouse machines: they're simply the annual spec bump of the lowest end Apple computers with DCI-P3 displays, Wifi 6, and now the new the Apple Silicon M1 SoC.
They have the same limitations as the machines they replace - 16 GB RAM and two Thunderbolt ports.
These are the machines you give to a teacher or a lawyer or an accountant - folks who need a decently performing machine who don't want to lug around a huge powerhouse machine (or pay for one for that matter). They're still marketed at the same market segment, though they now have a vastly expanded compute power envelope.
The real powerhouses will probably come next year with the M1x (or whatever). Apple has yet to decide on an external memory interconnect and multichannel PCIe scheme, if they decide to move in that direction.
Other CPU and GPU vendors and OEM computer makers take notice - your businesses are now on limited life support. These new Apple Silicon models can compete up through the mid-high tier of computer purchases, and if as I expect Apple sells a ton of these many will be to your prime (most profitable) customers.
In fact, I suspect that Apple - once they recover their R&D costs - will be pushing the prices of these machines lower while still maintaining their margins - while competing computer makers will still have to pay Intel, AMD, Qualcomm, and nVidea for their expensive processors, whereas Apple's cost goes down the more they manufacture. Competing computer makers may soon be demanding lower processor prices from the above manufacturers so they can more readily compete against these models.
I believe the biggest costs for a chip fab are startup costs - no matter what processor vendors would like you to believe. Design and fab startup are expensive - but once you start getting decent yields, the additional costs are silicon wafers and QA. The more of these units Apple can move, the lower the per unit cost and the better the profits - and the more resources they can dedicate to the next generation M1x (not that Apple's hurting for R&D capital 😄).
Of course, Apple will probably never crack the hardcore Windows fanboys, but I believe these represent a very tiny percentage of the Wintel market. Most Wintel purchases are predicated on price - not religion.
The single core score especially is really very impressive. For a mobile CPU too it's madness! It'll be interesting to see if Apple manages to pile on loads of cores for Mac Pro or high-end iMac users, the AMD Threadrippers are stomping on everything multi-core right now, with multicore scores of 30054 in Cinebench.
My 2019 i9 iMac gets 10118 multicore in Cinebench, 1233 single, and OpenGL compute of 50,000.
And this is the first iteration of Apple’s SOC. What lies ahead when the next generation arrives for the iMac and iMac Pro. As the trolls have been beating us over the head reminding us that Apple will never be the number one market leader in sales there is no doubt now that those same trolls have almost nothing left to denigrate Apple’s products. Performance metrics just proved that.
Not to be picky but Apple has been building their own SoC since the iPhone 4. Yes, this is the first Mac-specific version of their SoC design, and a fantastic first version, but it still relies heavily on the A-series chip designs (as stated in the article). I created a spreadsheet showing single, multi and Metal scores. Some Metal scores were from recent results, not in the regular listings. They show the first generation of the M1 SoC is blowing away the latest A-series chips especially in Metal scores. This is the biggest difference the new M1 SoC has.
MAXED OUT...... 16GB Ram....... ROFL. Tim Cook is the best comedian!
I am sure higher end Macs will have more RAM for configuration but do NOT take RAM as the primarily reason. RAM is so overrated nowadays. There are only few uses that would benefit from more RAM such as data-intensive apps (like extremely high resolution images) and virtual machines. If we still use HDDs, then more RAM would make a significant difference but nowadays more and more computers only use SSDs which are extremely fast enough as an effective swap page and still provide robust performance. Also, Mac compresses RAM on the fly effectively doubling the RAM for free.
MAXED OUT...... 16GB Ram....... ROFL. Tim Cook is the best comedian!
I am sure higher end Macs will have more RAM for configuration but do NOT take RAM as the primarily reason. RAM is so overrated nowadays. There are only few uses that would benefit from more RAM such as data-intensive apps (like extremely high resolution images) and virtual machines. If we still use HDDs, then more RAM would make a significant difference but nowadays more and more computers only use SSDs which are extremely fast enough as an effective swap page and still provide robust performance. Also, Mac compresses RAM on the fly effectively doubling the RAM for free.
There are some things you simply need more RAM for - video editing is a big one. I view these as first generation machines targeted at the more general consumer market. As such, 16GB is fine. I fully expect Apple to come out with an M2 (or D2 or ???) version of their processor that can handle more RAM as well as more advanced graphics capabilities. If they don't have plans to do so it would mean they plan on permanently dividing their product line between Apple Si and Intel chips and I seriously doubt that is the case.
MAXED OUT...... 16GB Ram....... ROFL. Tim Cook is the best comedian!
I am sure higher end Macs will have more RAM for configuration but do NOT take RAM as the primarily reason. RAM is so overrated nowadays. There are only few uses that would benefit from more RAM such as data-intensive apps (like extremely high resolution images) and virtual machines. If we still use HDDs, then more RAM would make a significant difference but nowadays more and more computers only use SSDs which are extremely fast enough as an effective swap page and still provide robust performance. Also, Mac compresses RAM on the fly effectively doubling the RAM for free.
There are some things you simply need more RAM for - video editing is a big one. I view these as first generation machines targeted at the more general consumer market. As such, 16GB is fine. I fully expect Apple to come out with an M2 (or D2 or ???) version of their processor that can handle more RAM as well as more advanced graphics capabilities. If they don't have plans to do so it would mean they plan on permanently dividing their product line between Apple Si and Intel chips and I seriously doubt that is the case.
Video effects use a lot of RAM in software like After Effects and Resolve. The following video shows some memory usage in these kind of use cases. 32GB looks ok, 16GB would be paging to disk:
If someone is doing effects on 1080p footage, each uncompressed frame in memory will use 1920 x 1080 x 24-bit = 6MB per frame x 30FPS = 186MB RAM per second x 60 seconds = 11GB RAM just for the current state. Including any undo caches will increase it more. 4K is 4x higher, 10-bit and 60FPS is higher too. The software can write frames to the SSD but RAM is better to use where possible otherwise it has to read frames, make the change and write it back to disk every time you scrub the timeline or adjust a slider.
The M1 chips have shared memory too, Intel MBPs have 16GB + 4-8GB video memory. Apple may be waiting for DDR5 memory before moving the higher end models, the following suggests Q3 2021 for commercial DDR5:
Apple has the advantage with custom chips in not having to wait for motherboard manufacturers and chip manufacturers to support it, they can start using the memory as soon as it's being manufactured. DDR5 should increase density and bandwidth, which helps for the GPU. The 16" MBPs and iMacs should be able to get a 32GB option, possibly 64GB. There's not many use cases above 64GB but for those scenarios, they could have an additional external memory system or non-volatile setup like Optane:
$695 for 128GB would be fine. It's not as fast as normal RAM but still better than an SSD with much better write endurance. Options above 64GB would mainly apply to the iMac and Mac Pro.
What needs to be said in every review, several times when looking at multi core benchmarks, is that Apple’s chips, and this goes for all of them now here, are not really 8 core chips. According to testing, all four of the high efficiency cores together equal one high performance core, approximately. So what we’re really taking about is a 5 core M1 compared to 6 and 8 core chips.
the reason I mention this is because Desktop chips have all cores performing the same. So the comparison is really biased towards those chips. When one realizes that this is really a 5 core equivalent, it’s much more impressive.
The MacBook Pro's slight advantage over the Mac Mini tells me that Apple could probably tweak a lot more performance from the Mini if they were willing to sacrifice a bit on its aesthetics to have a more aggressive thermal management subsystem. Maybe a mini cheese grater design rather than the current cigar box design?
What are you basing that on? Based on the numbers DuhSesame (re)posted, the pro and the mini are identical--even with the 30-minute test. Calling 7763 versus 7716 a "slight advantage" is like saying that something selling for $99.39 is "slightly cheaper" than something selling for $100.00. While technically true, a 0.6% difference just statistical noise.
What these results tell me is that the M1 chip inside all the new Macs are identical (except in terms of memory and that sadly disabled 8th GPU core in the entry-level air), and that (unlike the very first MacBook Air, way back when) the lack of a fan not a huge deal (9% difference at 10 minutes, 18% different at 30 minutes).
That 8th core is much more likely a result of low yield, the same as we saw with Apple’s A12x vs the later A12z. I know some writers, and some people on YouTube have been saying that it’s disabled, as though Apple, for some reason did it deliberately, but it’s highly unlikely.
Yes, I'd bet that the 7-core chip is the result of a bad yield, or failed 8-core chips in a yield. It's being repurposed to a lower end unit, and thus some overhead is lowered, instead of just tossing otherwise functional chips. It also helps establish an entry level Mac and isn't an unknown practice among other manufacturers.
High speed flash storage has removed a lot of dependence on RAM for light and medium lifting. Too many people living in this usage world keep saying that lots of RAM is overrated and unnecessary "with today's SSDs". They're right, but only within they're own little world. Ok, it's a pretty big world but not the whole world, as they might believe.
Even if you don't render a lot of high-end, busy video, there are still RAM sucks. For a light lifter like me, that's browsers and web pages. I frequently have a lot of web pages open with sometimes two browsers. Browsers eat memory and left on their own, suck up a lot of it. I often get that warning something like "this web page is using significant resources...".
I'm hoping somebody's "real world" M1 testing includes loading up the CPU/GPU with a lot of browser pages for a lengthy time. There's been a lot of talk that SSDs have obviated the need for RAM, but that's a short-sighted view.
The M1chip seems to have demonstrated among a number of reviewers it's ability to do much more with less RAM. There's a lot of users who will find 8GB and 16GB enough or more than enough for their use with an M1 computer.
But there are a lot of high-end users who do still and will need more. I believe that will be coming as soon as the next M1 refresh. That goodness may well be reflected in the price, which might be one reason why we didn't see them this time.
All of these M1s are entry level machines in each of their categories that provide more functionality than probably any entry level Mac before. Decrying their suitability for high-end work that they weren't tasked for is silly at best and the stuff haters are made of.
Any objective doubters should scour the web for thorough reviews and if they don't see what ability they cherish tested, should prevail on the tester to consider it. That would be an actual constructive act instead of baseless speculation and petty bitching.
This article could be a lot more concise if you left out all the restated stuff from the articles and news of the last several days.
While true, there are many people for whom AI is there only or major source of information, so writing an article that has a broader reach makes more sense. It fends of inevitable questions from those people.
AI as routinely demonstrated its desire to offer something for everybody, and a review of new Mac iron isn't the place to skimp. Helping a less informed user make an informed choice is a good thing. Let's not forget that many of the people who read AI's and others reviews are not always the same who post to forums. Of course, I'm not counting the complete idiots who post complete nonsense.
And yes, I do find M1 won't surpass the 16-inch in Cinebench slightly disappointing.
Most people are looking at these first Apple Silicon Macs wrong - these aren't Apple's powerhouse machines: they're simply the annual spec bump of the lowest end Apple computers with DCI-P3 displays, Wifi 6, and the new Apple Silicon M1 SoC.
They have the same limitations as the machines they replace - 16 GB RAM and two Thunderbolt ports.
These are the machines you give to a student or teacher or a lawyer or an accountant or a work-at-home information worker - folks who need a decently performing machine with decent build quality who don't want to lug around a huge powerhouse machine (or pay for one for that matter). They're still marketed at the same market segment, though they now have a vastly expanded compute power envelope.
The real powerhouses will probably come next year with the M1x (or whatever). Apple has yet to decide on an external memory interconnect and multichannel PCIe scheme, if they decide to move in that direction.
Other CPU and GPU vendors and OEM computer makers take notice - your businesses are now on limited life support. These new Apple Silicon models can compete speed-wise up through the mid-high tier of computer purchases, and if as I expect Apple sells a ton of these many will be to your bread and butter customers.
In fact, I suspect that Apple - once they recover their R&D costs - will be pushing the prices of these machines lower while still maintaining their margins - while competing computer makers will still have to pay Intel, AMD, Qualcomm, and nVidea for their expensive processors, whereas Apple's cost goes down the more they manufacture. Competing computer makers may soon be squeezed by Apple Silicon price/performance on one side and high component prices on the other. Expect them to be demanding lower processor prices from the above manufacturers so they can more readily compete, and processor manufacturers may have to comply because if OEM computer manufacturers go under or stop making competing models, the processor makers will see a diminishing customer base.
I believe the biggest costs for a chip fab are startup costs - no matter what processor vendors would like you to believe. Design and fab startup are _expensive_ - but once you start getting decent yields, the additional costs are silicon wafers and QA. The more of these units Apple can move, the lower the per unit cost and the better the profits.
So ... who should buy these M1 Macs?
If you're in the target demographic - the student, teacher, lawyer, accountant, or work-at-home information worker - this is the Mac for you.
If you're a heavy computer user like a creative and don't simply want a light and cheap computer with some additional video and sound editing capability for use on the go - I'd wait for the M1x (or whatever) next year. You'll probably kick yourself next year when the machines targeted at _you_ finally appear.
Comments
What these results tell me is that the M1 chip inside all the new Macs are identical (except in terms of memory and that sadly disabled 8th GPU core in the entry-level air), and that (unlike the very first MacBook Air, way back when) the lack of a fan not a huge deal (9% difference at 10 minutes, 18% different at 30 minutes).
If Apple sold a first-party Retina display that wasn’t $5000, and if the RAM wasn’t capped at 16GB, I might’ve seriously struggled to NOT buy the new mini. But...
So I’m still waiting to see what comes of the “smaller Mac Pro” rumors, and what they release without these constraints...
In the mean time, since boot camp is nearing the end, I’m also looking at buying a PC so I can finally move beyond 2011 with my gaming... sigh.
The M1 is single chip - there is no variation. (Although some have a GPU core physically disabled.) The SoC is designed to throttle back when necessary. This has been demonstrated in the Air... it went from 3.2GHz (10W) and throttled down to 2.3GHz (7.5W) after running the CineBench test for 30 minutes. The MacBook Pro has active cooling so it can sustain higher performance for much longer before its cramped interior hits a thermal ceiling and the Mac mini has a lot of empty space inside allowing for an even higher thermal ceiling, it is basically capable of running indefinitely without needing to throttle. Variations in benchmark results should be expected - operating systems have a lot going on in the background and never is there a point when any computer is doing the exact same thing at the exact same performance level. So I wouldn’t worry about the mini’s performance in any these charts.
Furthermore, I believe this model of Mac mini is a stop gap version. Put out simply to supersede the DTK. All the advantages of the this first Mac specific SoC is intended for laptops (LP DRAM, low wattage), so I tend to think that Apple has also developed more performant desktop variants that we will see after all laptop models make the transistion.
Of course, Apple will probably never crack the hardcore Windows fanboys, but I believe these represent a very tiny percentage of the Wintel market. Most Wintel purchases are predicated on price - not religion.
My 2019 i9 iMac gets 10118 multicore in Cinebench, 1233 single, and OpenGL compute of 50,000.
Edit: Couldn't run Unlimited and the Extreme seemed to crash at the end of the test. It also seemed like it was locked at 120 FPS. So not very useful.
If someone is doing effects on 1080p footage, each uncompressed frame in memory will use 1920 x 1080 x 24-bit = 6MB per frame x 30FPS = 186MB RAM per second x 60 seconds = 11GB RAM just for the current state. Including any undo caches will increase it more. 4K is 4x higher, 10-bit and 60FPS is higher too. The software can write frames to the SSD but RAM is better to use where possible otherwise it has to read frames, make the change and write it back to disk every time you scrub the timeline or adjust a slider.
The M1 chips have shared memory too, Intel MBPs have 16GB + 4-8GB video memory. Apple may be waiting for DDR5 memory before moving the higher end models, the following suggests Q3 2021 for commercial DDR5:
https://www.theverge.com/circuitbreaker/2020/10/7/21506883/ddr5-ram-memory-first-modules-sk-hynix
https://www.techradar.com/news/samsungs-superfast-ddr5-ram-will-arrive-in-2021-but-dont-plan-to-upgrade-your-pc-just-yet
Apple has the advantage with custom chips in not having to wait for motherboard manufacturers and chip manufacturers to support it, they can start using the memory as soon as it's being manufactured. DDR5 should increase density and bandwidth, which helps for the GPU. The 16" MBPs and iMacs should be able to get a 32GB option, possibly 64GB. There's not many use cases above 64GB but for those scenarios, they could have an additional external memory system or non-volatile setup like Optane:
https://www.tomshardware.com/news/intel-optane-dimm-pricing-performance,39007.html
$695 for 128GB would be fine. It's not as fast as normal RAM but still better than an SSD with much better write endurance. Options above 64GB would mainly apply to the iMac and Mac Pro.
the reason I mention this is because Desktop chips have all cores performing the same. So the comparison is really biased towards those chips. When one realizes that this is really a 5 core equivalent, it’s much more impressive.
High speed flash storage has removed a lot of dependence on RAM for light and medium lifting. Too many people living in this usage world keep saying that lots of RAM is overrated and unnecessary "with today's SSDs". They're right, but only within they're own little world. Ok, it's a pretty big world but not the whole world, as they might believe.
Even if you don't render a lot of high-end, busy video, there are still RAM sucks. For a light lifter like me, that's browsers and web pages. I frequently have a lot of web pages open with sometimes two browsers. Browsers eat memory and left on their own, suck up a lot of it. I often get that warning something like "this web page is using significant resources...".
I'm hoping somebody's "real world" M1 testing includes loading up the CPU/GPU with a lot of browser pages for a lengthy time. There's been a lot of talk that SSDs have obviated the need for RAM, but that's a short-sighted view.
The M1chip seems to have demonstrated among a number of reviewers it's ability to do much more with less RAM. There's a lot of users who will find 8GB and 16GB enough or more than enough for their use with an M1 computer.
But there are a lot of high-end users who do still and will need more. I believe that will be coming as soon as the next M1 refresh. That goodness may well be reflected in the price, which might be one reason why we didn't see them this time.
All of these M1s are entry level machines in each of their categories that provide more functionality than probably any entry level Mac before. Decrying their suitability for high-end work that they weren't tasked for is silly at best and the stuff haters are made of.
Any objective doubters should scour the web for thorough reviews and if they don't see what ability they cherish tested, should prevail on the tester to consider it. That would be an actual constructive act instead of baseless speculation and petty bitching.
While true, there are many people for whom AI is there only or major source of information, so writing an article that has a broader reach makes more sense. It fends of inevitable questions from those people.
AI as routinely demonstrated its desire to offer something for everybody, and a review of new Mac iron isn't the place to skimp. Helping a less informed user make an informed choice is a good thing. Let's not forget that many of the people who read AI's and others reviews are not always the same who post to forums. Of course, I'm not counting the complete idiots who post complete nonsense.