Binned M2 Pro vs. Unbinned M1 Pro

[aveona]

Hey folks,

Is the binned M2, with 6 performance cores and 4 efficiency cores, worse in performance than the unbinned M1 Pro with 8 performance cores and 2 efficiency cores?

My use case is studying, multitasking with alot of tabs open, lightroom and a bit of gaming.

The unbinned M1 pro with 1TB had the same pricing as the binned M2 pro with only 512 GB.

Storage isn't an issue for me as I have multiple external storage drives.

But did I make a mistake? And should I return the M2 for the M1 in terms of performance?

Thanks in advance!

 

[rwhg]

Not sure we will know about CPU performance until Geekbench scores appear.

However my understanding (very, very uneducated layman’s perspective) is that the cores themselves are more powerful. So the binned M2 Pro with 10 cores may well be more powerful than the unbinned M1 Pro with 10 cores, despite the ratio of efficiency to performance cores being different.

If you’ve ordered a binned M2 Pro system, you have 14 days to return it for a full refund if it doesn’t meet your needs.

 

[ChristianMorris]

Hey folks,

Is the binned M2, with 6 performance cores and 4 efficiency cores, worse in performance than the unbinned M1 Pro with 8 performance cores and 2 efficiency cores?

My use case is studying, multitasking with alot of tabs open, lightroom and a bit of gaming.

The unbinned M1 pro with 1TB had the same pricing as the binned M2 pro with only 512 GB.

Storage isn't an issue for me as I have multiple external storage drives.

But did I make a mistake? And should I return the M2 for the M1 in terms of performance?

Thanks in advance!

Hi,

The M2 and M1 Pro are both powerful processors and the difference in performance will likely be small. The M1 Pro has a higher number of performance cores, which may give it an edge in certain tasks that are heavily reliant on single-threaded performance. However, the M2 has a higher number of efficiency cores, which may give it an advantage in tasks that can take advantage of multiple cores.

In your use case of studying, multitasking with a lot of tabs open, Lightroom, and some gaming, the M2 is likely to perform well. The M2 is likely to be more efficient when running multiple background tasks and will handle multitasking better.

As for storage, you mentioned that it is not an issue for you since you have external storage, so if you are satisfied with the performance of the M2, then you should keep it. The decision to return the M2 for the M1 Pro should depend on your specific performance needs and preferences.

 

[xraydoc]

Hi,

The M2 and M1 Pro are both powerful processors and the difference in performance will likely be small. The M1 Pro has a higher number of performance cores, which may give it an edge in certain tasks that are heavily reliant on single-threaded performance. However, the M2 has a higher number of efficiency cores, which may give it an advantage in tasks that can take advantage of multiple cores.

In your use case of studying, multitasking with a lot of tabs open, Lightroom, and some gaming, the M2 is likely to perform well. The M2 is likely to be more efficient when running multiple background tasks and will handle multitasking better.

As for storage, you mentioned that it is not an issue for you since you have external storage, so if you are satisfied with the performance of the M2, then you should keep it. The decision to return the M2 for the M1 Pro should depend on your specific performance needs and preferences.

I agree with this analysis. The base M2 Pro and the upper-end M1 Pro will likely be pretty evenly matched within a few % points up or down on both the CPU and GPU side.

The base M2 Pro will likely have better battery life, however.

 

[tstafford]

I agree with this analysis. The base M2 Pro and the upper-end M1 Pro will likely be pretty evenly matched within a few % points up or down on both the CPU and GPU side.

The base M2 Pro will likely have better battery life, however.

And the M2P will have better single core performance for things like web browsing. My M2 MBA is faster at these tasks than my MBP14 M1P.

 

[John90976]

Hey, I was wondering the same thing. Thanks to reviewers already benchmarking the M2 Pro MacBooks, and that info being public on the geekbench site, there is an easy way to think about it:

10 core m1 pro = the 10 core m2 pro in multicore performance, roughly 12K in geekbench.
M2 pro will have better single-core performance (1950>1750).

So in terms of value, the base, binned 10-core M2 Pro is equal to the full M1 Pro which would've cost an extra $300, it is delivering better value.

The performance gains in the 6 p-cores + 2 additional e-cores allowed them to make the chip at the same performance level as M1 Pro with fewer power-hungry p-cores. I guess that is good, having more e-cores on the laptop certainly makes sense, but I am not a fan of losing p-cores. Luckily, you can pay $300 more for 2 more p cores (31% more multicore performance, geekbench score of 15K)....and 3 more gpu cores? lol

 

[TSE]

Do not go for the binned M2 Pro. Do not go for the M1 Pro. Go for the M2 Max. And if you can't afford it, save.

 

[John90976]

Do not go for the binned M2 Pro. Do not go for the M1 Pro. Go for the M2 Max. And if you can't afford it, save.

why? why recommend the model with double the GPU when they are asking about the CPU? You can get the full 12-core in M2 pro as well. No reason to get Max unless you know you need the GPU, additional display outs, and memory bandwidth/capacity(96gb).

 

[TSE]

why? why recommend the model with double the GPU when they are asking about the CPU? You can get the full 12-core in M2 pro as well. No reason to get Max unless you know you need the GPU, additional display outs, and memory bandwidth/capacity(96gb).

He's asking for advice and that's my advice.

 

[Queen6]

He's asking for advice and that's my advice.

Which is plainly bad. OP has not really expanded on the usage. Barring the gaming factor a base M1 Air would easily suffice. M2 Max is vastly overpriced and very much overkill unless you have a specific need.

As for the OP M1/M2, he/she wont be disappointed either ways. Me I'd go with the cheapest Mac for the study and game on the PC, be way cheaper and a far better experience in the long run...

Study implies battery longevity M1/M2 Air or pushing hard M1 13" MBP with it's active cooling. I really like the 14" MBP form factor, but it doesn't get remotely close to the 13"s battery run time and that cancels it out...

Q-6

 

[Sagnet]

Hey, I was wondering the same thing. Thanks to reviewers already benchmarking the M2 Pro MacBooks, and that info being public on the geekbench site, there is an easy way to think about it:

10 core m1 pro = the 10 core m2 pro in multicore performance, roughly 12K in geekbench.
M2 pro will have better single-core performance (1950>1750).

So in terms of value, the base, binned 10-core M2 Pro is equal to the full M1 Pro which would've cost an extra $300, it is delivering better value.

The performance gains in the 6 p-cores + 2 additional e-cores allowed them to make the chip at the same performance level as M1 Pro with fewer power-hungry p-cores. I guess that is good, having more e-cores on the laptop certainly makes sense, but I am not a fan of losing p-cores. Luckily, you can pay $300 more for 2 more p cores (31% more multicore performance, geekbench score of 15K)....and 3 more gpu cores? lol

The binned M2 might have equal or better performance than the unbinned M1. But the M2 cores do run at a higher clock frequency, and I wonder if that means that the M2 will generate more heat, and make the fans spin more, than on the M1?

 

[John90976]

The binned M2 might have equal or better performance than the unbinned M1. But the M2 cores do run at a higher clock frequency, and I wonder if that means that the M2 will generate more heat, and make the fans spin more, than on the M1?

Yes, that is a concern, on geekbench I see the M2 max hovering between 13-15K points in multicore. And between 1772 and 2030 in single-core. I hope those are flukes because those are massive 10-20% variances that make it more like a 10-core M1 Pro! (1725 single core/12, 230 multi-core) I have to assume it is due to thermal throttling.

 

[Juicy Box]

The binned M2 might have equal or better performance than the unbinned M1. But the M2 cores do run at a higher clock frequency, and I wonder if that means that the M2 will generate more heat, and make the fans spin more, than on the M1?

If everything else is the same, I think you are correct in your concerns about more heat/fan speeds, but the cooling systems could be improved on the new Macs.

The 14” M1 Max MBP thermal throttled due to the less effective cooling system compared to the 16”. If Apple improves the cooling on the new Macs, especially on the 14”, then these Macs could run with the same or less fan speeds/noise than the old ones.

I wonder what the M2 Pro Mac Mini cooling will look like. There is plenty of room for a beefier system, hopefully Apple took advantage of the abundant space in the MM case.

 

[DHagan4755]

Max Tech's analysis of leaked benchmarks.

 

[donawalt]

Sorry for the uneducated question, what is binned vs. unbinned? I may have missed it, I don't see a definition of that anywhere...

 

[ilikewhey]

Sorry for the uneducated question, what is binned vs. unbinned? I may have missed it, I don't see a definition of that anywhere...

binned is a chip that failed the performance needed for its intended purpose, however to throw away that chip is a waste of money, instead mfr just disable the cores on these "failed" chip and put it in a lower category. hence the term binning.

there are always some silicons that is not the same performance wise on the same wafer, so there will always be some binned chips.

 

[MyiBill]

Are all the pro chips binned or just the ones on the lowest model?

 

[John90976]

Are all the pro chips binned or just the ones on the lowest model?

In short, yes. They all have binned options, 7 and 8-core gpu options in m1. 8 and 10 core gpu options m2. 30/38 core gpu options in m2 max.

M1/M2 pro has the most variance with them binning against both CPU and GPU cores, but the base model is always the binned one with the 6 p cores and slightly lesser GPU. They seem to bin base m1/m2 and m1/m2 max chips solely off of GPU cores. I really feel like the 6 p core m1/m2 pro should be called M2 Plus as it is in the middle between base m2 and full m2 pro. They don't really advertise that you are getting 25% fewer p cores than in the full chip. They don't seem to pull that much trickery with any of the other chips so I think it could do for a different name. As they design more chips there is definitely room for more segmentation. Like, arguably, they'd never want to make the mac mini have a pro chip, but because of the way the current 3-chip structure lines up, it gets one for the current segmentation purposes.

 

[Sagnet]

As a side note, while waiting for the reviews: When a chip is binned, is it usually the case that most/all of the cores have some kind of degradations? Or is it normally just the disabled ones, and all the others perform well within specs?

Reframing the question: Do the cores on an unbinned chip have higher average quality than the (active/enabled) cores on a binned chip do?

 

[QuentinWilson]

As a side note, while waiting for the reviews: When a chip is binned, is it usually the case that most/all of the cores have some kind of degradations? Or is it normally just the disabled ones, and all the others perform well within specs?

Reframing the question: Do the cores on an unbinned chip have higher average quality than the (active/enabled) cores on a binned chip do?

If the cores are enabled, then they will have been tested to be well within spec. You don't have to worry about that. It's not any more risky to buy binned chips for the customer, just a way for the manufacturer to increase yields.

There is always what's known as the silicon lottery where some chips will just perform plain better than others, but that's only relevant when they're operating outside of spec. I.e., someone is tinkering with them. Overclocking, undervolting, that sort of thing. Not something you will or can do on a Macbook.

 

[donawalt]

Thank you @QuentinWilson @ilikewhey @John90976 for the thoughtful replies about binning etc., this has been very informative!

 

[playtech1]

My expectation is that there will be little real-world difference between the CPU speeds of the 10 Core M1 Pro vs the 10 Core M2 Pro. At best the improvements seem to be about 20% in raw CPU performance - which is a nice incremental uplift, but not so big that there will be any software that M2 Pro can run that M1 Pro cannot run.

Where M2 Pro shines will be in the package it comes in - WiFi 6E, HDMI 2.1 are nice additions and the upgraded neural engine and media engine could make a real difference if you use software that takes advantage.

Personally I would take an M1 model with 16GB / 1TB over an M2 model with either 8GB or 512GB.

I am interested to see what the GPU upgrade offers too - does it get close to low-end M1 Max territory?

 

[Sagnet]

My expectation is that there will be little real-world difference between the CPU speeds of the 10 Core M1 Pro vs the 10 Core M2 Pro. At best the improvements seem to be about 20% in raw CPU performance - which is a nice incremental uplift, but not so big that there will be any software that M2 Pro can run that M1 Pro cannot run.

The M2 will probably perform better. But will it come at the expense of more heat, noise and shorter battery life? That’s part of the equation.

 

[NiKeZz]

Would it be wiser to buy a M1 Pro 8core 16/500 or the M2 16/500? I’m a professional photog looking to upgrade our base editing station.

 

[John90976]

Would it be wiser to buy a M1 Pro 8core 16/500 or the M2 16/500? I’m a professional photog looking to upgrade our base editing station.

Considering the discount you will get, and the much faster drive, I would reccomend M1 pro compared to m2 pro. I cant fathom paying $2K/2.5K for a drive that is half the speed. Just mad.