M1 Pro vs. M1 Max: Real-World Performance Test

[Fabercon]

These machines should definitely last ten years. I can't believe the amount of horsepower on these things.

What do you know? They DO build them like they used to.

Apple is BACK, with a VENGEANCE.

Steve Jobs would be proud as hell.

What a way to start the decade!

Keep in mind this is still only the first generation with a lot of room for Apple to grow with CPU cores. If M2 Pro/Max shrinks the die and only adds an additional block of Cores (4 performant) without touching the GPU... it would still be a pretty significant update (at least 20% increase to multicore) without a lot of changes to the current Pro/Max design.

 

[Wando64]

What do you mean you "cranked up the samples to 1024" in the Logic Pro test?

If you mean you set the I/O Buffer Size to 1024, this would actually *decrease* the load on the CPU. A larger buffer size means that the computer has to process fewer buffers in the same time frame, at the cost of higher latency.

Setting the buffer size as small as possible would get the most stress on the system. That decreases the latency in the processing chain, which means the computer deals with more overhead, and has a tighter window to complete the processing of each buffer.

Yup. And on top of that, a project with 100 midi tracks is a pointless and meaningless test.

 

[centauratlas]

Our 16 inch Pro Max comes tomorrow (I think) from an initial Dec ~7 delivery. It sounds like I need one for myself but this is to replace the wife's 2013 MacBook Pro 15". I can't wait to give it a try though since I'll have to set it all up, migrate etc. It will make her 2013 machine look glacial I am sure, but it is also probably going to blow my 3 year old MBP out of the water. Maybe next year or 2023 I'll replace mine.

It is nice to see all these great reviews with real world performance matching or surpassing the hype.

I was a little skeptical at the announcement when all the details weren't out, but it sounds like a great Mac Book Pro update.

I do hope that these last at least as long as her 2013 machine, preferably longer. Her 2013 MBP is running well, but the battery is starting to swell and since it is a royal pain to swap it, it made sense to replace it. Merry Christmas to her.

 

[Luposian]

If i'm going to buy a Mac, it's going to be the highest end. I don't mess with "good enough". But, that notch... that kills me ever getting a MacBooks Pro. I'll get the highest end M1 Max Mac Mini when it comes out.

 

[SenorWhyMe]

can they run crysis?

yes

 

[Zdigital2015]

I don't need Max's performance, but I would pay the premium if the extra horsepower means Apple will support that notebook longer (e.g., OS upgrades) than one with a Pro chip of the same vintage. Is that typically the case?

No…both are going to be supported for the same amount of time. That’s not a reason to spend more money on the Max.

 

[Zdigital2015]

Very happy I went with the base model, it’s way more than I could ever need. The screen is perfect and I’m liking the new design.

You’re still rocking a primarily iPad Pro workflow right? Or has that changed?

 

[Tagbert]

I thought that the CPUs were pretty much identical between them... and assumed the single and multicore scores to be about the same.

The only major difference I can think about is the memory channel of M1 Pro: 200 vs M1 Max: 400.

Or is this one of those cases where the high performance mode comes into play?

These machines were not spec’d the same. It was more of a test between top and bottom. In this case, the heat capacity of the 16” was probably not a factor. It doesn’t sound like either machine was getting hot. this was more about testing how much performance difference you get if you spend money on the most expensive model vs the least. Seems like it depends on how much money you can make per minute vs how long it takes to render things.

14”
8-core CPU
16-core GPU
16GB RAM

16”
10-core CPU
32-core GPU
32GB RAM

With 2 more cores, you can see why the multi-core performance was higher. The single core performance was identical.

Wits twice the GPU cores, the 16” did a lot of the renderings in about 66% of the time of the 14”.

 

[Tagbert]

Yeah I guess you're right, I guess I'm trying to say that I think there should have been more negativity towards the keyboard. Perhaps folk were just so relieved that it was a major upgrade from the butterfly keyboard! It's something I'll get used to in no time, it's not terrible.

Seems like it would be nearly the same as the pre-butterfly keyboard. That is how people described the most keyboard on the most recent MBP when they replaced the butterfly with scissor switch keyboards in 2019. It has been compared to the magic keyboard. Definitly not a mechanical keyboard (thank god) but not the clacky and failure-prone butterfly.

 

[coachingguy]

COMPARISON REQUEST: 2018 and/or 2019 MBP compared to 14" and/or 16" M1 MBP in real world tests.

For those of us with laptops that are a couple of years old, it would be interesting to see how much faster a video would render, Logic Pro would export, and files would copy. If upgrading would save me an hour or two a week, I could make a business case to my boss on how purchasing me a new M1 MBP would actually save the company money.

I’ve got a late 2018 MBP i9, 32, 1TB and work got me a base MBA… it’s faster editing pictures and video with 15-20 tabs open in Safari than the MBP… So, I ordered the 14” base… setting it up now. ? Can’t wait to see how it performs against all 3 tomorrow.

Coachingguy

 

[jlc1978]

Yeah I guess you're right, I guess I'm trying to say that I think there should have been more negativity towards the keyboard. Perhaps folk were just so relieved that it was a major upgrade from the butterfly keyboard! It's something I'll get used to in no time, it's not terrible.

Nothing compares to the old IBM mechanical keyboards, but having that experience would require a 2” thick MBP.

It seems to take 5 minutes to watch a 5-minute video, no matter what I do. Also, I notice that MacRumors forum posts take just as long to read on either machine.

You’re reading them wrong.

 

[Tagbert]

I don't need Max's performance, but I would pay the premium if the extra horsepower means Apple will support that notebook longer (e.g., OS upgrades) than one with a Pro chip of the same vintage. Is that typically the case?

Generally, when Apple declares a model no longer eligible for new OS updates, it is because it does not support a particular hardware feature that has been deemed necessary to the OS or has components that just cannot be supported any more. Usually you get OS support back 7+ years. This year, they dropped a few models that were only 6 years old because they used an Nvidia video card that was no longer supported.

Short story: Apple will probably support any of these models for exactly the same time. Since these are all Apple tech with Apple CPUs and GPUs, they very well might support them longer than 7 years.

 

[CoMoMacUser]

Generally, when Apple declares a model no longer eligible for new OS updates, it is because it does not support a particular hardware feature that has been deemed necessary to the OS or has components that just cannot be supported any more. Usually you get OS support back 7+ years. This year, they dropped a few models that were only 6 years old because they used an Nvideo video card that was no longer supported.

Short story: Apple will probably support any of these models for exactly the same time. Since these are all Apple tech with Apple CPUs and GPUs, they very well might support them longer than 7 years.

This is the kind of concise, straightforward answer I was hoping for. Much appreciated!

 

[BuddyTronic]

But how does it perform watching cat vids on YouTube?

Well the 16" MBP with 8TB SSD and maxed out RAM (64GB) does pretty well.

I guess to describe it, I'd say....

It's Snappier!

 

[macman_canada]

How about compare the 2018 MacBook Pro 2.9 GHz 6-Core Intel Core i9, 32 GB 2400 MHz DDR4, Radeon RX 5700 XT 8 GB vs. M1 Max 10-core CPU, 32-core GPU and 16-core Neural Engine, 64GB unified memory??

 

[Durafab]

I just wish that (most) music software companies would update their software to M1 native. Most are still saying that their software is “compatible” with Rosetta on M1 Macs. Izotope. Native instruments. Arturia. All these major plug-in developers are suggesting that we run Logic in Rosetta. I believe that they would have had the dev kit 18 months ago, July 2020? Izotope is saying that their developers are hard at work with Mac Silicon, and expect to have native versions some time in 2022. Thats helpful. Funny thing though, thats a subscription service. So people like me are paying monthly for something we can’t even use (I paid a year in advance). I’m not going to run Logic in intel emulation. Native Instruments is basically saying that they’ll have it when they have it.

But the kicker is that Waves, of all companies, is now native with their new V 13. Hey NI, Izotope, and Arturia, Waves beat you to it! So did Serum, Valhalla, and several more. Those are on our machines, and I’m not sure we need anything else.

 

[szw-mapple fan]

Almost every YouTube channel compares M1 Pro in 14" and M1 Max in 16 inch. I would really like to see more comparisons for the M1Pro and M1 Max in the same chassis to see actual performance differences but more importantly battery life differences between the two M1 variants.

 

[Fomalhaut]

Just do a search for 14" M1 Pro vs 14" M1 Max on YouTube. There are soooooo many videos covering all kinds of comparisons of the new M1Pro/Max machines on YT.

Are there though? I put in that exact string “14" M1 Pro vs 14" M1 Max” into YouTube and didn’t find a single one that compares Pro and Max in the same spec 14” . There are lots of comparisons of 16” vs 14” or 14” with different memory or Pro CPU / GPU configurations, but not exactly 14" M1 Pro vs 14" M1 Max with the same memory spec.

Any links please?

 

[magbarn]

Received a 16" m1 max a few days ago, works great but I'm surprised to have seen no mention of the keyboards in any reviews.

Sure, it's better than the butterfly keyboards but it's a long way from feeling "mechanical" or even the retina mbp I'm upgrading from

Still a totally awesome machine, it would just have been nice to know that the keyboard is still rather "meh".

You’ve been spoiled. You didn’t have to deal with the years of purgatory of the butterfly keyboard.

 

[theorist9]

1) It's interesting the performance of the .5 TB and 1 TB SSD's is the same. Wonder if there's a performance jump in going to higher sizes (more lanes) and, if so, where that happens. One youtuber (Tally Ho Tech) said the jump is between 1 and 2 TB, but didn't explain where he got that info.

2) This is odd--they have the same cores, so the single-core scores shouldn't be that different. In particular, there seems to something wrong with the score for the M1 Pro--that's lower than what I've typically seen, which is in the mid-1700's. Wonder if he just got a slow chip through the luck of the draw. There's of course some inter-unit variation in chips, but that seems like a lot:

3) I thought the owner of GB explained that his software can't accurately assess the Metal score for the 32-GPU model, which is why it's not ~80k, as expected. That should have been mentioned in MR's review.

 

[amartinez1660]

It is somewhat reassuring see my ($$$) 10-core i9/5700XT 2020 iMac is still holding up fairly well to the entry level MacBook Pros here. Definitely a fair lower compared the Pro chip's compute scores (scores 1313 & 9748) and mid-way on the Metal score (~5800) between Pro and Max. It would for sure be crushed on any tasks that have dedicated computation cores, such as processing ProRes. It is also excluded from those handful of M1-only MacOS features.

Of course this computer takes a lot more wattage to run overall and is prone to fan noise despite the larger body. The M1 chips are certainly impressive, especially for power efficiency!

The Geekbench CPU/GPU scores on the M1 Pro/Max do no justice to the real world either…

Some benchmarks compete against workstation class CPUs, entering desktop “tons of cores AMDs” (forgetting how may, 32 I think) and saw today a comparison of a Surface Laptop with RTX graphics cards that on GB has 2x the posted scores of the baseline M1Pro yet 3DMark wildlife benchmark (a game-like test) spits out RTX ~8000 points vs M1Pro ~9250

To note too, Blender GPU rendering (and overall) is being reworked to be compatible with Metal with Apple engineers lending a hand. That’s when these test might take a new spin.

I got the same iMac by the way but 8-cores instead of 10. I have zero regrets and only good things to say, my daily workstation… but I’m pretty sure if an iMac M1 Super Duper appears soon I’m gonna be finding justifications ideas effortlessly to buy it.

 

[Makanmata]

Anybody have any idea when the updated 13" MPB will be released?

 

[sideshowuniqueuser]

Are there though? I put in that exact string “14" M1 Pro vs 14" M1 Max” into YouTube and didn’t find a single one that compares Pro and Max in the same spec 14” . There are lots of comparisons of 16” vs 14” or 14” with different memory or Pro CPU / GPU configurations, but not exactly 14" M1 Pro vs 14" M1 Max with the same memory spec.

Any links please?

I dunno mate, I'm not going to search for ya, but maybe have a look at Max Tech, he said he ordered about 12 different models to test. If he doesn't have one, maybe message him and ask if he can do it for you.

 

[theorist9]

Received a 16" m1 max a few days ago, works great but I'm surprised to have seen no mention of the keyboards in any reviews.

Sure, it's better than the butterfly keyboards but it's a long way from feeling "mechanical" or even the retina mbp I'm upgrading from

Still a totally awesome machine, it would just have been nice to know that the keyboard is still rather "meh".

Lots of the youtube videos I've seen discuss the KB. Check out, for instance, Lisa Gade (Mobile Tech Review). At 10:10 she mentions the keyboard is decent, but still finds the travel too short at 0.9 mm (which may be a bit shorter than that on the 2019 16" MBP, though she's not sure). For comparison, both the 2014-2015 era MBP, and the XPS 15, use 1.3 mm, and the Thinkpads (which many consider the best laptop keyboards) are 1.5 - 1.8 mm. Those three values (1.3, 1.5, and 1.8 mm) are 44%, 67%, and 100% larger than the MBP's 0.9 mm, respectively.

Personally, I really like my Logitech K811 which has both a very soft touch (very little effort to depress the keys) and (as best I can measure it) ~2.1 mm of key travel.

Of course, the new MBP's 0.9 mm is still better than the butterfly's 0.7 mm. Lisa does say it's very tactile, so you could almost describe it as the lowest-travel mechanical keyboard you've ever used, though she's being a bit tongue-in-cheek.

 

[gaximus]

I don't need Max's performance, but I would pay the premium if the extra horsepower means Apple will support that notebook longer (e.g., OS upgrades) than one with a Pro chip of the same vintage. Is that typically the case?

M1 chips will probably all loose support at the same time. But I’m sure that the Max will perform better on macOS(whatever) ten years from better than the other two.