Apple MacBook Pro 14 2023 M3 Max Review - The fastest CPU in a 14-inch laptop

[Plum]

@A not sure what you are on about.

Some people say that there are use cases in which windows laptops are superior, and you are trying to disprove them by showing that there are things that don't run on x86 laptops at all?

[A]

Some people say that there are use cases in which windows laptops are superior, and you are trying to disprove them by showing that there are things that don't run on x86 laptops at all?

You need to follow the conversation before commenting that you are not sure what's this about.

It started about efficiency. Guy came in with 4090 performance banner. I've shown him that that performance banner can work both ways, if you cherry pick tests.

[A]

P.S. We kinda need to agree what is cherry picking first actually, showing that 4090 is faster using small models noone is using anymore or using real 2023 AI workflows when it's slower )

[Toortle]

youtube.com/watch?v=jaM02mb6JFM

Or you can just watch a video of how 4090 is faster on 7B model (noone actually uses those)
slightly faster on 13B model (probably around the best you can run on x86 laptop)
and M Max RAPES 4090 on 70B model (which is de-facto most balanced for running locally today) - and he explains that building x86 rig able to run it on GPU is worth more than M Max and uses way more power than M Max.

Fun starts at 5:47

At 8:52 you can also see performance/power charts that are quite crucial in AI productivity, 4090 loses.

Wait, so you want to say that the only purpose of M3 Max Macs is AI/ML (because they clearly lose in everything else such as content creation, gaming, engineering software (95%+ of it doesn't even exist for macOS nor it can be emulated) and so on?

And then you compare a $6,000+ M3 Max Mac with a $2,000 GPU? Ok, how about adding 2x 4090 + i9 14900K then into the equation to get a 40% faster CPU, 64 GB or RAM and 48 GB of VRAM for about the same price? pcpartpicker.com/list/LXt3mD

Or you can slot inside one Quadro RTX 8000 (4608 CUDA cores, 576 Tensor and 72 RT) for the same amount of VRAM and to save $300, a GPU specifically made for AI/ML. Add two of them, you get 96 GB of VRAM, and with just 20% higher price for the whole PC you get roughly 300% higher overall performance than a fully maxed M3 Max with 128 GB.

[A]

Wait, so you want to say that the only purpose of M3 Max Macs is AI/ML (because they clearly lose in everything else

How did you come to conclusion "they lose in everything else" looking only at Blender benchmark? ))) It's on par with most top-end x86 CPUs, it's extremely fast in image and video processing, and is plenty fast for most productivities.

Ok, how about adding 2x 4090 + i9 14900K then into the equation to get a 40% faster CPU, 64 GB or RAM and 48 GB of VRAM for about the same price? Add two of them, you get 96 GB of VRAM, and with just 20% higher price for the whole PC you get roughly 300% higher overall performance than a fully maxed M3 Max with 128 GB.

Is it portable? And what is the energy consumption compared to M3 Max's 60Watts?

So you kinda agree that power efficiency-wise x86 will be nowhere near M Max and to beat LAPTOP using 60W you need a desktop rig with top end CPU and two top-end GPUs for a total of what, 800-1000W? ))))

[Toortle]

Wait, so you want to say that the only purpose of M3 Max Macs is AI/ML (because they clearly lose in everything else

How did you come to conclusion "they lose in everything else" looking only at Blender benchmark? ))) It's on par with most top-end x86 CPUs, it's extremely fast in image and video processing, and is plenty fast for most productivities.

There is no conclusion there at all, it's just the fact that it's nowhere near with anything comparable for the same amount money, be it in video editing, 3D modelling, creation and animation (Blender needs GPU more than CPU). You are here trying to portray M3 (Max only?) Macs as an ultimate machine capable to be faster and better than literally anything on the market for any existing purpose, while it happens to be clearly incorrect. Of course it is plenty fast for most productivities - at twice the cost of another equally fast laptop.

Ok, how about adding 2x 4090 + i9 14900K then into the equation to get a 40% faster CPU, 64 GB or RAM and 48 GB of VRAM for about the same price? Add two of them, you get 96 GB of VRAM, and with just 20% higher price for the whole PC you get roughly 300% higher overall performance than a fully maxed M3 Max with 128 GB.

Is it portable? And what is the energy consumption compared to M3 Max's 60Watts?

So you kinda agree that power efficiency-wise x86 will be nowhere near M Max and to beat LAPTOP using 60W you need a desktop rig with top end CPU and two top-end GPUs for a total of what, 800-1000W? ))))

No one even says that x86 more power efficient but why is power efficiency even important at all for a plugged in machines? M3 Max's GPU performance on battery is heavily nerfed, some 40% according to pretty much every existing review and measurements on them so far.

So you would rather get 101.7 Watt usage (measured in this existing review here) than a 300-700% faster PC with higher power consumption and with complete user reparability and freedom to upgrade/replace anything you want whenever you want? And let's also mention that your PC (desktop or laptop) won't die from literal dust: youtube.com/watch?v=AH5F73qPtQQ

[Toortle]

So you would rather get 101.7 Watt usage (measured in this existing review here) than a 300-700% faster PC with higher power consumption and with complete user reparability and freedom to upgrade/replace anything you want whenever you want?

For the same price, I forgot to add.

[A]

There is no conclusion there at all, it's just the fact that it's nowhere near with anything comparable for the same amount money, be it in video editing, 3D modelling, creation and animation (Blender needs GPU more than CPU). You are here trying to portray M3 (Max only?) Macs as an ultimate machine capable to be faster and better than literally anything on the market for any existing purpose, while it happens to be clearly incorrect. Of course it is plenty fast for most productivities - at twice the cost of another equally fast laptop.

I never portrayed it like this, we were talking about _efficiency_ against x86 laptops and all of a sudden you decided to prove me something i know with your Blender shenanigans. Why not Geekbench? Geekbench is less convenient for you )

Of course it is plenty fast for most productivities - at twice the cost of another equally fast laptop.

"Equally fast" laptop that can't perform equal number of tasks and productivities*

No one even says that x86 more power efficient

This whole conversation was about efficiency when you've arrived.

M3 Max's GPU performance on battery is heavily nerfed, some 40% according to pretty much every existing review and measurements on them so far.

No

So you would rather get 101.7 Watt usage (measured in this existing review here) than a 300-700% faster PC with higher power consumption and with complete user reparability and freedom to upgrade/replace anything you want whenever you want?

People buy LAPTOP because they want LAPTOP. The fact that you have to use desktop and cram two GPUs in it to beat laptop is already a win for laptop, despite the fact desktop will be faster.

[A]

For the same price, I forgot to add.

Comparing desktop to laptop prices is VERY convenient for you )))

[Toortle]

I never portrayed it like this, we were talking about _efficiency_ against x86 laptops and all of a sudden you decided to prove me something i know with your Blender shenanigans. Why not Geekbench? Geekbench is less convenient for you )

My Blender shenanigans? It was you who mentioned it first, I didn't say anything about it until my last comment. I mean if you think that CPU is more important for Blender then you are free to run it at i9 with iGPU. And Geekbench, hmm... 7945HX 2882/17564 vs M3 Max 3096/21243. Then with that Ryzen you get 32 GB of upgradable RAM and an RTX 4090 with 16 GB VRAM. For $3,500, solid 35-ish% less than this M3 Max MacBook in this review.

"Equally fast" laptop that can't perform equal number of tasks and productivities*

So that laptop above (Strix Scar 17) can't perform equal number of tasks and productivities? Seriously? And you are dead-serious about you not trying to say that the M3 Max is an ultimate machine better than anything? Nice, early April 1st.

No one even says that x86 more power efficient

This whole conversation was about efficiency when you've arrived.

I'm afraid that you will have to quote me on that, to show my post where I said that because I never even mentioned x86 all, searched even with Ctrl+F, until quoting your own words in my previous post.

People buy LAPTOP because they want LAPTOP. The fact that you have to use desktop and cram two GPUs in it to beat laptop is already a win for laptop, despite the fact desktop will be faster.

I agree. Which is why the mentioned Strix Scar 17 for $2,000 is less is a much better choice as it's better than an M3 Max Mac no matter how high you spec that M3 Max - even with 128 GB you can freely upgrade that ROG to 128 or more and you can use an eGPU with 48 GB of VRAM - and still be $1,800 cheaper than that fully maxed M3 Max MacBook with higher performance. Plus you can play any game you want on it in full experience, as opposed to MacBook with halved fps (in the best scenario) 😁

[NikoB]

No problem, you are also getting smarter and smarter every day because of our conversations.

Just in a week you've accepted that
a) MiniLED is cool
b) Apple is the best at efficiency today
c) GPTs can be run locally and do not require HUNDREDS OF TERABYTES OF RAM

Maybe it's not all lost for you and one day you will be less of a clown.

Pathetic shameful bot A, lies outright and seems to be already using the forceful help of the moderator (no brain, no arguments other than bullying, so we use force when we lose - familiar?), who, to his shame and disgrace, ignores truth and science for the sake of bullying and cheap gaslighting. Which only confirms the shame of this site more and more. There aren't many readers here. But whoever reads all this sees...

[NikoB]

What's wrong with this super expensive laptop?

We were promised RAM speeds of 300GB/s(for top config old 400GB/s from M2 MAX) for M3 Max. The result of the test is almost 3(4 for top max config ) times less, because the real efficiency of the memory controller is only 40%(<40 for max). Despite the fact that 400GB/s was declared in the M2 Max - but we did not see tests, only declarations.

Where does this impact the most? Yes, just when igpu is working. For it, the decisive factor is the bandwidth of the RAM, which is shared with the system and other devices.

Go ahead. Let's look at the performance of the SSD - a pitiful only poor 53MB/s@4k IOPs, despite the fact that even such a slow memory(vs official declaration) pumps ~10 GB/s at 4k IOPs, according to the test results. The difference is 200(!) times in favor of the RAM, which is at least 64GB, which reduces the need for swap with such a slow SSD to a minimum.

Well, the bots above have already talked about the slow (by response) screen. PWM is ok - 14kHz not visible for all. Can only add the strange (if not surprising) reluctance of the authors of Mac Book reviews to carry out hardware screen calibration and post the results after it.

It can also be noted that despite the premium segment and exorbitant price, a 14" laptop weighs as much as 15.6 x86 laptops, which is clearly too much for this segment. However, Apple fans are apparently pumped-up people.

The author does not write what the real noise is in intensive surfing in the maximum performance profile (where impulse performance is maximum, and this is what we need). I suspect that this laptop is not at all quiet in such surfing.

I would like to have noise tests in 4k@60fps and 8k@60fps mode (if supported by the system) on the same, for example, YouTube. Will the laptop be completely silent after at least half an hour of video playback? A simple test that immediately shows who is who.

Temperatures at the keyboard level are too high. This laptop is definitely dangerous to use with the lid closed with a high load on the processor with an external monitor - which could potentially lead to faster degradation of the miniLED panels whose critical operating temperature is approximately 50C.

In about another 3-4 years, Apple will also reach a dead end with technical processes in terms of performance per 1W - selling a new one with minimal improvement will be more and more difficult, especially at such prices. So, what is next? Increase TDP like Intel cheaters? Or is it again selling "rhinestones" instead of real progress in performance per 1W?

But is all this interesting to those who will actually buy it? After all, they buy Apple... ;)

[Toortle]

For the same price, I forgot to add.

Comparing desktop to laptop prices is VERY convenient for you )))

Not at all. It is you who linked the video of a guy doing AI/ML where even he said that a 36 GB is incapable to complete anything and that 128 GB is basically required. And it is very convenient for you to basically claim at this point how Mac users are all doing AI/ML on the go (so they want laptops) instead at home with equally priced PC for up to 700% higher performance per $.

[NikoB]

It is clearly visible that the whole "discussion" comes down to the maximum removal of the most valuable comments from the top next to the review text. If the editors change the arrangement of comments at the end of the review to normal - the old ones will be first, as it should be, and all the bots will immediately start wagging their tails in anticipation articles, because whoever writes that comment first and will read the majority. The topic is as old as forums and comments on the Internet.

[NikoB]

Any decent and sensible person can easily determine who the liar is here by reading your stupid nonsense and what I write. So relax, you lost a long time ago and showed how much dumber and less educated you are than me.