Has Apple deliberately nerfed the M3 Pro CPU? And for what reason?
From Apple’s slides starting at 10:29:
M3 = 35% faster CPU than M1; 20% faster than M2
M3 Pro = 20% faster CPU than M1 Pro; No comparison to M2 Pro was given! 🤔
M3 Max = 80% faster than M1 Max; 50% faster than M2 Max
When Apple announced the M2 Pro they claimed it was 20% faster than M1 Pro. So are we to assume M3 Pro has no performance improvement this gen?
They’ve reduced the number of performance cores from eight to six, and as per the OP memory bandwidth at 150GB/sec is lower than the 200GB/sec of the M1 Pro.
It seems reducing the number of performance cores in favour of efficiency cores has eliminated any overall performance uplift M3 Pro had over M2 Pro. We’ll have to wait for benchmarks to be certain, but I’m sure Apple’s omission of a comparison to M2 Pro is very telling.
These things already had incredible battery life, I’m not sure why Apple would choose to sacrifice performance for more battery life? The people buying these machines, myself included, are pros that need performance, and the rest of the M3 family has CPU performance improvements, so why not M3 Pro?
You must log in or # to comment.
Gpu cores m3 for machine learning
Does anyone know if the M3 generation is on N3B or N3E?
I think the M1 Pro was too compelling of an option so they nerfed it to get more differentiation and force more folks to max. It may not work this generation because folks see what’s going on but eventually it will lead to higher margins.
The base M1 chip is still super compelling. I think Apple painted themselves into a corner with the M1 MacBook Air. I have one since 2020 and I’d buy another if it died. If you aren’t doing heavy workloads the M1 is fine even fanless. And I’ve toyed with some 4k video in iMovie so it’s not like I’m babying the thing.
Only place I feel like I’d really benefit from a better chip would be 3d modeling/rendering, heavy video / photo editing, etc
I think Apple has already hit general adoption for speed but I think they are working on 2 key aspects that were not ready for the M1/2.
- PCI lanes M1/2/3 all seem limited in this respect
- CPU/GPU extensions (ray tracing, etc… that you’re seeing on M3)
- Memory management
I think the M3 is their next iteration to tackle the above which is why it was at a throw away conference the day before Halloween. It’s not meant to be a generational shift, its iterative.
People like long battery life (a staple of Apple Macs) so its an easy sales feature while they focus on some more core functionality in future chips. Nothing wrong with the approach, also why I assume this is non-pro ATM, the IO/PCI lanes.
Just my speculation tho.
Yeah…but still no 27 inch iMac.
I mean it’s also cheaper than the M2 Pro spec for spec where I live (to the tune of £100 or £200 based on MSRP, though the M2 series models are now cheaper due to sales) so I think it’s probably an attempt to upsell M-series Base users to the M-series Pro.
This is probably also why the M3 Base made it to the 14 inch, minus the black colour and a criminal 8GB of base RAM and 256GB SSD (even if realistically it’s probably fine for a fair few people, I’d expect Apple to eat the minimal actual cost on a machine that’s £1700) to make comparisons favouring spending the extra for the M3 Pro easier.
M3 has 512GB SSD on the 14" MBP. Only the iMac has 255GB SSD.
It better to say the put effort into bespoke designing it rather than just taking part of the Max chips design… why did they put this effort in? I expect the reason is yields are lower on 3nm and cost per mm2 is much higher… You can see from the die shots how everything is packed much tighter on the M3 Pro compared to the M2 pro (this takes a lot of work to do and still have the internal frabic within the chip). Apple clearly wanted to shink the die size down otherwise they would have had to increase the prices of Macs with the M3 Pro.
Poor Mac mini gets shafted. Woulda liked to see 8 performance cores somewhere, considering the M3 Max moved up to 10 performance cores for the base
Can we expect benchmarks to pop up before the official release?
Single core performance will still be higher. Multi core performance might even be tiny bit smaller vs M2 pro. M3 pro will be significantly more efficient vs M2 pro as well.
I think what you’ve put is well-written and informative, and I agree with your analysis of the M3 Pro performance gains. It’s disappointing that Apple didn’t provide a direct comparison to the M2 Pro, and the reduced number of performance cores and lower memory bandwidth suggest that the performance gains over the M2 Pro are likely to be minimal, if any.
They didn’t seem to post a GPU comparison between M2 Pro and M3 Pro either? Perhaps I missed that as I’m a Max user.
However, as you mention, the M3 Pro still offers a significant performance boost over the M1 Pro, and it’s likely that the overall performance will still be excellent. For users who are upgrading from an M2 Max, the performance gains of the M3 Max should be more noticeable, especially in terms of CPU performance.
Overall, I think your post is a fair and balanced assessment of the M3 Pro and M3 Max performance gains. It’s important to note that Apple hasn’t released any benchmarks yet, so it’s too early to say for sure what the real-world performance difference between the M2 Pro and M3 Pro will be. However, based on the information that we have so far, it seems likely that the performance gains will be modest (I suspect around the 5% mark which Apple don’t really wish to tout as an overwhelming gain, and there may be some losses where active memory is key. Perhaps the dynamic cache will make up for some of this, but its disappointing to see a loss in bandwidth at this price scale.
I’m upgrading from M2 Max 12-core CPU, 38-core GPU 96GB ram, to M3 Max 16-core cpu, 40-core GPU 128 ram, and I’m hoping that will offer significant performance boosts, based on Apple’s claims, however if benchmarking does indicate otherwise, I’ll cancel my order and stick with the M2 Max for another year (well I’ve had it 9 months).
Why I upgrade so soon? It’s a laptop not a phone and the difference is small. Just curious on three reason ing not wanting to offend
Not offensive at all. I’m currently working on neurolinguistic models in deep learning as a lecturer and researcher at a university in London. These models require real-time processing of data feeds from MRI and EEG to be integrated into deep learning models for the study of consonant and vowel production activity in the brain. I can’t afford any lag in this process, and we’re in the process of upgrading to 9T MRIs, which will provide a significant bandwidth increase. Previously, our fMRI scans had a resolution of 1-2 mm, while higher-resolution scans, such as diffusion tensor imaging (DTI), can have a resolution of 0.5 mm or less.
Assuming a 1 mm resolution, the prefrontal cortex, with a volume of approximately 700 cm³, would contain around 700 million pixels. Wernicke’s area, with a volume of around 5 cm³, would contain about 5 million pixels. So, with the upgrade, we are dealing with 2.8 billion and 20 million pixels, respectively. However, most of these are filtered out when no activity is present, which is typically around 90% of the time, reducing it to approximately 300 million pixels during live processing. A 4k display contains around 10 million pixels, so we essentially have the equivalent of 30 of them as inputs. However, I’m only focusing on selective areas at any given time for deep learning, which amounts to around 10 million pixels. Running this through deep learning in real-time to detect patterns is extremely resource-intensive. So, every upgrade is highly beneficial. Functions like ray tracing and dynamic caching can also be utilized to support this. To reduce memory wastage in simplex analysis, as opposed to static memory running the same amount as in compound analysis, where you lose 40-50% of your GPU memory, it’s crucial.
Additionally, my university’s computing resources leave much to be desired, and the only alternative is using GPU farms, which introduce significant delays. I want to proceed as autonomously as possible. I can also connect my MacBook Pro to my Mac Pro (M2 Ultra with a 24-core CPU, 76-core GPU, and 192GB RAM) in my office using a 10Gbps LAN to use transcoder and share the workload, providing much lower latency compared to off-campus GPU banks (~10ms to 150-200ms for return).
I also have a semi-professional hobby of 3D modeling, which is again extremely resource-intensive when rendering motion from a complex mesh with dynamic shading. Both the increase in CPU and GPU power should help speed up this process, although it doesn’t have to be real-time but can be time-constrained.
I am somewhat disappointed that memory bandwidth hasn’t seen an increase with the introduction of ECC RAM, as it has been around for a while and seems like the next logical step. It’s available in the Mac Pro, offering 800 GB/s bandwidth, and there doesn’t appear to be a limitation in offering it in the MacBook Pro (although Apple’s additions to ARM are not typically open source, making it hard to know for sure until a teardown). Additionally, it’s worth noting that the M3 MacBook Pro offers nit syncing with the studio but not with the XDR display, which is more commonly used in research. Sitting at a desk for 12 hours running models and switching between multiple screens with different nit settings that don’t refresh simultaneously can strain the eyes. Perhaps they’ll add this feature later, or it may not be supported. In the meantime, I’ll run a backend like I did before, using delta time to keep them in sync, but that disables the promotion."
I do also actually replace everything every year basically. I’ve only skipped a few models of iPhone and Apple Watch. where the updates just seemed totally insignificant, so maybe a bit of an Apple tech obsession (well tech otaku all around really lol)
Wow amazing thanks for insight
The university must have a nice tech budget
As a M2 Pro owner the release of the M3 Pro just months after the M2 Pro really annoys me. Either the M2 Pro was a trick or the M3 Pro is a trick. They can’t possibly have made significant progress in a few months.
Apple is not beholden to Intel anymore and can control their release cycle better. The M2 chips were supposed to launch in Nov 2022 and got delayed. Ergo, the M3 stuff technically is on schedule. Also, you got what you paid for and finally, looks like the M3 Pro chip is a side grade at best compared to the M2.
They didn’t spend a lot of time comparing M3 to M1 so I assumed that there was little incentive to move up from an M2Pro. I do like that black, though.
Yes. Starting to feel like the M1 and M2 MBPs are gonna be the “2015 MBP” for the next decade
Apple claims M3 p-cores are 15% faster, and e-cores are 30% faster.
For M2, e-cores were 40% of p-cores
Base M3 Pro -> (5×1.15+6×0.4×1.3)/(6+4×0.4) is 17% faster than M2 Pro
Upgraded M2 Pro -> (6×1.15+6×0.4×1.3)/(8+4×0.4) is 4% faster than M2 Pro
Base is still decent bump. Otherwise, just jump straight to M3 Max.
Upgraded M2 Pro -> (6×1.15+6×0.4×1.3)/(8+4×0.4) is 4% faster than M2 Pro
Upgraded M2 Pro? You mean the 12 core CPU version?
Sorry, I meant to compare upgraded M3 Pro to upgraded M2 Pro.
Will edit my comment to reflect that.