Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Well, yes? It's much easier to make performance gains when you're not near the current practical maximum.
It’s a lot easier to make performance gains when you don’t nerf your products on purpose by reducing cores or memory bandwidth.
 
  • Like
Reactions: ric22


Apple's new M3 Pro chip with a 12-core CPU offers only marginally faster CPU performance compared to the M2 Pro chip with a 12-core CPU, according to a Geekbench 6 result spotted today by Vadim Yuryev, co-host of the YouTube channel Max Tech. This is only a single benchmark result, so further results are needed to ensure accuracy.

m3-pro-chip.jpg

Apple announced new 14-inch and 16-inch MacBook Pro models with M3, M3 Pro, and M3 Max chips on Monday. This benchmark result appears to be for the 14-inch model, based on the "Mac15,6" model identifier listed.

The result shows the M3 Pro has a single-core score of 3,035 and a multi-core score of 15,173. If these scores are accurate, the M3 Pro is up to 14% faster than the M2 Pro in terms of single-core performance, but only up to 6% faster in terms of multi-core performance for the most demanding tasks and workflows.

Both the M2 Pro and M3 Pro are equipped with up to a 12-core CPU, but the M3 Pro has only six high-performance cores and six efficiency cores, while the M2 Pro has eight high-performance cores and four efficiency cores. So while the M3 Pro is manufactured with TSMC's 3nm process, compared to 5nm for the M2 Pro, the chip's resulting performance gains are diminished due to it having two fewer performance cores. The M3 Pro also has 25% less memory bandwidth and one fewer GPU core compared to the M2 Pro.

By limiting the M3 Pro's number of high-performance cores, Apple has created more differentiation between the M3 Pro and the M3 Max, which has up to 12 high-performance cores. However, as a result, the M3 Pro is only marginally faster than the M2 Pro.

Geekbench 6 results from earlier this week showed that the M3 Max is up to 45% faster than the M2 Max, while the standard M3 chip is up to 20% faster than the standard M2 chip, so the M3 Pro is by far the least improved chip of the series. Of course, the M3 Pro is still a significant upgrade for those coming from an Intel-based Mac. M3 Pro is also up to 20% faster than the M1 Pro chip, which is a decent improvement.

As always, benchmarks provide a useful reference point, but real-world performance can vary.

The new MacBook Pro models are available to order now, and M3 and M3 Pro configurations will begin arriving to customers and launch in stores on Tuesday, November 7. M3 Max configurations will launch later in November.

Article Link: M3 Pro Chip Barely Faster Than M2 Pro in Unverified Benchmark Result
So basically what you're saying is that I should buy an M1 Pro lol
 
  • Like
Reactions: mr_jomo
I honestly think most Mac users don’t care about ‘performance’ or ‘cores’. They want to be seen drinking coffee with the latest MacBook. Only way to show you have the latest is buy black. So base black M3pro will be the biggest seller. Look at MKBHD, he’s even said his current MacBook is great. But he will ‘need’ black. That’s the majority of buyers.

Yes this is what I call Mac passion
 
  • Like
Reactions: Urban Splash
I don’t think this will turn out to be true. They are difficult colors to capture in photos/videos, so I look forward to some naked eye comparisons in store.

And… no fingerprint protection whatsoever instead of at least having some.

There don’t seem to be any comparisons on YouTube atm.
it is true
Is dark grey. Midnight blue is darker
The only true black mac was the plastic one from decades ago
 
I think the performance of the M3 max chip will also spell doom for the M3 Ultra in the Mac Studio.

I wouldn’t be surprised if they keep the M3 Ultra chip for the Mac Pro tower so they can squeeze even more $ out of their customers.
 
It’s all marketing talk, M1 Max is still over kill for 99% of people, they just do a new colour and tweak the brightness in software like the app vivid does and make people want it
 
  • Like
Reactions: Chuckeee
not surprisingly- I think it was the only M3, where they did not show the M2 curve in the event.
Should be more energy efficient given the lower performance core count + GPU and AI gains …
 
14% and 6% are decent improvements. Why would you expect more for an annual upgrade? We don't need a revolution every year...just incremental improvements.

And there is a difference between wanting and needing more powerful chips.

Jeez what's wrong with people who have no expectations nowadays. Everyone should be demanding the fastest and most powerful processor from Apple, not being satisfied with lacklustre crap that just manage to scrap past.

With that attitude of not expecting only the best, there won't be innovation and we'll be stuck with crappy incremental updates.


If you're so satisfied with current tech, then you can be a laggard and just keep using the first gen iPhone forever. After all, there won't be a "need" for you to upgrade every year.


Steves Jobs would never have settled for anything less than the very best, including large performance leaps every year. Expectations and the reality of being able to deliver or not is another thing, but at least if there's no expectation, there would be zero chance of innovation.
 
14% and 6% are decent improvements. Why would you expect more for an annual upgrade? We don't need a revolution every year...just incremental improvements.
Such improvements are barely noticable for the average user. I didn’t even notice much of a difference when going from a 2015 mbp to a 2019 i9 in single core apps. With multi core apps the story was a bit different but over all: mehh.
Early this year I switched to the 16” mbp m2 max (with a huge discount). Now that’s the kind of improvement I’d like to see every time I replace my mbp. Guess I’ll have to wait for at least the m6 max to re-experience that.
 
  • Like
Reactions: killhippie
It's interesting to see that at 3nm there is already one class of M3 chips that only see 6% speed improvement.

This seems to suggest that the new iMac and MacBook Pro are good buys because 1) 2 nm-based chips are probably at least two years away and 2) speed improvement will be marginal for the M4 series which will still be 3nm-based.
 
For those who follow chips, the modest performance boost of M3 is not a surprise. Taiwan Semiconductor(TSM) started struggling with the N3 process node shrink in 2020. Die shrinks which between previous nodes were 2-2.2, are barely 1.7 in N3. More importantly n5 to n3 brought zero shrink on SRAM memory meaning no improvement in speed or memory density. So Apple was forced to adapt chip design architecture and hence we see reduced memory bandwidth and the importance of the announcement of improved memory allocation. They also made significant changes in processor core work allocation, which has made geekbench score suspect in the near term. So that all said I would argue the real story of a sea change in the relationship between chip hardware, circuit design, OS is introducing fundamental changes in the whole system design stack. It is also very clear that node shrink marketing (ie n5 to n2b n3e or n3d in TSM case) has become functionally meaningless. I can provide references for those wanting to geek out.
 
Most likely M2 Pro users will not upgrade to M3 Pro, right? :-D
And it is still faster than Snapdragon Elite at 80W...using half of the power.

So yeah, some people are complaining about performance but there isn't anything better in that category anyway. Apple is still on top. ~15000 in 45W envelope..is pretty good.

For new/Intel users it is decent mac. For M1 Pro, M2 Pro, there is no reason to upgrade.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.