Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
It's very difficult to take these numbers out of context and infer much of anything from them. Unless one is a chip engineer familiar with all of the complexities of this processor, then one cannot infer much from these data points.

These processors are extremely complex and there are so many components and dependencies. This really requires more expertise than what most of us have to pass useful judgement on this. I mean we have the 3 nm architecture; doesn't that count for something? Electrons have to cover smaller distances, so maybe not as much bandwidth is required? This is a very simplistic argument but it also illustrates why this is a topic where you cannot just cherry pick a number and make assumptions, because they most likely will be wrong.

Get degrees in physics, math, electrical engineering, etc...and then study the actual architecture of this chip thoroughly and then one can start maybe making useful judgements.

The main point is that these are more powerful processors and when tests are run I think we will see that. How they get there doesn't matter.
 
that would be the m3 max not this one.
Only some of the M3 Max: “the scaled-down M3 Max with 14-core CPU and 30-core GPU has only 300GB/s of memory bandwidth, whereas the equivalent scaled-down M2 Max with 12-core CPU and 30-core GPU featured 400GB/s bandwidth”
 
  • Like
Reactions: G5isAlive
I’ll reserve judgement on performance until some real world numbers, but on paper this certainly seems like a step backward.

I’m with others who feel this event had an aire of desperation from Apple. From the press leading up to it, the time slot (hello primetime Monday night), to the event itself - it just felt rushed. At least there weren’t any cringeworthy skits 🤷
 
I disagree! Who thinks 350 vs 400 GB/s is of any importance for his work will get the top tier model anyways.
For 99.9% of the users, it doesn’t make any difference.
You disagree with what? Last year it was slower SSD's for the same price. "It doesn't make any difference". This year, slower memory performance. "It doesn't make any difference". Next year, less cores. Every step of the way, it's the same price or higher. People should care.
 
You will love it
If I was buying one, I'm sure I would. I have an M1 Pro MacBook Pro and don't need to upgrade for another 1 - 3 years, but this looks like a really nice computer for people coming from something older. It would even be a nice upgrade for me, I just can't justify the cost at present for what I'd do with it.
 
This is outrageous. Apple should improve raw specs with every iteration, not go backwards in terms of memory bandwidth and Neural Engine. I'd expect the number of CPU cores to increase IN ADDITION to performance and efficiency gains for each individual core.

Only the benchmarks matter. We used to have the same kinds of pissing matches over raw GHz numbers in CPUs. Turns out GHz is not a great measure of actual performance.

And it's also ridiculous whenever I see them not comparing with the direct predecessor (i.e. M2). They should squarely compare M3 against M2, and this goes for A17 Pro too when they compared it against A14 instead of A16. Same with comparing against "the most popular Windows PC" etc

They are addressing people who bought M1s in 2021. This is the sensible thing, and I’m glad to see they aren’t trying to upsell people who just bought M2s a year ago. Laptops aren’t phones.

They have a tight time/attention budget to work with in a sales event and I think they used it effectively. Us nerds can do the real comparison later against whatever we want.
 
Only some of the M3 Max: “the scaled-down M3 Max with 14-core CPU and 30-core GPU has only 300GB/s of memory bandwidth, whereas the equivalent scaled-down M2 Max with 12-core CPU and 30-core GPU featured 400GB/s bandwidth”

lol ok then the 'best chip' is the m3 max option 2
 
You disagree with what? Last year it was slower SSD's for the same price. "It doesn't make any difference". This year, slower memory performance. "It doesn't make any difference". Next year, less cores. Every step of the way, it's the same price or higher. People should care.
The worst is they don’t even bother letting people know that the SSD speed was reduced. Obviously, they would never, but I bet many have no clue that they’ve tiered SSDs based on the storage size.
 
Frankly, the more I read about these M3 Pros the more satisfied I am with my M2 Pro.

Same. I would *love* to have an M3 Pro for no other reason than space black...but even with a discount (veterans/gov't), I'd likely pay a $400 premium if I trade my M2 Pro in. If there are some tangible speed increases and what have you, it might be worth it, but based on everything I'm reading right now, we're talking about minor advances.

I'm firmly in the camp of "It would've been nice to know about this a month ago when I bought my M2 Pro." LOL (But mainly/only for the Space Black color.)
 
The mid-range models are too good, so they are nerfing them to push people to spend more on the top end models. Clearly done to boost the revenues in the short term in the face of declining sales (via increased revenue per device).

As I have said before (and will say again), this strategy does work well if you are popular and make good products, but eventually you will push people past the point where they are willing to spend more and sales/revenue will fall off a cliff.
So, Apple is only making all these “mid-range” options because they want everyone to buy the higher end options, and these mid-range ones will….end up in a landfill?

I understand what people are saying with this theory, but they make it seem so absolute lol. Yes, the pricing structure will encourage SOME people to just go for an upgraded model, but many people will choose the mid-range options. And Apple wants to sell those people a computer.
 
  • Like
Reactions: wegster
I wonder what this means, overall, in terms of performance for the M3 Pro.



This also suggests — and Srouji’s artwork showing the three SoCs seems to confirm — that the M3 Pro, unlike the M1 Pro and M2 Pro, is no longer a chopped version of the Max. The non-suffix SoC was always a separate layout, but now the Pro is also separate from the Max. Interesting.

But overall, it sure reads like: the Pro’s performance barely changes at all; if you want that, you now have to upgrade all the way to the Max. And the Max, conversely, will have worse battery life due to fewer e-cores.
Yup, I remember this from the rumors that came out. It seemed odd, but now there’s more differentiation between the 3 SoCs and I think that will be better long-term.

The base M leans toward efficiency. It sips power and has (potentially) fan-less thermals. The Pro variant will be the balance between performance and efficiency, it draws more power and needs a fan, but retains most of the battery life. It makes the most sense in a laptop, hence on the desktop it only shows up in the upper-tier mini. The Max is (relative) performance over power. We now see this reflected in not just the GPU core count but the CPU count as well.

So we’re seeing the biggest shifts in the base (more efficient, less power/heat) and the Max (big performance jump with more performance cores, presumably at a cost of some battery life). The Pro was already more balanced, but now we’re seeing that balance all the way down to the SoC design itself.

I’m curious to see if Apple continues to keep the same clock speeds when the Max/Ultra arrive for their desktop systems. Historically that has been true, but Apple seems to be leaning into differentiating their products lines and their desktop systems are competing against Intel boxes that are more performant because power draw/thermals matter less. The Studio & Pro tower certainly have the PSU and cooling systems to handle higher clocked Ultras especially if the underlying silicon is upwards of 30% more efficient
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.