I think this is even more impressive than m3 max, why? Because be better than just 9months older m2 pro BY REMOVING p cores and some bandwidth its an achievement. M3 max is the same routine like others you get more by adding more
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
M3 Pro Chip Barely Faster Than M2 Pro in Unverified Benchmark Result
- Thread starter MacRumors
- Start date
- Sort by reaction score
It's mainly due to the fact that M3 Pro has 6 Performance Cores and 6 Efficiency Cores while M2 Pro has 8+4.
People who are pushed to buy the M3 Max is just a fortunate side effect, but this was not the main reason Apple updated the M3 Pro and Max as they did. Before the release of the M3, the Mac lineup was a muddled mess. The MBA and 13” MBP were too close to each other and was confusing for customers. There wasn’t much difference between the Pro and Max M-series chips. That created two points of confusion for people who can’t figure out which machine to buy. So before this release, there were basically two tiers of laptops: 1) the MBA and 13” MBP in one tier, and 2) M-series Pro and Max in the second tier. That was it.
Now, with this release, there are four distinct product categories within the laptop lineup. You have 1) students and non-professional users, 2) the entry level Pro for light business users like product managers or marketing people, 3) moderate pro users, and 4) high-end pro users.
Apple went from 2 to 4 categories, each with clear separation with the other categories. From a business perspective, this needed to be done. Over the last couple of years, we’ve heard so many complaints about the messy product lines. Now that it’s sorted out, people have totally missed the real reason it was done.
I liked when the only difference between the Pro and Max chip was the GPU. That made sense and was easy to understand.
Now Apple is tying better CPU with better GPU and more memory with better GPU. If you want better one thing, you are stuck paying for better everything. Kind of just smacks of desperation to me.
My guess is Apple executives are under pressure to maintain pandemic level profits, and are trying to do so with sleazy moves like this. Either that or Apple significantly overplayed for N3B production.
Now Apple is tying better CPU with better GPU and more memory with better GPU. If you want better one thing, you are stuck paying for better everything. Kind of just smacks of desperation to me.
My guess is Apple executives are under pressure to maintain pandemic level profits, and are trying to do so with sleazy moves like this. Either that or Apple significantly overplayed for N3B production.
It's faster and uses less power.
Case closed.
This whole article is a storm in a teacup.
Would be interesting to find out why Apple chose to have a more efficient processor (marginally) vs a faster one (also marginally). But it's more of an acadamic topic.
In terms of product differentiation it might be that that creates a bigger gap to the m3 max so people who need a ton of CPU power go to the m3 max, and people who don't need that go to the m3 pro & get the better battery life that benefits their workflow more than CPU power they don't need.
Not apologizing for Apple here but IMO the M3 is the first time they have a lineup of machines and processors that really makes sense.
Whereas before, the MacBook Air was the same speed as the biggest M1 Max 16" beast in single CPU - which is the primary score that matters for day to day browsing and so on.
If they start to differentiate them better that's all good with me.
Case closed.
This whole article is a storm in a teacup.
Would be interesting to find out why Apple chose to have a more efficient processor (marginally) vs a faster one (also marginally). But it's more of an acadamic topic.
In terms of product differentiation it might be that that creates a bigger gap to the m3 max so people who need a ton of CPU power go to the m3 max, and people who don't need that go to the m3 pro & get the better battery life that benefits their workflow more than CPU power they don't need.
Not apologizing for Apple here but IMO the M3 is the first time they have a lineup of machines and processors that really makes sense.
Whereas before, the MacBook Air was the same speed as the biggest M1 Max 16" beast in single CPU - which is the primary score that matters for day to day browsing and so on.
If they start to differentiate them better that's all good with me.
So it is still faster and will probably be much more efficient in daily use. I will run mine in low power mode and let it serve me cool, quietly and forever.
Why even bother having binned and uncoiled versions of the same chip? M3, Pro, and Max would be the perfect number of options!!
Because its not realistic. For both consumers and apple. Nobody is upgrading their laptops every year.
Still will wait for real world results (and verified ones at that).
But it's not surprising to me. Both Pro chips have similar stats (and the M3 is less in some of those). I was hoping the chip being more efficient along with Ray tracing might make more of a difference.
If not, I'll just probably go with a refurbished M2 Macbook Pro.
But it's not surprising to me. Both Pro chips have similar stats (and the M3 is less in some of those). I was hoping the chip being more efficient along with Ray tracing might make more of a difference.
If not, I'll just probably go with a refurbished M2 Macbook Pro.
incremental performance increases. Exponential price tag increases
Still need to clarify the Mac Pro. iPad lineup could use some much needed cleaning up as well.
m3 pro price literally is the same.
always got to remember 80% of macrumors just say random crap.
As several here have said, this is lineup rationalization. It means we get a couple of machines that don't really get an upgrade this year, but it means there are a much wider range of choices (oddly NOT including an 8 performance core machine, which was the sweet spot for a couple of years - that really IS odd).
Counting performance cores, the M1 and M2 lineups went:4,6,8,8,8,16, with the 16 being desktop only. The bottom of that lineup is great (nobody was clambering for a 5-core or a 7-core). Why three different 8-cores, though? With two names?And that's a big gap between 8 and 16 cores.
The new lineup goes:4,5,6,10,12 (20,24 assuming the Ultra is what we think). The biggest gap in there is that there isn't an 8-core. The gap between 12 and 20 cores is both smaller (percentage-wise) and farther "out" - the 12 core Max is a very high performance chip, satisfying more needs without going to the Ultra.
They pushed performance out quite a bit by creating a higher-end option than they'd ever had before (12-core Max - of course it's more expensive, it's only the fastest laptop in the world by a significant margin). The 10-core Max takes over the price and performance space of the higher end 8-core Max with a nice upgrade.
There's definitely a chip missing (and one has been added right above the base M3). The 5-core Pro doesn't really fit a space where anything has been before - it's more than a base chip, but less than the previous smallest Pro. The 6-core Pro is a nice upgrade to the previous 6-core Pro, but not a replacement for the 8-core Pro or the lower-end Max.
They need one more chip, another cut-down of the Max (or call it . 8 Performance cores. If they offer an Ultra version of it as well, their lineup would go:4,5,6,8,10,12(16,20,24). That's a REALLY nice lineup, eliminating the gap between 6 and 10 cores AND making the gap between the top Max and the Ultra much smaller.
There's talk of both an Extreme (double Ultra) and, on the far opposite end, a very low-end Mac for the education market (might it use an A-series chip with two or three performance cores?
Counting performance cores, the M1 and M2 lineups went:4,6,8,8,8,16, with the 16 being desktop only. The bottom of that lineup is great (nobody was clambering for a 5-core or a 7-core). Why three different 8-cores, though? With two names?And that's a big gap between 8 and 16 cores.
The new lineup goes:4,5,6,10,12 (20,24 assuming the Ultra is what we think). The biggest gap in there is that there isn't an 8-core. The gap between 12 and 20 cores is both smaller (percentage-wise) and farther "out" - the 12 core Max is a very high performance chip, satisfying more needs without going to the Ultra.
They pushed performance out quite a bit by creating a higher-end option than they'd ever had before (12-core Max - of course it's more expensive, it's only the fastest laptop in the world by a significant margin). The 10-core Max takes over the price and performance space of the higher end 8-core Max with a nice upgrade.
There's definitely a chip missing (and one has been added right above the base M3). The 5-core Pro doesn't really fit a space where anything has been before - it's more than a base chip, but less than the previous smallest Pro. The 6-core Pro is a nice upgrade to the previous 6-core Pro, but not a replacement for the 8-core Pro or the lower-end Max.
They need one more chip, another cut-down of the Max (or call it . 8 Performance cores. If they offer an Ultra version of it as well, their lineup would go:4,5,6,8,10,12(16,20,24). That's a REALLY nice lineup, eliminating the gap between 6 and 10 cores AND making the gap between the top Max and the Ultra much smaller.
There's talk of both an Extreme (double Ultra) and, on the far opposite end, a very low-end Mac for the education market (might it use an A-series chip with two or three performance cores?
Yield problems. When creating wafers of dozens of chips, there are bound to be imperfections that make several of the chips defective. Useless chips cost money that cannot be recouped. So what companies do to salvage some of the chips is to find out what's still usable. Say it designed an M3 Max chip with 40 GPU cores, but discovered a bunch of the chips had 1-10 defective GPU cores. Without binning, they'd have to toss the whole chip in the trash. With binning, they can disable 10 of the cores and sell it as a binned 30 core chip even if 39 of them were still good (to keep down excessive skus). If you've bought a binned chip, you know there's at least one defective CPU or GPU core that's been disabled.
Things are especially bad for the new N3B 3nm process where yield was only around 60%. If Apple has to throw away 40% of the chips, they'd have to nearly double the price on those chips. If they can salvage a good percentage of those 40% bad chips, they can recover most of the costs.
"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."
The paragraph above is the reason why the difference between M2 Pro and M3 Pro is not that much. I believe, Apple has strategically made the decision to put more mileage between the Pro and Max chips and this will be the path going forward.
The paragraph above is the reason why the difference between M2 Pro and M3 Pro is not that much. I believe, Apple has strategically made the decision to put more mileage between the Pro and Max chips and this will be the path going forward.
Yes, but all this takes time. Last year, they cleaned up the iPhone line, and this year the Mac laptop line. Maybe next year they'll clean up the iPad line, which would account for rumors of a second iPad Air size. It sounds like Apple is preparing to revise the entire line, top to bottom, likely a reason they released no iPads this year.
Just think of this scenario. People have complained the Airs are too close to the Pros, so we might end up with 11” and 12.9” Airs with M3 and 11.1” and 13” iPad Pros with M3 Pro chips.
Last edited:
FWIW, I (and I imagine many others) looked at the M1 Pro vs. Max price/performance, and decided I was saving a bundle and getting more than enough performance out of the M1 Pro.
I think the new lineup is rational and while I hate the price increases, the market will determine what they're willing to pay. I'll be sticking with my M1 Pro for a while, the cost to meaningfully beat its performance is out of my budget for the foreseeable future.
I think the new lineup is rational and while I hate the price increases, the market will determine what they're willing to pay. I'll be sticking with my M1 Pro for a while, the cost to meaningfully beat its performance is out of my budget for the foreseeable future.
You could almost argue, do we need a new MacBook every year?
6% snt noticeable at all, And that's why you don't see it in the marketing materials.
Upgrading a machine because of 6% difference should have somebody put in an institution. Because it's crazytown.
14% .. barely enough to be noticeable. Still not worth it.
M3pro = dont do it. I would argue that real gains start around 30%, but that's just me. This is Apple at their worst, stripping down a level of product so that the higher end looks good. Just unfortunate and sad.
but new users are buying this year’s machines, and existing Intel or M1 users may be due for an upgrade. They want an upgrade on 2-3 year old machine, with 2-3 years worth of improvements so it has to keep moving every year
Some people will tear down Apple no matter what they do. They have products that are too close in performance to each other and Apple gets criticized for confusing people with undifferentiated products. Apple differentiates their products and they get criticized for making the high end look good. Which way do you want it?