Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
If Apple can’t make an M2 Extreme chip with twice the cores of the M2 Ultra, then they should just delay the Mac Pro.

It would be embarrassing to have a 2023 Mac Pro that is only marginally faster than a year old Mac Studio.

Refurb Mac Studio Ultra’s are going to fly off the shelves!
 
For example, the Intel Core i9 13900 is almost 50% faster than the Intel Xeon W 3.2 GHz, which is the base CPU of the 2019 Mac Pro.
No, the i9 is 100% faster than the Xeon. The latter is 50% slower than the i9.

But suppose you NEED server features, such as ECC memory and very large amounts of RAM available.
If you have money to burn, you can go instead with an AMD EPYC 9659, which is almost 14% faster single thread, and scores 70% faster in CPU Mark:
No, the Epyci9 is 300% faster than the Xeon. The Xeon 75% slower than the Epyci9.

I just had to correct these basic math errors, because they were so completely wrong.
 
Last edited:
You said that the i9 is 100% faster than the Xeon, and then 300% faster.
Which is it?
Also, you have to point out the exact model, because obviously there are several Xeon models and several i9 models.
 
Now, regarding the Mac Studio: the simple fact that Apple is avoiding re-releasing the Studio so it will not conflict with the Mac Pro shows that they consider the two very similar, which is very telling. This shouldn't EVER happen.
 
You said that the i9 is 100% faster than the Xeon, and then 300% faster.
Which is it?
Simple typo, as the latter i9 should be the Epic, which should be clear since I am quoting you. I've made a clear edit in the post to help you out.

Also, you have to point out the exact model, because obviously there are several Xeon models and several i9 models.
I am quoting models you found, so that should be pretty clear to you.
 
They mean the i9 is 100% faster, and the EPYC is 300% faster.

Huh? They’re referring to the very specs you’ve posted.

It's just that the way you worded, it makes it sound that all i9s are faster than all Xeons, which is obviously not true. But it was just poor wording. It's all clear what you mean now.

Regarding the data, I was basing myself on the data from cpubenchmark.net (as you can see, the images are in the previous topic pages).

How did you get 100% and 300%, respectively? Can you please clarify? Was it based on the absolute single thread / multi-thread numbers?

Of course, 100% and 300% makes the comparison even less favorable for the Mac Pro. Ouch!
 
Because it’s comparing in the other direction.

If B is 300% (four thirds) faster than A, then A is 75% (three quarters) less fast than B.

Ooooh, I see. Sorry for my mistake. I should have done the percentage calculation manually.

Either way, the fact that a cheap $700 CPU is 100% faster in 5 years is scary – and it's even more scary that Apple kept those prices.

That also speaks volumes on how fast technology has moved. We tend to think speed increases have slowed down, but in theory a server processor should take much longer to become obsolete – not just five years!

People also like to point out that Intel runs hot and inefficient.
It's true that it can improve, but this cheap $700 now spends HALF the energy of the obsolete, five-year-old Xeon. That is impressive too.
 
I agree if they haven’t figured that out yet then they shouldn’t release something half assed, and the Intel MP should continue to fill that gap until they do. I just hope that’s not too far away. With all the rumors of an ASi MP coming real soon one can only hope they’ve got it right.

Apple shouldn't leave it there to fill a gap – not at the current prices, at least.

Check the numbers we're discussing. A cheap $700 CPU is 100% faster than the old Intel CPU, and the server-grade material (AMD EPYC) can easy get to being 300% faster. All while costing $15,000 – $20,000 (the CPU costs $11,000 and 1.5 TB RAM costs around $1,300, so I'm giving some room here for the other server components).

That means that if you actually want a server grade material, current chips can achieve much better performance for half to a third of the price.
 
That also speaks volumes on how fast technology has moved.

That, but also, how long Intel was stuck in their Skylake-derived / 14nm microarchitectures. Ice Lake, Sapphire Rapids, etc. were huge boosts in part because the previous years had been smaller boosts.

It's true that it can improve, but this cheap $700 now spends HALF the energy of the obsolete, five-year-old Xeon. That is impressive too.

Welllllll, actual power consumption of a 13900KS while being benchmarked is far higher. https://www.techspot.com/review/2607-intel-core-i9-13900ks/

On average we saw a package power of 280 watts with a peak of 300 watts


These power measurements show that the Core i9-13900KS will rarely reach its 320W power limit with standard applications — there simply isn't enough thermal headroom due to the difficulty of dissipating this amount of heat from such a small area (thermal density).

I'm actually not sure where that TDP of 125W comes from, because it ain't Intel: https://www.intel.com/content/www/u...-36m-cache-up-to-6-00-ghz/specifications.html

Processor Base Power
150 W
Maximum Turbo Power
253 W
 
In theory, you could still underclock it to get a lighter power consumption and it would still be more powerful than the Mac Pro.

Yes. (There's other considerations, such as lack of ECC support.)

But also, the M1 Ultra is already much faster than the Xeon W-3245, while burning about a third the power.
 
Not particularly well. 😉

I had one for a short time. I’m not certain but if I recall correctly the entire Windows OS ran in a window in a dedicated Mac app. There was little integration if any between the Mac and Windows OS’s. But it ran natively. Better than emulation, for sure and better than virtualization even since it was the only thing running on that CPU, dedicated RAM, etc. just sharing monitor space, keyboard and mouse.

Of course someone please correct me if I’m remembering that wrong.
Thanks for sharing your experience. I suspect this tech could be done much better today.
 
  • Like
Reactions: Detnator
Why would they do that?
They would sell their Mac Studio at a loss to buy a new machine that is just fractionally better?
For me it would be to get a quieter machine. I have a Studio Max with the whine, but haven't got to that point yet.
 
If Apple can’t make an M2 Extreme chip with twice the cores of the M2 Ultra, then they should just delay the Mac Pro.

It would be embarrassing to have a 2023 Mac Pro that is only marginally faster than a year old Mac Studio.

Refurb Mac Studio Ultra’s are going to fly off the shelves!

The Ultra is already embarrassing.
It's slow and overpriced.
It's barely 50% faster in many tests compared to the M1 Max and considerably slower than Intel 13th gen with a mid tier graphics card.
 
It’s true FCP is no longer an indie filmmaker darling. But fun fact: 2020 Oscar winner “Parasite” is said to have been edited in FCP7, on a Macbook Pro.
Nice! I still think it proves my point though, they used Final Cut from 2011, the last time Apple's "Pro" software/hardware was relevant.
 
The iMac Pro came very late to the game, towards the end of that design form. You could say it was the Audi 80 Sport edition. The Mac Studio was and is a different and separate line of computers, with multiple configurations. It is entirely possible that it could go the way of the G4 Cube but is that unlikely? The G4 Cube quickly drew criticism for its underwhelming performance. By contrast the Ultra M1 Studio still holds its own in benchmark tests. It might be unwise for Apple to launch a M2 Ultra Mac Pro and a M2 Ultra Studio at the same time. However, it is entirely in Apple’s strategy to greatly inflate the price of the baseline M2 Mac Pro and put it way beyond a M2 Studio. Considering the R&D Hindenburg of the Mac Pro 2013, it is also possible Apple can still design itself into a corner. The Studio by contrast doesn’t diverge from the current Apple ecosystem. When the Mac Pro 2013 arrived it was pulling in both directions. It had limits on expansion but also tried to be a Mac Pro cheese grater of old. If anything the portent of history repeating itself for the Mac Pro M2 is more concerning than the fate of the Studio. A Mac Pro with limits on graphics cards, non upgradable ram, it all sounds eerily familiar to a certain trash can idea. Maybe the Studio is a dead duck but I’d be more worried about the Mac Pro’s future.
 
Last edited:
I bought a base model M1 Mac Mini when they first came out, not knowing it would be so good that it would quickly become my main computer. I have been wishing I specced it out more for a long time. I need a new machine and today I almost bought a Mac Studio M1 Max 32 core because of these dumb rumors that it won't be updated, and I'm tired of waiting.

But I decided to hold off and suffer longer, because these dumb rumors make absolutely no sense to me. I think Mark Gurman is full of sh*t with his predictions... he just makes up 50 things and then when 2 of them turn out to be right everyone praises him for his amazing track record. It would be absolutely ridiculous for Apple to stick an M2 Ultra in a giant empty cheesegrater case and call it a Mac Pro. I just don't think they are that dumb. Unless they make something way more impressive than that and actually have a reason for PCIE slots (AMD GPUs), it just seems illogical.

I bet if there is anything to these M2 Ultra rumors, it's actually about the next Mac Studio, not a Mac Pro. If Apple was serious about the Mac Pro I don't think they would have waited this long to update it. They could have slapped an M1 Ultra in the cheesegrater case a long time ago if they really think that's all it takes.
 
I'd love to see an M2 Ultra Mac Studio (192 gb ram would be very tempting), I'd also love to see a Mac Pro that you can stick multiple M2/3 Ultra's in. It will be interesting cuz everyone is expecting the AS MP to break convention in some way in order to hit the memory amounts and GPU performance the target audience for the MP expects.
 
I bought a base model M1 Mac Mini when they first came out
...
I almost bought a Mac Studio M1 Max 32 core because of these dumb rumors that it won't be updated, and I'm tired of waiting.
...
But I decided to hold off and suffer longer, because these dumb rumors make absolutely no sense to me. ...
I'm sorry to hear about your suffering with that M1 Mac mini as you wait, but maybe you'd like to swap machines and wait with my 2014 i7 MBP. 😂

I feel the same way on the Mac Studio M1 Max. I won't be happy if an M2 (or M3) Studio shows up later this year (as it should have already). At the same time, putting what I want into the Mac mini Pro brings the price too close to the Studio.
 
  • Like
Reactions: Serqetry and spaz8
I'm sorry to hear about your suffering with that M1 Mac mini as you wait, but maybe you'd like to swap machines and wait with my 2014 i7 MBP. 😂

I feel the same way on the Mac Studio M1 Max. I won't be happy if an M2 (or M3) Studio shows up later this year (as it should have already). At the same time, putting what I want into the Mac mini Pro brings the price too close to the Studio.
Lol, I know... "suffer" is relative, could be a lot worse.

Exactly, the pricing on the M2 Pro Mac Mini is really dumb next to the Studio. Apple will need to tweak the price points somehow when the M2 Studio comes out... easiest thing is probably increase the base price and have it start at 64GB RAM minimum, that would remove overlap with the Mini that can only go to 32GB.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.