There was a time when we could read about actual rumors here. Now the "rumors" are that after M3 comes M4 and that it could be faster and support more memory.
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.
Apple's M4 Mac Studio: What We Know So Far
- Thread starter MacRumors
- Start date
- Sort by reaction score
While this would be very atypical, it's not totally impossible, and the lack of M3 studio/pro machines suggests it's much more likely than normal.
Normally, the Mx chips are bigger brothers to a particular aX chip, shipping some time after. The Ax ship in September, but Apple *could* ship some in June. They just don't have enough of them! They need to ramp production and store the chips so they can have enough to get even close to satisfying demand for the new iPhones.
So Apple's got the core designs done long before that, and if they have the bandwidth (ie, a big enough design team), they really could ship an Mx in June. Volume is a concern - they don't want to take too much silicon away from the iPhone - but volume for the Studio and Pro are so much lower than the iPhone's, it does seem manageable.
This would have a significant advantage for Apple, in that the high-end machines would actually be the most powerful by far, rather than barely keeping up with the newest laptop in half the benchmarks (and falling rather behind in single-core).
So, do I think that this is going to happen? Not so much. But I wouldn't bet against it.
This is nonsense twice over. The Pro and Max chips have the same bandwidth regardless of which chassis they're in. The Pro has 150GBps, and the Max 300 or 400 depending on which version of the chip you have.
As I've pointed out before for similar pieces, the title of this article is a lie. You know *nothing* so far. You have rumors, and by all means report on them. But call them what they are.
At least a couple of times recently you've done the right thing (not used the word "know"). It's a shame you're backsliding.
Hopefully that “M4" plate on top is of better quality than that FineWoven-Stuff. Ideally it also should have adjustable ambient led-lights. Also i am not sure if that plate on top eats too much space and the whole Device will not fit under the Screens.
Outlier, the most any Mac Pro before could support was 128GB...
Thunderbolt 5 would be nice, but if Apple is moving the Mn-series to a yearly cadence, they might get away with waiting to introduce TB5 with the M5-series...
And the (all-new) Mac Pro Cube...!
RE: M4 Extreme
While I would love to see a Mn Extreme, I recall an older rumor stating no Extreme until M5; maybe Apple wants to work out any kinks with the monolithic M4 Ultra first, then move to the Extreme in the M5-series, with a faster/wider interconnect than UltraFusion...?
No M3 Ultra because Apple is skipping it for the M4 Ultra...
Because Max Tech tells us so...!
Because he is bombastic...
Last edited:
Technically the only people who "know" anything about the M4 chip and a possible new Studio are working for Apple.
Mac Studio is an overpriced piece of crap. Apple introduced MacStudio and display to increase revenue margins on their side. Zero benefit for the consumers.
So what I'm reading from this article is we don't actually know anything?
So the title should probably have been: What we have HEARD so far
The worst ever clickbait channel about apple is Matt talks tech.
The guy knows absolutely nothing and constantly repeats his topics with the same content. Then he asks his viewers to write in the comments which Apple devices they are planning to buy. Of course he's not interested in that at all, it's all about earning more money.
This nutcase produces nothing but hot air without any basis.
At least Max tech also has interesting content.
Hope I'll be able to afford an M4 Ultra!
I'll be due for an upgdrade in 2025.
I'll be due for an upgdrade in 2025.
No way Apple is wasting the bulk of its “A18/M4/AI in the Apple ecosystem” hype narrative on products as low sales, low revenue as iPads or Macs.
Apple leads each new generation with iPhone as it outsells, out-earns, and gets more talk than all other Apple devices by far.
Before next iPhones drop, we only get products with SoCs based on last years high-end iPhone SoC.
Apple is not turning its product strategy and the Apple supply chain on its head this year just to sell a few M4 Macs before September 2024.
The bulk of Marc German’s predictions are based on nothing else but how predictable Apple’s product launch schedule has become:
iPhone always leads the next generation of SoC in september. So it’s a given that we’re not getting anything better than current gen SoCs in Macs and iPads until after the next iPhone has dropped.
It’s like clockwork, every single year.
So it’s really just about guessing which Macs and iPads sold well enough last time to warrant a refresh with the new SoC after hype for the latest iPhone has started ceasing.
iPads and Macs dropping before iPhone 2024 event will thus be no better than M3.
M4 is on 3nm compared to M2 at 5 nm and that shrinks die size. With a smaller die, it makes sense to do M4 Ultra as a dedicated chip (slightly larger than M2 Ultra, but not by much, as some components no longer need unnecessary doubling).
Based on this I predict Apple with add Ultra fusion interface to dedicated M4 Ultra and also offer M4 Extreme by combining two M4 Ultras.
My M4 core count predictions:
M4 - 4p 4e 12g
M4 pro - 6p 6e 24g
M4 max - 14p 4e 48g
M4 ultra - 28p 4e 96g
M4 extreme - 56p 8e 192g (2 x M4 ultra)
Based on this I predict Apple with add Ultra fusion interface to dedicated M4 Ultra and also offer M4 Extreme by combining two M4 Ultras.
My M4 core count predictions:
M4 - 4p 4e 12g
M4 pro - 6p 6e 24g
M4 max - 14p 4e 48g
M4 ultra - 28p 4e 96g
M4 extreme - 56p 8e 192g (2 x M4 ultra)
I couldn’t possibly surf the web with anything less than the M4 pro max ultra extreme (made of 8 x M4 extreme’s). Any news on when that will hit the Apple Store?
I love my MacStudio, and looking forward to whatever update they have in store for us. The form factor is nearly perfect so no need to change that.
I want to upgrade my 2015 iMac. I want a machine capable of all high end tasks like final cut pro, photoshop, blender etc. I was hoping mac studio would get updated soon then i could get it with studio display. I don't want to wait till 2025 for a mac studio with m4. Should I get something else? or wait?
Na. Not complaining about rumors, just statements of fact that imply more. what do we know so far? nothing. what rumors, speculations, we have heard? I get it MacRumors is a rumor site so that should be implied, but I think they are carrying it a bit too far and even the readers seem to forget it's all just rumors until released.
The Mac Pro has internal PCIe slots which provide a lot more I/O bandwidth than you can get with external Thunderbolt-to-PCIe enclosures. If you need lots of fast internal storage, or multiple specialist I/O, networking or AV cards that can take full advantage of the bandwidth of 8x or 16x slots you need this (and will presumably be prepared to pay through the nose for it). That's the only distinction that counts, and the only reason for the Mac Pro's continued existence. As I understand it, most of that PCIe bandwidth came from the surplus SSD interface on the M2 Ultra's second die.
If the - quite considerable - Thunderbolt 3/4 I/O of the Mac Studio Ultra is enough for you (and that's going to improve when TB5 rolls out), then there's no point in getting a Mac Pro-like "big box o'slots" just to get extra CPU and GPU cores and RAM capacity.
Part of the disappointment with the 2023 Mac Pro was down to the name (and cosmetic design), which invited comparison with the 2019 Mac Pro. I think it helps to mentally re-name things:
Mac Studio -> New Mac Pro (the trashcan, done mostly right)
2023 Mac Pro -> Mac Pro PCIe (for those who need lots of PCIe)
If you want something like the 2019 Mac Pro then Apple Silicon isn't the tool for the job - adding external, expandable RAM and 3rd party discrete GPU support undoes some of the key advantages of Apple Silicon. Using conventional third-party GPUs would mean that the Mac could never be better (or offer something significantly different) than AMD's (or NVIDIA's if hell freezes over) latest chipset.
If Apple thinks there is a niche for a specialist AI server system (as has been rumored) and wants to go head-to-head with Nvidia's Grace/Hopper, that's a whole new ballgame. I don't think the 2019 Mac Pro-style "general purpose personal workstation" has a rosy enough future to warrant developing a whole new die for: the market is being eaten by ever-more-powerful laptops and small-form-factor systems at one end, and pay-for-what-you-use cloud computing at the other.
Why does it need one? It's a box to hold the logic board, fans, PSU and sockets together. The only possible fault is that it maybe needs a plastic panel in the lid to improve WiFi/Bluetooth - I don't think that's gonna lead to a rush to upgrade. It's different with laptops where there are displays, speakers mics to upgrade and a pressure to make them as thin and light as the technology will bear - especially the low-end models which sell as fashion items to back-to-school kids.
More to the point, if Apple planned the M3 series as a stopgap annual upgrade for the main Mac cash cows - the MBA and MBP - and never planned to make a M3 Ultra there was no point in including the interconnect on the M3 Max.
Speculation. It's looking more likely that there simply won't be a M3 Ultra - could be wrong, but if the rumours of a rapid across-the-range M4 roll out within 12 months are true it's getting a bit late to release a M3 Ultra.
What is suggestive is that Apple changed the system with the M3 range - with M1/M2 the Pro, Max and Ultra were all permutations on the same die design - the "Pro" was 2/3 of a "Max" die (even called "chop" during development) and the Ultra was 2x Max dies. That's clearly not the case with the M3 Pro and Max having totally different dies, so really all bets are off for the M3 Ultra.
Meanwhile, although all the M3s are still better than their M2 counterparts but, relative to each other, the Pro has become slightly less powerful - and more focussed on power efficiency for laptops (which takes off some of its shine for desktop machines) - and the Max has become significantly more powerful all-round, stealing ground from the M2 Ultra.
So, really, the desktop range is ripe for a shake-up - anything is possible: the Studio could get dropped if the Mini can cope with a M3/M4 Max, the Studio Ultra could get dropped if the Studio Max is now good enough, the Mac Pro could very easily be left to languish as-is (just like the last two Mac Pros and the iMac Pro)... or may be the M3/M4 "Ultra" could turn out to be a "scalable" die like the M1/M2 Max with 1x, 2x and 4x (or Grace/Hopper-like CPU die+GPU/AI) combos showing up in future Studio, Mac Pro and Hypothetical AI server machines... its fun to speculate (disappointment is the most likely outcome).
