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.
M4 Mac Studio and Mac Pro Launching in Mid to Late 2025
- Thread starter MacRumors
- Start date
- Sort by reaction score
Chicken and egg situation.
Apple doesn’t update the chips and they’ll never be competitive.
I am deciding between a Macbook Air M3 en a Macbook Pro M3. I was leaning to the MBA because of portability. But now the MBP will likely get an M4, I’m starting to have doubts. It’ll probably be as bulky and as heavy as the MBP M3… I’m not a power user, but still tempted by power (and the screen).
If they‘d only upgrade the Air to M4 - but that won’t happen until spring, I guess.
If they‘d only upgrade the Air to M4 - but that won’t happen until spring, I guess.
This is true to a point, but it simply isn’t a big market segment any more.
They’re quite competitive elsewhere.
True, but with that attitude why bother releasing the products at all?
The M2 Max is barely half the reticle limit.
Because people who use mac pro SoC dont care about incremental updates like 10% increase
From m2 ultra to next year SoC will be night and day
Its the same thing with v12 engines..but with different purpose one to burn money the other to make money
It's not like the Mac Studio is terrible. It could be more powerful if given more thermal headroom and a SoC (or non-SoC approach) with higher peak performance, sure. But there's diminishing returns to that.
Max and Ultra chips will never come before the base chip. They would be awfully expensive to produce and they take longer to develop than base chips. Apple won’t be sitting on a ready design for months on end waiting for the Max and Ultra to finish development.
Target customers of Ultra care mostly about multi-core performance, where an older Ultra would still beat a newer base chip.
Makes no sense for the Ultra to come first.
Target customers of Ultra care mostly about multi-core performance, where an older Ultra would still beat a newer base chip.
Makes no sense for the Ultra to come first.
Dang, behind the Ultra Studio/Pro individuals, there’s gathering of people waiting for just-current-gen-Max Studio to get their daily chores done bit faster, bit easier, bit further. We won’t be served this time?
I understood the M3 Max was close to 600mm2 and the reticle limit is (I understand) about 848mm2. So maybe not exactly pushing the limit, but it's certainly not enough to create a monolithic M4 Ultra on a single die, unless there has been some very clever redesign!
Right now, ultra SKU are on a 2-year refresh cycle. M4 Ultra will be there until M6 series Ultra chip.
Of course Apple might revise that with changes on M5 packaging https://www.macrumors.com/2024/07/04/apple-m5-chips-advanced-packaging-tsmc/
But in the case of Mac Studio with the Max SoC, Apple seem to be exactly "sitting on a ready design for months". They could release an Mx Max Mac Studio at pretty much the same time as the Mx Max MacBook Pro.
I think the Mac Studio Max and Ultra have different markets and lifecycles. I bet they sell a lot more Mx Max Studios than the Ultra version, partially because of the price difference, but also because the Ultra is a lot more likely to be a business expense for professionals rather than a personal one for enthusiasts.
I would like to see the Max Studio marketed as a high-end consumer desktop - parallel to the high-end MBP - and for the Ultra Studio to have a longer refresh cycle more similar to the Mac Pro. Apple could create an Ultra version every alternate iteration. If we had had an M3 Max Studio, a lot of people would have bought it I think. Next year we can have M4 Max & Ultra, then in 2026 an M5 Max only, then M6 Max & Ultra in 2027.
Last edited:
Honestly, if you give me a Mac mini that can run 3 monitors (HDMI:4K@120+2xThunderbolt at 4K@60) you can keep your Studio... but everybody is betting on a shrinking Mac mini so I'm not optimistic :/ My M1 still runs fine, but I feel like I outgrew it...
Last edited:
I heard that they were and were not working on a Quantum Chip. We may or may not see that in the future or maybe not.
I'm hoping they don't bork the Mini Pro by compromising performance by shrinking it - either by removing useful ports like Ethernet and HMDI, or by reducing its thermal capacity by giving it a tiny heatsink and fan. If the 5-port USB rumor is true, I would hope for:
3 x TB 4 ports (rear)
2 x USB-C (20Gb/s?) (front)
1 x HDMI
1 x Ethernet (1GbE, upgradable to 10GbE)
1 x 3.5mm headphone
IEC power plug with internal PSU.
I'm OK with USB-A going as long as I still have 5 high-speed ports.
It's quite simple actually, the yields of the max and ultra are much lower then of the regular M series and even the Pro. Therefore it is not possible to upgrade the Studio and Mac Pro at the same time as the other Macs due to low or no stock for the M Max and M Ultra SOCS. What do you think what would happen in customer base if the Studio and Mac Pro got upgraded at the same time as the other Macs but the BTO options are greyed out because of no stock or if not greyed out a BTO would have a delivery time of 3 months or longer depending on the yields? A year later in my opinion is ludicrous though but a few months later then the rest would be acceptable. Probably the yields of the Max and Ultra are very disappointing.
I believe so, give or take.
M1 Max is made on 5nm (N5 TSMC) with 57B transistors measuring 432mm².
M2 Max is made on 5nm (N5P TSMC) but with 67B transistors. The logic density of the N5 and N5P process is the same. That should land it around 507mm². Or around 59% of the reticle limit.
Factor in N3B density reduction of 42% (density 1.7x) compared to N5/N5P for M3 Max and there is still room to grow. SRAM logic doesn't scale much though.
So the overall transistor count rose by around 62% (67B to 92B transistors) but then we need to factor in the density reduction from the N3B process so around 608mm² seems like a good guess or around 71% of the reticle limit 😅
But they could make a new larger die for the Mac Studio if they had any ambition but I am not currently limited by the CPU core count but by the GPU. I would much rather we stayed around the same CPU core count but massively increased the GPU core count.
Last edited:
That's a good point, and certainly true in the case of the Ultra.
I can also imagine Apple wanting to delay a Max Studio to allow for pent-up desire for the equivalent Max MBP 14" & 16" to be met, in order to maximize the sales of these more expensive machines. I would be OK with a Max Studio coming about 2-3 months after the launch of the equivalent MBP. That should be enough time for the initial rush of interest in the MBP to have died down to levels that they can assign a proportion of the SoC production to the Studio.
I doubt it’d be the same. Luke Miani made a graphic that showed the M4 iPad Pro is already higher than the M3 Pro.
I’m saying if the M4 Mac Pro and M4 Mac Studio are the same performance as each other like the M2s were.. it would be absurd..
I wonder whether Apple could release a new mini for the M4 base chip in a smaller guise as per the leaks, and then use the existing Mac mini form factor for the M4 pro … and call it a Studio Mini.
So it could have additional ports on front and maybe an SD slot.
This would fill the gap to allow the studio/pro for later release (or even a form factor refresh)
So it could have additional ports on front and maybe an SD slot.
This would fill the gap to allow the studio/pro for later release (or even a form factor refresh)