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.
New Mac Pro and an 'Extreme' Chip: Here's What the Latest Rumors Say
- Thread starter MacRumors
- Start date
- Sort by reaction score
Someone with more chip knowledge… why can’t they just make Ultra/Extreme chips with the same scaling logic as the Pro/Max? Like, take the Max and add even more GPU cores, without doubling the entire chip. Is it a limitation on the physical chip size at that point?
- When is a new model expected? The second half of 2025, likely October to December
Pre-Orders begin 12/31. Shipping begins 2/1/26.
Well, except that, maybe unlike Intel, this reluctance to update the high-end Apple Silicon Macs isn't hurting Apple in any significant way:
MacBooks Pro and Air are by far Apple's best selling and most profitable Macs and thus the ones that get the most frequent updates.
I don't even need to pull out any graphs of the literal sales numbers because the proof lies in the average days between product updates for each Mac, respectively:
- MacBook Pro: 384
- MacBook Air: 411
- Mac Studio: 454
- iMac: 551
- Mac mini: 732
- Mac Pro: 938
But if you've been following Mac product releases over the last 10-20 years, and have observed what Macs are most commonly spotted "in the wild", it's quite easy to see that there's nothing personal or abstract about why some Macs seem to get new releases almost every single year, while others don't see any new releases for years and almost appear to be abandoned by Apple.
-Apple makes Macs for profit. And Macs Pro, whether due to high manufacturing costs or low sales, simply aren't generating enough profit to warrant frequent product releases.
I wouldn't be surprised if we never see a new Mac Pro again and Apple instead starts offering a(n even) higher-end Mac Studio with an "Extreme" chip.
The Mac Pro is the flagship Mac for SERIOUS creative professionals, scientists and world changing applications
It is due for a release with architecture for multi-slot M5 Ultra / Xtreme expandability. Multi RAM upgrades up to 8TB of internal RAM. Up to 80TB of SSD, and hidden low profile wheels
Plus an optional internal bay to keep your coffee hot
It is due for a release with architecture for multi-slot M5 Ultra / Xtreme expandability. Multi RAM upgrades up to 8TB of internal RAM. Up to 80TB of SSD, and hidden low profile wheels
Plus an optional internal bay to keep your coffee hot
Last edited:
film, audio, broadcast television studios, film sets, etc.
I run a creative studio, and for 90% of the work, a M1 MBP does the job. In fact an air probably will be fine for 70% of the work..... it's only that 10% where the real power is needed.
I have always supported the idea of mobile lightweight first, with remote into dedicated heavy task desktops when required. It works amazingly well.
An extreme chip mac pro is great for this way of working although we use PC's for these tasks now as I no longer trust Apple to deliver. We will be upgrading to 5090's in the PC's when they come out and retire the 4090's that have worked very hard and deserve a rest
A far far easier thing to do than rely on Apple in this area.
Computers with backplanes have now for several years not been of relevance to almost all consumers.
Only those with obscure interfaces (such as data acquisition), or who wish to have highly specialized embedded processors, need a device like the Mac Pro.
Which means not many people, and thus the slow cadence of updates.
Yet as this keypost is not a new story but simply something to fill up space, let me speculate:
1) long term reliance on PCIE is pretty much assured, as Apple learned its lesson with NuBus.
2) the chip limitations for CPUs is so difficult going forward that Apple is exploiting its current advantage now, to roll out new models quickly, to take as big a market share as it will ever have in the hopes that in years to come when changes are much slower that the Mac can remain in the marketplace.
Accepting those two things: Apple is more likely than not to give the Mac Pro one more round of life.
Only those with obscure interfaces (such as data acquisition), or who wish to have highly specialized embedded processors, need a device like the Mac Pro.
Which means not many people, and thus the slow cadence of updates.
Yet as this keypost is not a new story but simply something to fill up space, let me speculate:
1) long term reliance on PCIE is pretty much assured, as Apple learned its lesson with NuBus.
2) the chip limitations for CPUs is so difficult going forward that Apple is exploiting its current advantage now, to roll out new models quickly, to take as big a market share as it will ever have in the hopes that in years to come when changes are much slower that the Mac can remain in the marketplace.
Accepting those two things: Apple is more likely than not to give the Mac Pro one more round of life.
Because the chip size is too big. Max chip is already as big as RTX 4090 which makes it quite expensive and yet difficult to mass produce.
Also, unlike Intel and AMD, only Apple uses Apple Silicon and therefore they cant make various chips. Adding more GPU cores means you need a whole new chip since it's SoC.
They can and more than likely will end up doing just that…
M4 Ultra 20|24 CPU, 60|76 GPU
M4 Extreme 24|28 CPU, 96|128 GPU
This could be an SoC design where the GPU is not part of it but is connected over the Ultra Fusion interconnect. The main SoC (through binning) could offer 20|24|28 CPU cores, and then there would be two GPU variants; an Ultra GPU 60|76 and an Extreme GPU 96|128.
Something I mentioned like 4 years ago. Multiple slots for multiple cpu/gpu packages.
If I was a Billionaire I would buy one and MAX it to the extreme. Then just use it for email. Thats it.
Millions of folks have already figured out out how to escape his abusive upgrade prices.
Yeah, but the research and development for multi-die packaging would have to make financial sense for the MacBook Air and MacBook Pro FIRST before they’d spend any money on it. And, for Apple, that’s not a situation that’s likely to exist.
Air & Pro do not need high-end multi-die packaging solutions, SoCs are fine for those, almost like they were developed to use SoCs in the first place...
The high-end desktop/workstation/server markets is where multi-die packaging can shine; and Apple using their own silicon in their Apple Intelligence servers is where the R&D can be done to also allow these high-end packages into the Mac Studio/Mac Pro line-up of desktop/workstation products...
I await the all-new Mac Pro Cube Personal Workstation, powered by the Mn Extreme multi-die packaged "chip"... ;^p
The high-end desktop/workstation/server markets is where multi-die packaging can shine; and Apple using their own silicon in their Apple Intelligence servers is where the R&D can be done to also allow these high-end packages into the Mac Studio/Mac Pro line-up of desktop/workstation products...
I await the all-new Mac Pro Cube Personal Workstation, powered by the Mn Extreme multi-die packaged "chip"... ;^p
I thougth the Extreme could have been where you are allowed to add multiple Ultras on a board or something to add more power and ram.
This… I don’t understand why a “green” company is content issuing one-off hardware.
Even if I had the money to purchase a Mac Pro, just knowing it would quickly fall behind in ability, without any upgrade path, would ruin it for me.
Let users bring it into the Apple Store for an overhaul, at a reasonable discount.
It worked for Gillette!
Case = Handle
Board = Blade cartridge
People that don't use or need a MacPro will never understand. Being able to put extra drives and PCI cards INSIDE your tower is something we will happily pay extra for.