Multi display video wall, security camera monitoring station, really any sort of monitoring / dashboard station. I use 3 for my main workflow, but could happily drop in some others to put reference material on, etc. If you have a Wacom Cintiq, that's one of your display allotment.
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.
MP All Models Mac Pro X,Y…WHAT IF…
- Thread starter ZombiePhysicist
- Start date
- Sort by reaction score
Multi display video wall, security camera monitoring station, really any sort of monitoring / dashboard station. I use 3 for my main workflow, but could happily drop in some others to put reference material on, etc. If you have a Wacom Cintiq, that's one of your display allotment.
Displaylink adapters / docks that use them would be OK for stuff like reference material, possibly a Cintiq too, though obviously not for interactive 3D.
Sure you can always argue alternatives, but what's the point to anything in life if you can't do things the specific way you want to do them?
A cintiq is abslutely going to need huge bandwidth - it's a 10 bit 4k display, AND it needs a 120hz refresh.
But again, whether you need it or not, that's the fundamental Max/Ultra difference - more Thunderbolt, and more displays. Some folks need that, just like some folks needed the Mac Pro when all the hoi poloi in the Mac-Media-douche-o-sphere were saying "an iMac or a Macbook Pro are good enough for everyone"
Well if you do a lot of 3d work, or AI, where the GPUs tend to matter more, that could be a significant boost and offer a decent life span. True, the pure CPU bound stuff gets dated. But even there, the 28core held it's own for a good while more than the 3 months of the M2Ultra to M3max.
Yeah, abslutely, my point on this has always been that most of the difference, and most of the stuff that gets eclipsed generationally is doing the same things a bit faster - a better GPU in an older machine is definitely a benefit if you're doing realtime stuff that's outside the capability of a newer lower-end GPU, and I'd certainly categorise that with More Displays & More Thunderbolt as the purpose of an Ultra.
But for most things, when the difference is speed of a process that is not realtime-critical, its hardware flexibility that defines the Max / Ultra difference.
The M(N+1) Max equals M(N) Ultra phenomenon will be relatively rare. Yes, it's true of M2 to M3, but the M3 Max is a very unusual chip. How often are you going to manage 50% more P-cores plus a die shrink (plus a modest clock speed boost) in one generation? It's the kind of jump we used to see decades ago, not the kind we see now. Even in the M2 to M3 transition, it was only the Max that did that - both the base and Pro models saw more typical improvements.The M3 Ultra is likely to be another beast, because an Ultra is typically a double Max, so it should get the advantages of the Max
Assuming Apple continues to have annual improvements that are considerably better than Intel (because Apple Silicon has less legacy architecture), but not crazy 50% year over year jumps, I'd imagine that it will average M(N+2) or even M(N+3)Max=M(N) Ultra, perhaps with the occasional extra-large jump.
Assuming Apple continues to have annual improvements that are considerably better than Intel (because Apple Silicon has less legacy architecture), but not crazy 50% year over year jumps, I'd imagine that it will average M(N+2) or even M(N+3)Max=M(N) Ultra, perhaps with the occasional extra-large jump.
Apple has, astonishingly, maintained the old Moore's Law cadence in the top MacBook Pro chip from i9 to M3 Max. They went:
Core i9 - M1 Max (doubling performance) - M2 Max (15% improvement) - M3 Max (50%+ improvement). An M3 Max has raw CPU performance roughly 350% of the last Core i9 MBP (while cutting power consumption in half)...
If the M3 Ultra is what we're expecting, they'll have maintained a similar cadence in mid-level pro desktops, starting from a late generation i9 iMac. It would take a miracle (in the form of an M(N) Extreme) to put the Mac Pro on that cadence.
They had a complete architecture shift in there, of course, but 350% in almost exactly four years has been unheard of since the late 1990s or early 2000s.
Can they keep it up? Almost certainly not - because nobody has been able to in twenty years. GPU performance has come closer than CPU performance. At best, Apple will manage to improve the M-series chips at a GPU-like rate of 15-20% annually, while Intel and AMD are stuck at 5-10% (with occasional partial architecture shifts like Bulldozer to Zen yielding better results).
Even if they "only" do that, they'll keep pulling away. Intel and AMD will have to abandon x86 to come close (and the legacy Windows house of cards will fall). If the new Qualcomm chips are where they're saying they are on the power/efficiency curve, and they scale like Apple Silicon, the x86 (and derivatives) architecture may be in trouble, or it may have such a legacy that it stays around, despite being increasingly slower than the competition. One HUGE advantage that Apple has is that their OS seems to be relatively architecture-agnostic, while Windows is tightly bound to x86.
MacOS has gone from 680x0 (pre OS X, but also NeXTStep, the foundation of OS X) - PowerPC (both classic and OS X) - x86 - Apple Silicon in 40 years. Windows (and DOS before it) has fundamentally been x86-bound that whole time. If Apple needs another architecture shift in 10 years, they can do it. I'm not at all sure Windows can (Windows on ARM notwithstanding)...
Core i9 - M1 Max (doubling performance) - M2 Max (15% improvement) - M3 Max (50%+ improvement). An M3 Max has raw CPU performance roughly 350% of the last Core i9 MBP (while cutting power consumption in half)...
If the M3 Ultra is what we're expecting, they'll have maintained a similar cadence in mid-level pro desktops, starting from a late generation i9 iMac. It would take a miracle (in the form of an M(N) Extreme) to put the Mac Pro on that cadence.
They had a complete architecture shift in there, of course, but 350% in almost exactly four years has been unheard of since the late 1990s or early 2000s.
Can they keep it up? Almost certainly not - because nobody has been able to in twenty years. GPU performance has come closer than CPU performance. At best, Apple will manage to improve the M-series chips at a GPU-like rate of 15-20% annually, while Intel and AMD are stuck at 5-10% (with occasional partial architecture shifts like Bulldozer to Zen yielding better results).
Even if they "only" do that, they'll keep pulling away. Intel and AMD will have to abandon x86 to come close (and the legacy Windows house of cards will fall). If the new Qualcomm chips are where they're saying they are on the power/efficiency curve, and they scale like Apple Silicon, the x86 (and derivatives) architecture may be in trouble, or it may have such a legacy that it stays around, despite being increasingly slower than the competition. One HUGE advantage that Apple has is that their OS seems to be relatively architecture-agnostic, while Windows is tightly bound to x86.
MacOS has gone from 680x0 (pre OS X, but also NeXTStep, the foundation of OS X) - PowerPC (both classic and OS X) - x86 - Apple Silicon in 40 years. Windows (and DOS before it) has fundamentally been x86-bound that whole time. If Apple needs another architecture shift in 10 years, they can do it. I'm not at all sure Windows can (Windows on ARM notwithstanding)...
To be fair, Windows from NT onwards, which is the ancestor of the current Windows has run on X86/X64, MIPS, Alpha, IA32, IA64, PowerPC and ARM.
The biggest problem the Mac has, is that macOS is melting into a puddle of garbage, as more and more of it, and its bundled apps, are replaced by parts of iOS (I'm going to bet that's the root of the PCI SSDs randomly disappearing at boot).
The processor is the least of the Mac's problems - we're already past peak utility for maOS, and into the phase where the Mac can do less of the things an iPad can't do with each release, as it homogenises into a UI skin for iOS.
It really seems like all Mac users are in for the special treat Mac Pro users have been experiencing. An almost spiteful resentment of their existence.
The real problem is Apple has waaaaay too many bozos/fiefdoms and needs a massive culling like Steve Jobs did on several occasions, resulting in huge runs of creativity and productivity. Apple is now becoming the new aol/yahoo. A company that buys other companies but is composed of so many do nothing bozos, it’s a company where other companies/technologies go to die. It’s now big tobacco, with the iphone, making money despite itself and its idiot management.
Yes, but (outside of the server space), 95% of installs have always been X86/X64. Most software, especially things that need performance, has never offered any other choice. Yes, you have all those choices for Windows itself (at different times, often not with consistent releases and support), but how many have a supported, updated native release of Photoshop? Resolve? ArcGIS? AutoCAD? Some of them haven't even had native MS Office!
Windows is, and always has been an X86/X64 system (with a little more diversity on the server side). It is currently trying to assemble a viable ARM user base in addition. It doesn't help that BY FAR the fastest widely available Windows on ARM systems are made by Apple. Yes, someone's going to point to an ARM server or maybe a custom box made by a tiny company and say "pshaw, that's faster than any MacBook Pro or upcoming Mac Studio", but there aren't even 10,000 of them sitting on people's desktops or any at all in people's briefcases.
What might have been.
Imagine MacOS on one of these.
No need to reinvent to reinvent the wheel for best in class performance.
Dell Precision 7875 Review: 96-Core Threadripper Pro and Dual RTX GPU Powerhouse
The Dell Precision 7875 is a highly impressive high-end workstation with an AMD Ryzen Threadripper Pro 7000WX series CPU.www.storagereview.com
I love me some Threadripper but Dell's implementation of them in their product has been awful.
What's wrong with it? Only skimmed the article, but SR said it was one of their favourite workstations - no workload too large, and quiet while doing it.
The job of a CPU is to run software; if Windows software is all x86, which it is, then the CPU needs to use that too.
It would be interesting to know the extent to which the ARM ISA contributes to AS efficiency, and how much is the result of process node and features like unified memory and integration into an SoC. If a 5nm x86 SoC were 20% less efficient than AS, but compatible with all existing Windows software, that would be a trade off many would accept.
For a desktop, I wouldn't want an SoC design even though it offers better energy efficiency - there's too many inherent downsides.
They are pretty restrictive with power delivery to the CPU due to an undersized cooling system. The benchmarks they get on the 96 core is lower than what you’ll get from a boutique or DIY 64 core even at stock speeds.
They also have compatibility issues with any NVMe drives that are not OEM. They won’t even boot with a u.2 or u.3 drive on the bus.
They are decent looking and well behaved in lower 24-32 core config, if you don’t care about the nvme issues. At the higher end you’re better off with BOXX, Puget, or DiY.
The Dells, HPs and Lenovos of the world have their own versions of the walled garden that they and their corporate, government , Hollywood and institutional customers demand for various certifications and procurement rules. They probably sell more workstations in an hour than Apple sells Mac Pros in a year.
As you say, for creatives and small studios the BOXX, Puget and DIY companies give them much more flexibility for creating affordable and custom solutions that use standard, nonproprietary components.
Last edited:
Luke Miani has just posted a video about his Mac Pro 2023 model. Which he has owned for 6 months.
I have been waiting for him to do a video on it, for months 🤣
I have been waiting for him to do a video on it, for months 🤣
Could he be more narcissistic? Seriously, because he doesn’t need more than 8tb of Ssd no one needs it? The drives he used in there are near idiotic. 2 12tb spinners when 24tb models are out? Super crap pci riser cards when you can get 64tb 12gb/sec ssds brainlessly from OWC?
His use case is a joke, which is fine, but that he can’t see beyond his own joke use case, is not.
His suggestion for a neutered 1-size fits all 2 slot box is repellant to me. His “think”ing is dismissed since he can’t fathom anything outside his selfish tiny self centered world view.
Last edited:
That might be the case. But I fail to find any other video talking about the latest Mac Pro, except for the early reviews.
And in light of the Mac Pro 2019 NVME Raid drives failing to mount thread, using Sonoma. I thought it might be of some value.
And in light of the Mac Pro 2019 NVME Raid drives failing to mount thread, using Sonoma. I thought it might be of some value.
Please don’t take my criticism of his analysis to be criticism of your sharing it here. It is appreciated. And hearing others’ opinions, even when we/I disagree, is very much the point of the forums.
But based on previous history isn’t he a kind of click bait / borderline troll?
Why did he need all those maxed out 5,1s back in the day??
What’s the need to even watch his channel? I’m not. His load of hot air is completely irrelevant to me and use cases I have for my computers.
Yes, I know he has to make money somehow…
I haven't seen Luke's video and I think I'll skip it, but from recent experiences on Mac, I think the AS Mac Pro concept is sound.
I don't base this statement on how I would like things to be, or unrealistic wishful thinking, but rather trying to plan ahead based on the direction of Apple and macOS—but also the general development of Mac software.
The Mac Studio makes sense. It's the most compact, and the cheapest, solution to get max AS power without buying a display that you need to scrap or resell when you want an AS upgrade.
In terms of Mac compute power, the Mac Studio is as far as it goes and it will most likely remain that way. The writing has been on the wall for a while.
The Mac Pro then, is nothing more than a beautiful and Maccy way to dress up and internalise various Mac Studio expansion solutions. These solutions will likely remain limited to storage and specialty PCI cards for video, audio, or networking. Or other niche stuff.
This new version of the "Mac Pro concept" is different from what it was. I'm still fine with the name Mac Pro. I guess we'll see how it fares in various markets. It would be good if the additional cost above a same-for-same Mac Studio could be kept low, as it's effectively a 'case plus expansion possibility' cost.
I like the current design. Skip the MPX modules then, but provide a couple of empty nvme slots and bays for spinning drives (I still like to have internal "unlimited" space for things that I temporarily offload) in addition to a few PCI slots.
Would still love that stage 2 Afterburner card, but this time loaded with hardware raytracing capability. But as I said: no wishful thinking for me.
I don't base this statement on how I would like things to be, or unrealistic wishful thinking, but rather trying to plan ahead based on the direction of Apple and macOS—but also the general development of Mac software.
The Mac Studio makes sense. It's the most compact, and the cheapest, solution to get max AS power without buying a display that you need to scrap or resell when you want an AS upgrade.
In terms of Mac compute power, the Mac Studio is as far as it goes and it will most likely remain that way. The writing has been on the wall for a while.
The Mac Pro then, is nothing more than a beautiful and Maccy way to dress up and internalise various Mac Studio expansion solutions. These solutions will likely remain limited to storage and specialty PCI cards for video, audio, or networking. Or other niche stuff.
This new version of the "Mac Pro concept" is different from what it was. I'm still fine with the name Mac Pro. I guess we'll see how it fares in various markets. It would be good if the additional cost above a same-for-same Mac Studio could be kept low, as it's effectively a 'case plus expansion possibility' cost.
I like the current design. Skip the MPX modules then, but provide a couple of empty nvme slots and bays for spinning drives (I still like to have internal "unlimited" space for things that I temporarily offload) in addition to a few PCI slots.
Would still love that stage 2 Afterburner card, but this time loaded with hardware raytracing capability. But as I said: no wishful thinking for me.
I’m afraid that the reason there aren’t many videos talking about the latest Mac Pros is because very few were sold. The silence is deafening, sad and entirely predictable. The Apple Silicon Mac Pro was a phoned in, transition finale that will sell poorly and allow Apple to abandon it claiming there is no market for “expandable” HEDT computers. To quote Steve, no more “trucks” from Apple.
Sensible enough - none of us have any control over this really.
Sure, there will be some market for this. But it's becoming a niche of a niche of a niche.
Yet Apple deliberately priced it high - an additional $1000 or so over the 7,1. Presumably, they assumed that given the existence of the Studio Ultra, it would sell in even smaller numbers than before, and increased the price to justify its production. And if it fails to sell at that higher price, it'll get canned.
What do you like about the design? The 7,1 chassis / cooling? Given the ditching of MPX modules, and the fact it has neither M2 slots nor HDD bays (without adding them yourself), there doesn't seem to be much else to like.
I like the overall finish and build quality. I'm impressed by it when I see it, and I'm impressed with it on the rare occasions when I open it.
IF Apple now decides to ditch MPX and run it as AS+Expansion, I think they should redesign the inside to allow for plug-in storage. That's what I say above: I mean the user shouldn't need extra cards for nvme or caddies for HDDs. It should be part of the new design. The Mac Pro will still need a couple of PCI slots, but it might still be able to make it a bit smaller than it is now.
And I don't think it would be that niche to offer an M3 Ultra that a user can beef up substantially with off-the-shelf nvmes as time goes by. I wouldn't be surprised that quite a few users wouldn't blink to pay a $1500-2000 premium over the Mac Studio for this and a couple of open PCI slots.