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.
Mac Pro Enthusiasts Raise Concerns Over Upgrade Limitations of Apple Silicon
- Thread starter MacRumors
- Start date
- Sort by reaction score
I get it. Essentially, change is hard. When the first Mac came out, there were likely many folks wondering why didn’t the Mac have any slots? Why couldn’t it be more like the Apple II? I mean, how can you even do midi with it? And no way to run the leading spreadsheet of the day? What’s the point of that?
There are people that wanted a Mac that was, essentially, an Apple II build. Apple DID make the IIGS, but we saw what ended up being the product with a future.
You mean like they brought back only 2 years later in the SE with the SE PDS? And a year after that with the II and 6xNuBus slots?
Yes, and the MacII came out for about $10,000 in today's dollars. The Mac II is a good example of the pricing for the Mac Pro.
I get it. Essentially, change is hard. When the first Mac came out, there were likely many folks wondering why didn’t the Mac have any slots?
And guess what: computers today still have "expansion slots" (aka USB / USB-C), and PCI slots in the case of desktop computers.
So, while the actual slots from before may be dead, the concept itself still exists.
Yeah, but still wouldn’t run AppleII software natively, so, still, it was not what those folks that wanted a new thing (Mac) in an old usage factor (AppleII).
Suppose instead of each of these daughterboards being a RaspberryPI4 Compute Module, each was an M1 or M2 SoC module? This is what some of us think Apple might be releasing.
but, they'd need a major rewrite in macOS to handle this kind of resource. Applications? Well, the industry wisdom is that one "killer app" would do it. Like VisiCalc for Apple II. Suppose that killer app was a biotech or AI software? Something where distributed computing is currently done in the cloud, but where privacy concerns might dictate otherwise. Apple does seem to be fixated on data privacy these days, which is good. Somebody has to be.
but, they'd need a major rewrite in macOS to handle this kind of resource. Applications? Well, the industry wisdom is that one "killer app" would do it. Like VisiCalc for Apple II. Suppose that killer app was a biotech or AI software? Something where distributed computing is currently done in the cloud, but where privacy concerns might dictate otherwise. Apple does seem to be fixated on data privacy these days, which is good. Somebody has to be.
I cannot imagine that there is a big enough market for Apple to even consider spending the engineering time to make a Blade Cluster high performance computing system work.
Could Apple do it, sure.
Does a market exist, yes.
Would it ever have a positive RoI, almost certainly not.
VisiCalc was a (relatively) popular app. There is no "popular" app that needs distributed processor blade clusters.
^ Apple needs to figure out how to make a device a decent bit more powerful than a Mac Studio Ultra with just laptop parts, so it certainly seems like a plausible approach.. certainly when the alternative was trying to connect 4 max chips together was likely super expensive, low yeild and abandoned as the rumours go.
Apple needs to do something, as with the advent of 13th gen Intel it's clear they've lost the performance crown. Yes, if one looks only in the laptop space it looks a whole lot better (much, much better), but in the desktop space, if you don't care about low power and just want a fast desktop, the M2 Max and the i7-13700K have a clear performance disparity, in favor of the i7. 14th Gen Intel is a few months away (and due for another good jump, 'tock' rather than 'tick', if I have that right).
nanoreview.net
(Jump to the 'Benchmarks' section; the rest is bot-written / is mostly for entertainment)
Intel Core i7 13700K vs Apple M2 Max: performance comparison
We compared Intel Core i7 13700K (3.4 GHz) against Apple M2 Max (3.5 GHz) in games and benchmarks. Find out which CPU has better performance.
nanoreview.net
Well, if we're talking about skewed benchmark, then we can say that this shows AS is king?Apple needs to do something, as with the advent of 13th gen Intel it's clear they've lost the performance crown. Yes, if one looks only in the laptop space it looks a whole lot better (much, much better), but in the desktop space, if you don't care about low power and just want a fast desktop, the M2 Max and the i7-13700K have a clear performance disparity, in favor of the i7. 14th Gen Intel is a few months away (and due for another good jump, 'tock' rather than 'tick', if I have that right).
(Jump to the 'Benchmarks' section; the rest is bot-written / is mostly for entertainment)Intel Core i7 13700K vs Apple M2 Max: performance comparison
We compared Intel Core i7 13700K (3.4 GHz) against Apple M2 Max (3.5 GHz) in games and benchmarks. Find out which CPU has better performance.
Why do you think I showed a skewed benchmark? I included *Geekbench*, for heaven's sake. The definition of a Mac friendly benchmark. It still got crushed.
Your link was written in 2022 but included (at the latest) only Intel specs from 2019. Not what I'd call an informative or fair article except as a focused look at Macintosh performance, rather than overall performance.
Luckily, the performance crown just doesn't matter that much.
Apple doesn't even care about the performance crown as much as the "performance per watt" crown.
We are crossing a point where for most people, the slowest part of a computer system is the user.
As long as Apple's stuff isn't too slow, performance isn't a problem.
Right now, Apple's GPUs are borderline on being too slow for some users, mind you.
Cinebench favours Intel AVX. GB scores are close, with Intel burning way more energy, but I guess that's a win?
But I guess Apple is doomed?
GB scores aren't close. Single core within 5%, multicore within 40% isn't close unless you only do spreadsheets / Safari / etc. type things (ie unless you only use / need SC). If you're doing MC things (video proc, etc.) that's a huge jump. And that's just a basic i7. An i9 is that much faster: https://nanoreview.net/en/cpu-compare/intel-core-i9-13900k-vs-apple-m2-max - SC by 15%, MC by 60%.
To acknowledge a speed difference isn't saying Apple is doomed. It's saying we like Macs and we'd like them even more if they'd be more competitive, in speed and price, with Intel. I can buy this CPU and stick it in (or upgrade) a fully working Hackintosh, with MacOS, for $800 with 32GB and a motherboard. Add a fan, stick it in the case/PSU/GPU/etc., and I'm gtg.
That's what I'd like Apple's modern Mx arch to give me.
Hope springs eternal for June 5 and an M3 Ultra that's a lot faster at a reasonable price.
Hope springs eternal for June 5 and an M3 Ultra that's a lot faster at a reasonable price.
I think it is safer to assume: Apple only needs The M3 Ulta to be a bit faster and sell at high prices.
Unless the delta in performance between Macs and PCs is large enough that Apple's profits are hurt, Apple is making the right call for Apple.
We can "hope" for magical performance at magically low prices, but Apple has no interest in that.
Apple will make "fast enough to sell" and "high enough margins to mitigate their massive spend on low volume chips".
... but hey, if you want "OK" performance, Apple has some great stuff. My base SKU M1 Pro MBP is definitively "OK".
I've just purchased a new M2 Max MBP. My old 2019 Intel MBP had its fans on constantly even doing web browsing.
I use a mid-level 16C Mac Pro for my daily work, 2D animation and sometimes 3D rendering with an AMD 6900XT that I've added.
It's fair to say the AMD card is 3x faster than the M2 Max, but I've been really surprised at how good the MacBook Pro is at literally everything else. I was able to take my work on the road last week…
The big noticeable difference is my Mac Pro is often drawing 250W just for the graphics card, whereas the laptop maxes out at 75W.
My point is, if Apple can fit all this power into a portable computer, the AS Mac Pro is going to be killer.
I use a mid-level 16C Mac Pro for my daily work, 2D animation and sometimes 3D rendering with an AMD 6900XT that I've added.
It's fair to say the AMD card is 3x faster than the M2 Max, but I've been really surprised at how good the MacBook Pro is at literally everything else. I was able to take my work on the road last week…
The big noticeable difference is my Mac Pro is often drawing 250W just for the graphics card, whereas the laptop maxes out at 75W.
My point is, if Apple can fit all this power into a portable computer, the AS Mac Pro is going to be killer.
I just honestly wish Apple would make a path towards compatibility with industry standard GPUs. I'm glad to hear that some AAA games are coming to the Mac, but it feels like they could drop support for this in a couple of years and we'd be in the same spot as always with Macs and 3D in general.
I'm sure it's been said here, but the only reason we had to switch out entire studio to PC is because Macs just don't have good enough GPU power to compete. Everything in VFX is so GPU dependant now and Macs just can't render as fast (their fastest is still about half of what Nvidia can do). If they just pushed the GPUs then I feel the Pro market would explode... everything else is crazy fast for throughput but I just want faster GPUs
The ASi Mac Pro could just been a powered PCIe enclosure that supports GPU's plugged into a Mac Studio and it would of been miles better than the $3000 upgrade for an empty aluminum box of air, and the privilege of running some audio pci cards. I'm not sure they are even gonna sell 10,000 of these things.
Apple seems happy to punish pro users and make them wait years for a igpu that isn't 3 yrs behind AMD and Nvidia, at least when it comes to 3d, but that would mean Apple admitted they aren't as good as someone else at something. Why offer a band-aid when you can just pretend the issue isn't there.
Apple seems happy to punish pro users and make them wait years for a igpu that isn't 3 yrs behind AMD and Nvidia, at least when it comes to 3d, but that would mean Apple admitted they aren't as good as someone else at something. Why offer a band-aid when you can just pretend the issue isn't there.
This Mac Pro was almost exactly as I expected (if a bit more expensive and maybe a bit PCIe bandwidth starved). It'll have a market larger than 10s of thousands, but definitely not larger than 100s of thousands. Maybe that is enough for Apple, or maybe people inside Apple will succeed in getting it killed off.
There is a difference between what Apple is intending to do and what Apple is accomplishing.
I'm very confident that Apple was planning on an M* Quadra to be ready by now and having another:
- 2x more GPU performance would have been "enough" for most pro user needs.
- 2x more memory would have been "enough" for most pro user needs
- 2x more CPU and Neural Engines would have been great too...
Apple almost certainly had a plan that scaled reasonably well for Pros.
None of that changes the reality that Apple failed to ship an M* Quadra.
Hopefully in the next 2-3 years they will. If not, the Mac Pro will hobble along in it's weak state.
The biggest fear I have is that Apple will interpret the weak reception this half-donkey'd Mac Pro will get as a reason to not bother getting the real Mac Pro working :/
I agree, if the Asi MP could of 2x'd all the stats of the Ultra it would of been a celebration by the pro community. I'd like to think Apple tried to build the "extreme" chip.. and failed, plan B is still a 7+ months away. And this M2 Ultra MP, is just picking up the pieces and getting the eye of sauron pressure to "finish the transition" off them.
It does however feel like an almost insulting product though. I'm glad I'm not an audio guy, and have to seriously consider this thing. I do worry the abysmal sales of this I predict will be used against the Pro line. I just always wonder who the F the pro's apple is consulting with are?
It will be very interesting 15-ish months from now.. If the MP is gonna be updated in lock-step with the Mac Studio? This MP feels like $8K in E-waste ready to happen. With no upgradable parts, and there bing almost no difference from it compared to a lot of the ASi products that are updated yearly its gonna age very quickly. Again.. an external enclosure you can move from system to system sure would of been a nicer play.
It does however feel like an almost insulting product though. I'm glad I'm not an audio guy, and have to seriously consider this thing. I do worry the abysmal sales of this I predict will be used against the Pro line. I just always wonder who the F the pro's apple is consulting with are?
It will be very interesting 15-ish months from now.. If the MP is gonna be updated in lock-step with the Mac Studio? This MP feels like $8K in E-waste ready to happen. With no upgradable parts, and there bing almost no difference from it compared to a lot of the ASi products that are updated yearly its gonna age very quickly. Again.. an external enclosure you can move from system to system sure would of been a nicer play.