It didn’t take 2 plus years to just throw in an m2 Ultra and remove MPX.
DANG THIS IS SAD.
if Apple spent a fraction of the effort they put into the stupid headset on the Mac Pro, we’d have a world beating computer to rule them all. Instead we get an excuse to settle for the Mac Studio.
I think we will look back at this article as a very sad misprint. I can't find where video cards will not be supported.
Finally, Mac Pro brings PCI expansion to Apple silicon. It features six open expansion slots that support gen 4, which is two times faster than before. So users can customize Mac Pro with essential cards, including audio and video I/O, networking, and storage.
Right there in the keynote, Apple says the new Mac Pro will support video card for input and output.
Listed Tech specs for the Mac Pro.
Expansion
Six full-length PCI Express gen 4 slots Two x16 slots
Four x8 slots
One half-length x4 PCI Express gen 3 slot with Apple I/O card installed
300W auxiliary power available:
Two 6-pin connectors delivering 75W of power each One 8-pin connector delivering 150W of power
So you have two x16 slots that can support video cards. Granted the AUX power is only 300W total.
Again the same Aux power cables that are used on the 2019 Mac Pro are also compatible with the 2023 Mac Pro.
The Belkin AUX Power Cable Kit for Mac Pro provides a set of common AUX power cables for graphics cards and other AUX-powered PCI Express devices that enable connection to the 6‑pin and 8-pin AUX power receptacles in Mac Pro.
Enables PCI Express devices requiring AUX power to be connected inside Mac Pro
Provides cables needed to power typical graphics cards and AUX power PCI Express cards
The articles premise is most of the target market is missed by the new Mac Pro. Based upon what? Even without video cards, the expansion is highly wanted and useful. Just having internal storage is highly useful, because it is much faster than external storage.
Looks like a mistake by Primate Labs. That score is more than 2x the score of the M2 Max. Unless there are a lot of Compute sub benches that are memory bottlenecked which the M2 Ultra overcomes, hard to believe that score.
Geekbench compute scores vary a lot so can be quite unreliable but there are also two different models each of Max and Ultra and they don't say which they are testing, they describe them by CPU cores.
There are also different power levels in the OS, these two Max pages have different clock speeds. They also have different cooling capability. A Mac Studio will cool an M2 Max better than a Macbook Pro 14".
It looks like they have improved the Ultra GPU scaling though. We'll see with the Blender tests:
M1 Max (32C) is on page 5 with 1036. M1 Ultra (64C) is on page 4 with 1831.
M2 Max (38C) is already above M1 Ultra on page 3 with 1917. If M2 Ultra (76C) can scale to near 2x, that will bring it to the top of page 2. This is 1/3 from the top overall (and those others have hardware RT).
An M2 Ultra 76-core showed up on page 2 with 3412, 1.77x the M2 Max. A 4090 is about 3.8x, 4080 is 2.8x, 3090 is 1.8x.
To get near to the top, they'd need a 50% boost from M3 to make a 114-core Ultra and then add a second GPU chip that is equivalent to another 114-core Ultra. Or maybe 50% boost + hardware RT would be enough.
Interesting video on YouTube from Daring Fireball that I always like to watch the week of WWDC is "The Talk Show" and they talked about Mac Pro and Mac Studio with John Ternus and Joz. Yeah some the responses are kinda skirting the question but they do give good information in there as well and some reasons behind their decisions. Judging by their responses, it definitely sounds like both are here for the long run and they serve 2 different purposes.
It's a 2hr show but I always enjoy watching it. They always give some good information on the topics asked. Yeah there are certain things they can't talk about so sometimes it sounds like their answers are BS but that goes with any company. No company is gonna absolutely spill their guts on every single decision made.
Interesting video on YouTube from Daring Fireball that I always like to watch the week of WWDC is "The Talk Show" and they talked about Mac Pro and Mac Studio with John Ternus and Joz. Yeah some the responses are kinda skirting the question but they do give good information in there as well and some reasons behind their decisions. Judging by their responses, it definitely sounds like both are here for the long run and they serve 2 different purposes.
It's a 2hr show but I always enjoy watching it. They always give some good information on the topics asked. Yeah there are certain things they can't talk about so sometimes it sounds like their answers are BS but that goes with any company. No company is gonna absolutely spill their guts on every single decision made.
Here's the quote from Ternus on the question of PCIe GPUs (starts at 24:14): "Fundamentally, we’ve built our architecture around this shared-memory model and that optimization, and so it’s not entirely clear to me how you’d bring in another GPU and do so in a way that is optimized for our systems. It hasn’t been a direction that we wanted to pursue."
The key clause there is "do so in a way that is optimized for our systems." Could Apple itself do that, with Apple silicon via PCIe? Earlier in the interview, Ternus points out that, because of the Unified Memory architecture, the amount of memory available to the GPU in Apple silicon is orders-of-magnitude larger than in any other GPU. So maybe this hypothetical Apple silicon PCIe GPU extension for the Mac Pro doesn't need to have its own memory, it draws on the system's Unified Memory. So then it's an optimization problem, in software. They talk about how they've been making big improvements in terms of optimization, this would fit into that effort.
Users that bought the 2019 Mac Pro as soon as it came out or shortly afterwards made a good investment. There is probably going to be continued demand for that product.
I think we will look back at this article as a very sad misprint. I can't find where video cards will not be supported.
Finally, Mac Pro brings PCI expansion to Apple silicon. It features six open expansion slots that support gen 4, which is two times faster than before. So users can customize Mac Pro with essential cards, including audio and video I/O, networking, and storage.
Right there in the keynote, Apple says the new Mac Pro will support video card for input and output.
Listed Tech specs for the Mac Pro.
Expansion
Six full-length PCI Express gen 4 slots Two x16 slots
Four x8 slots
One half-length x4 PCI Express gen 3 slot with Apple I/O card installed
300W auxiliary power available:
Two 6-pin connectors delivering 75W of power each One 8-pin connector delivering 150W of power
So you have two x16 slots that can support video cards. Granted the AUX power is only 300W total.
Again the same Aux power cables that are used on the 2019 Mac Pro are also compatible with the 2023 Mac Pro.
The Belkin AUX Power Cable Kit for Mac Pro provides a set of common AUX power cables for graphics cards and other AUX-powered PCI Express devices that enable connection to the 6‑pin and 8-pin AUX power receptacles in Mac Pro.
Enables PCI Express devices requiring AUX power to be connected inside Mac Pro
Provides cables needed to power typical graphics cards and AUX power PCI Express cards
The articles premise is most of the target market is missed by the new Mac Pro. Based upon what? Even without video cards, the expansion is highly wanted and useful. Just having internal storage is highly useful, because it is much faster than external storage.
They addressed no video cards several times, most recently in Gruber’s talk show.
The reasons why for this are complex. It's not just about drivers -- Apple has decided that it didn't need a way for the Apple Silicon processor to talk to an external graphics card, at all, under any circumstances.
Very well said, Mike!
But, I'm curious about this statement. I was under the impression it was more a driver type situation, and maybe eGPUs could even make a come-back.
I'm not even in the market for a Mac Pro, but I look to it to see Apple's potential and direction. In that regard, I really hope this is a stop-gap to get the transition done, because it is very disappointing.
For myself, I'm left with a bit of dilemma. I want GPU power for 3D stuff (and, sure, I'd love gaming). But, unless the M3 is a real leap (which it might be), I'll be left with buying a fairly expensive Studio (kind of my dream machine), that barely competes with with a $1500 gaming PC, and might be questionable about usability, outside of question of being competitive.
Or, I could buy a base-level mini, which will probably be just fine for most of my computing needs, and then get that entry to mid-level gaming PC to run certain 3D apps (and real gaming). Given how well Geforce Now works, there has to be some remote-control software that could put that PC on my Mac screen, usable for both 3D work and gaming, right? I'd spend less money, have a more capable setup.
I'd rather give Apple my money for a Studio. But, Apple has kind of abandoned the mid-to-high end marketing needing GPU capability, at least for now.
Mike Wuerthele said: Sure. To make a long and complex story simple, Intel chips on-die have the ability to address external graphics processors, like PCI-E ones. With that, all you need is a software driver.
Apple Silicon does not. All video work must be done on-die.
Thanks Mike... should have read the comments first.
... Yes, there is a very small percentage of users who need expandability, insanely fast graphics cards, and over a TB of RAM. They are not in Apple's target market.
Just to be clear, this isn't about 'insanely fast graphics cards' but trying to have adequate graphics performance, even compared to a fairly modest PC.
Even if the the new Mac Pro could match could match a PC with a single higher-end GPU (or the old Mac Pro with a single GPU, let alone 4+), I'd say it is a good start. I'd then wonder how they will be expanding that, but it would be a reasonable start. My issue here, is I don't feel it was a reasonable effort.
I’d guess that they would have liked to do more with AS Mac Pro, but for whatever reason were not able to get it done this year. Hopefully they will keep pushing forward. From taking a year longer than their 2 year transition goal, to failing to deliver the planned Jade 4C “Extreme” SoC in both M1 and M2 generations, seems like they really hit some roadblocks from their original intent. I highly doubt they would have given 2019 Mac Pro such dramatically better expansion capabilities if they knew they wouldn’t be able to maintain that level of performance after Apple Silicon transition.
One has to suspect the ongoing braindrain on the Apple Silicon design team is part of these stumbles. Let’s hope Srouji gets the big Mac back on track, and their patented multi-GPU tech comes to light soon!
Yeah, my hope is this was a stop-gap to complete the Apple Silicon transition. If they add ray tracing to the M3, that will close the gap a bit more. Then, hopefully, they'll eventually figure out a way to expand/scale that even more, as you say. It's just that we might be talking years away at this point. And, I'm not sure where that leaves the rest of the lineup (like Studio) on the GPU-front (hopefully the M3 version becomes acceptable in that regard).
Marvin said: By comparison, the M2 Ultra is faster than the 28-core Intel chip and faster (27TFLOPs, assuming both GPUs fully used) than the higher-end Radeon GPUs.
Is that really the case, though? Maybe in terms of TFLOPs, but w/o hardware for RT and such, won't certain 3D tasks will perform very poorly compared to a Radeon? I've watched several videos using Studios or MBPs in different 3D/CAD apps and comparing them to PCs (even laptops) where they don't do so well. I suspect some of this is software optimization, but I have to wonder how much is hardware related.
My understanding (maybe even based on a link you gave me some time ago?) is that the GPU approaches are quite different, with advantages and disadvantages (not just TFLOPS numbers). As far as this applies, the problem is that most of the market is developing for, and accustomed to, AMD/Nvidia, so where Apple falls short, it will be quite noticeable in workflows.
Marvin said: The only market worth targeting is the enthusiast market that would today buy an i9-13900K + Nvidia 4090 for around $4k. The M2 Ultra is within 30% of the CPU and probably 1/3-1/4 the GPU (lower when using hardware raytracing) for $7k.
Yes, this is really the issue, and not just for 3D people, but people wanting to game on the Mac. But, i think it goes even further than that. What about someone who'd buy a mid-level gaming PC with a 3070 or 3080? Then we're in the under $2k range. I still don't think the M2 Ultra will compete, but even the M2 Max becomes expensive. If the GPU is on the table, Apple has no mid-to-high end anymore, though better lower-end than most of their past lineup. It becomes more about justifying any spending beyond a mini or MBA.
Marvin said: ... According to this, Nvidia sold about 30 million desktop GPUs in 2022. Notebook GPUs would be at least as many and their GPU revenue was $12b.
$12b / 60m units = $200 ASP.
High-end GPUs like the 4090 ($1600+) sell around 1m units per year. The majority of GPUs sold are consumer gaming cards:
The Ultra Mac Studio and Mac Pro are aimed at the market that buys single 4090 GPUs. The market above this ($5k+) is well below 1%. The HEDT/enthusiast market is around 5-10%.
The M2 Ultra falls short of a $4k i9-13900 + 4090 but it's at least competitively priced in the Mac Studio. Chasing after a minuscule portion of the high-end market isn't a high priority, it can easily wait for M3 for hardware RT and possibly some extra GPU chip.
Keep in mind that one huge factor impacting those GPU sales graphs, was the demand for GPUs (and increased prices!) due to crypto-mining. Yes, I sure how the M3 closes the gap more, or I'll be buying one of those PCs.
I just saw the M2 UItra Metal result on GB: 281,948, which IS better than I can get with my eGPU, which gives me 191,426 on GB 6.1. So I guess Apple has accomplished what I was hoping they would.
Just be sure you compare the apps/workflows you run, not just GB scores. For example, I've seen YouTube videos of people in 3D/CAD apps on the M1 Ultra where they perform incredibly well in one regard, while becoming almost unusable in others.
I'm trying to remember, but in one 3D sculpting app, the interface got so laggy, the guy straight out said he wouldn't be able to use it. I think his daily-driver was a several year old PC laptop with an RX580. (What wasn't clear, is how much of that problem was software optimization vs hardware... but at the point where it kills your workflow, it doesn't matter all that much.) On the other hand, he was able to replicate bunches of copies of the models (and insane amount of geometry) and the Mac Studio still worked, where he said his PC would have crashed.
macxpress said: The wedgeless MacBook Air is Apple's best selling laptop so that statement about it being a mistake is BS. Just because you don't like it doesn't mean the rest of the market doesn't.
You can't tell, though, if you don't have the choice. This is similar to Apple making iPhone screens bigger and concluding it's fine because people keep buying iPhones. My wife loved her wedge MBA, and hates the new form-factor... but she also wanted a new computer. So, she's dealing with a form-factor she doesn't like... the market didn't speak.
My own uneducated guess is that it's an Apple Silicon PCIe GPU/Unified Memory extension which doubles the GPU power of the Ultra, and not the rumored 4x "Extreme" design.
That would be great for Mac Pro users, but kind of suck for the rest of us. Hopefully there is some plan to give some more GPU power to Studio and laptop users, too.
I think we will look back at this article as a very sad misprint. I can't find where video cards will not be supported.
Finally, Mac Pro brings PCI expansion to Apple silicon. It features six open expansion slots that support gen 4, which is two times faster than before. So users can customize Mac Pro with essential cards, including audio and video I/O, networking, and storage.
Right there in the keynote, Apple says the new Mac Pro will support video card for input and output.
They addressed no video cards several times, most recently in Gruber’s talk show.
I think maybe the confusion here is over the difference between GPUs and video I/O cards.
I think we will look back at this article as a very sad misprint. I can't find where video cards will not be supported.
Finally, Mac Pro brings PCI expansion to Apple silicon. It features six open expansion slots that support gen 4, which is two times faster than before. So users can customize Mac Pro with essential cards, including audio and video I/O, networking, and storage.
Right there in the keynote, Apple says the new Mac Pro will support video card for input and output.
They addressed no video cards several times, most recently in Gruber’s talk show.
I think maybe the confusion here is over the difference between GPUs and video I/O cards.
Marvin said: By comparison, the M2 Ultra is faster than the 28-core Intel chip and faster (27TFLOPs, assuming both GPUs fully used) than the higher-end Radeon GPUs.
Is that really the case, though? Maybe in terms of TFLOPs, but w/o hardware for RT and such, won't certain 3D tasks will perform very poorly compared to a Radeon? I've watched several videos using Studios or MBPs in different 3D/CAD apps and comparing them to PCs (even laptops) where they don't do so well. I suspect some of this is software optimization, but I have to wonder how much is hardware related.
My understanding (maybe even based on a link you gave me some time ago?) is that the GPU approaches are quite different, with advantages and disadvantages (not just TFLOPS numbers). As far as this applies, the problem is that most of the market is developing for, and accustomed to, AMD/Nvidia, so where Apple falls short, it will be quite noticeable in workflows.
Marvin said: The only market worth targeting is the enthusiast market that would today buy an i9-13900K + Nvidia 4090 for around $4k. The M2 Ultra is within 30% of the CPU and probably 1/3-1/4 the GPU (lower when using hardware raytracing) for $7k.
Yes, this is really the issue, and not just for 3D people, but people wanting to game on the Mac. But, i think it goes even further than that. What about someone who'd buy a mid-level gaming PC with a 3070 or 3080? Then we're in the under $2k range. I still don't think the M2 Ultra will compete, but even the M2 Max becomes expensive. If the GPU is on the table, Apple has no mid-to-high end anymore, though better lower-end than most of their past lineup. It becomes more about justifying any spending beyond a mini or MBA.
Hardware RT definitely makes a difference for some things. Some apps only use it properly on Nvidia (Optix), AMD doesn't do so well in Blender:
The M2 Ultra is on page 2 and in the same performance range as AMD's 7900XTX as well as all the mid-range Nvidia GPUs.
Nvidia's RT means that anything above a desktop 3070/4060 is faster and the $1600 4090 is 3-4x faster. Nvidia is doing a great job getting optimal software and hardware working together:
For standard rendering, the GFXBench 4K Aztec 3D test has the 4090 highest again at 507FPS, M2 Ultra 299FPS, about 60% of the performance. M2 Ultra is pretty much even with AMD's highest end GPUs.
Given that Apple wouldn't support Nvidia, a Mac Pro would have offered those high-end AMD GPUs and allowed two of them. The M2 Ultra is equivalent to an Intel Mac Pro with 24-core Xeon W-3345 (250W) plus an AMD W7900 (295W), excluding RT cores.
Hardware RT takes time to implement properly and should allow for motion blur and shader execution order. Apple is doing the software side just now. If it can make it into M3, the M3 Ultra will have the ability to beat all the other GPUs in the Studio form factor but the M2 Ultra looks like it's performing well.
I thought Apple was waiting till the M3 chip was ready to release an updated Mac Pro?
I think this is what they wanted to do but I bet there were yet more delays with M3 so they couldn't keep delaying the release of a new Mac Pro. As I've said before, I don't think this is the Mac Pro they wanted to release. They just had to get something to market so they did the best they could with what they had at this time. I wouldn't be surprised if next year they release an M3 Extreme based Mac Pro with something like 2 Ultra's put together.
I think Apple quite simply painted themselves into a corner on the Mac Pro. When they started the transition, they had no issue with MacBooks, the iMac, etc. But the Mac Pro was something I think they knew would be an issue, but had 2-3 years to figure out. I suspect the Mac Studio was the result of not having solved the problem in time…that being the entire concept of the Mac Pro wasn’t congruent with the integrated architecture they had produced. Frankly, I think Apple looked at the problem and size of the true pro market, then designed this because they promised they would.
Interesting video on YouTube from Daring Fireball that I always like to watch the week of WWDC is "The Talk Show" and they talked about Mac Pro and Mac Studio with John Ternus and Joz. Yeah some the responses are kinda skirting the question but they do give good information in there as well and some reasons behind their decisions. Judging by their responses, it definitely sounds like both are here for the long run and they serve 2 different purposes.
It's a 2hr show but I always enjoy watching it. They always give some good information on the topics asked. Yeah there are certain things they can't talk about so sometimes it sounds like their answers are BS but that goes with any company. No company is gonna absolutely spill their guts on every single decision made.
Here's the quote from Ternus on the question of PCIe GPUs (starts at 24:14): "Fundamentally, we’ve built our architecture around this shared-memory model and that optimization, and so it’s not entirely clear to me how you’d bring in another GPU and do so in a way that is optimized for our systems. It hasn’t been a direction that we wanted to pursue."
The key clause there is "do so in a way that is optimized for our systems." Could Apple itself do that, with Apple silicon via PCIe? Earlier in the interview, Ternus points out that, because of the Unified Memory architecture, the amount of memory available to the GPU in Apple silicon is orders-of-magnitude larger than in any other GPU. So maybe this hypothetical Apple silicon PCIe GPU extension for the Mac Pro doesn't need to have its own memory, it draws on the system's Unified Memory. So then it's an optimization problem, in software. They talk about how they've been making big improvements in terms of optimization, this would fit into that effort.
There is an emerging system in Server space called CXL for exactly these sorts of shared memory pools in modular devices. AMD have said their aim to to bring that out of server space into consumer space. Mac Pro would seem to be key candidate for that and it would allow Apple to do multi-Ultra Mac Pros.
It does seem this MacPro is an inter-mediate step it where it could go in the future.
I’m curious how you’re fitting four 6900’s in the Mac Pro. I was only able to get two. Any examples?
I think Apple has once again drawn a line in the sand making it clear they are OK with true power users having to use PC’s. Apple is just bored by and unserious about truly high end computing. I’m guessing they are OK with folks with extreme use cases moving to commodity Windows machines.
I’m a Houdini user on my Mac Pro, use it for running special effects simulations and rendering. The Mac was already a slightly painful platform for 3D work because of lack of Nvidia support, but with a cap of 192GB Ram, simulation work is out the window. And 3D rendering is a joke without multiple GPUs. So, after being a lifelong Mac user, my next primary machine will have to be a Windows box, which makes me ill. But, I get it, I’m a niche user and Apple just isn’t in the unsexy, geeky, high end compute business.
Apple said they were working on a modular Mac Pro when they started this and still has not shown that ability. They had to hold something back that they couldn’t finish while working on the Vision Pro.
I can’t imagine them abandoning these users. They are showing a focus on games that I’ve never seen before.
Comments
DANG THIS IS SAD.
if Apple spent a fraction of the effort they put into the stupid headset on the Mac Pro, we’d have a world beating computer to rule them all. Instead we get an excuse to settle for the Mac Studio.
Listed Tech specs for the Mac Pro. So you have two x16 slots that can support video cards. Granted the AUX power is only 300W total.
Again the same Aux power cables that are used on the 2019 Mac Pro are also compatible with the 2023 Mac Pro.
The articles premise is most of the target market is missed by the new Mac Pro. Based upon what? Even without video cards, the expansion is highly wanted and useful. Just having internal storage is highly useful, because it is much faster than external storage.
https://www.apple.com/shop/buy-mac/mac-studio/12-core-cpu-30-core-gpu-16-core-neural-engine-32gb-memory-512gb#
There's M2 Max with 30 and 38-core GPUs. M2 Ultra with 60 and 76-core GPUs.
A 76-core Ultra would be expected to be more than 2x a 30-core M2 Max.
Here's an M2 Ultra at 223k:
https://browser.geekbench.com/v6/compute/538846
281 / 223 = 1.26x. 76-core / 60-core = 1.26x.
Here's M2 Max at 122k and 144k:
https://browser.geekbench.com/v6/compute/541704
https://browser.geekbench.com/v6/compute/538762
281 / 144 = 1.95x. (presumably 76-core Ultra vs 38-core Max)
223 / 122 = 1.82x. (60-core Ultra vs 30-core Max)
There are also different power levels in the OS, these two Max pages have different clock speeds.
They also have different cooling capability. A Mac Studio will cool an M2 Max better than a Macbook Pro 14".
It looks like they have improved the Ultra GPU scaling though. We'll see with the Blender tests:
Blender Opendata
M1 Max (32C) is on page 5 with 1036.
M1 Ultra (64C) is on page 4 with 1831.
M2 Max (38C) is already above M1 Ultra on page 3 with 1917.
If M2 Ultra (76C) can scale to near 2x, that will bring it to the top of page 2. This is 1/3 from the top overall (and those others have hardware RT).
An M2 Ultra 76-core showed up on page 2 with 3412, 1.77x the M2 Max.
A 4090 is about 3.8x, 4080 is 2.8x, 3090 is 1.8x.
To get near to the top, they'd need a 50% boost from M3 to make a 114-core Ultra and then add a second GPU chip that is equivalent to another 114-core Ultra. Or maybe 50% boost + hardware RT would be enough.
It's a 2hr show but I always enjoy watching it. They always give some good information on the topics asked. Yeah there are certain things they can't talk about so sometimes it sounds like their answers are BS but that goes with any company. No company is gonna absolutely spill their guts on every single decision made.
The key clause there is "do so in a way that is optimized for our systems." Could Apple itself do that, with Apple silicon via PCIe? Earlier in the interview, Ternus points out that, because of the Unified Memory architecture, the amount of memory available to the GPU in Apple silicon is orders-of-magnitude larger than in any other GPU. So maybe this hypothetical Apple silicon PCIe GPU extension for the Mac Pro doesn't need to have its own memory, it draws on the system's Unified Memory. So then it's an optimization problem, in software. They talk about how they've been making big improvements in terms of optimization, this would fit into that effort.
But, I'm curious about this statement. I was under the impression it was more a driver type situation, and maybe eGPUs could even make a come-back.
I'm not even in the market for a Mac Pro, but I look to it to see Apple's potential and direction. In that regard, I really hope this is a stop-gap to get the transition done, because it is very disappointing.
For myself, I'm left with a bit of dilemma. I want GPU power for 3D stuff (and, sure, I'd love gaming). But, unless the M3 is a real leap (which it might be), I'll be left with buying a fairly expensive Studio (kind of my dream machine), that barely competes with with a $1500 gaming PC, and might be questionable about usability, outside of question of being competitive.
Or, I could buy a base-level mini, which will probably be just fine for most of my computing needs, and then get that entry to mid-level gaming PC to run certain 3D apps (and real gaming). Given how well Geforce Now works, there has to be some remote-control software that could put that PC on my Mac screen, usable for both 3D work and gaming, right? I'd spend less money, have a more capable setup.
I'd rather give Apple my money for a Studio. But, Apple has kind of abandoned the mid-to-high end marketing needing GPU capability, at least for now.
Just to be clear, this isn't about 'insanely fast graphics cards' but trying to have adequate graphics performance, even compared to a fairly modest PC.
Even if the the new Mac Pro could match could match a PC with a single higher-end GPU (or the old Mac Pro with a single GPU, let alone 4+), I'd say it is a good start. I'd then wonder how they will be expanding that, but it would be a reasonable start. My issue here, is I don't feel it was a reasonable effort.
Yeah, my hope is this was a stop-gap to complete the Apple Silicon transition. If they add ray tracing to the M3, that will close the gap a bit more. Then, hopefully, they'll eventually figure out a way to expand/scale that even more, as you say. It's just that we might be talking years away at this point. And, I'm not sure where that leaves the rest of the lineup (like Studio) on the GPU-front (hopefully the M3 version becomes acceptable in that regard).
My understanding (maybe even based on a link you gave me some time ago?) is that the GPU approaches are quite different, with advantages and disadvantages (not just TFLOPS numbers). As far as this applies, the problem is that most of the market is developing for, and accustomed to, AMD/Nvidia, so where Apple falls short, it will be quite noticeable in workflows.
Yes, this is really the issue, and not just for 3D people, but people wanting to game on the Mac. But, i think it goes even further than that. What about someone who'd buy a mid-level gaming PC with a 3070 or 3080? Then we're in the under $2k range. I still don't think the M2 Ultra will compete, but even the M2 Max becomes expensive. If the GPU is on the table, Apple has no mid-to-high end anymore, though better lower-end than most of their past lineup. It becomes more about justifying any spending beyond a mini or MBA.
Keep in mind that one huge factor impacting those GPU sales graphs, was the demand for GPUs (and increased prices!) due to crypto-mining.
Yes, I sure how the M3 closes the gap more, or I'll be buying one of those PCs.
Just be sure you compare the apps/workflows you run, not just GB scores. For example, I've seen YouTube videos of people in 3D/CAD apps on the M1 Ultra where they perform incredibly well in one regard, while becoming almost unusable in others.
I'm trying to remember, but in one 3D sculpting app, the interface got so laggy, the guy straight out said he wouldn't be able to use it. I think his daily-driver was a several year old PC laptop with an RX580. (What wasn't clear, is how much of that problem was software optimization vs hardware... but at the point where it kills your workflow, it doesn't matter all that much.) On the other hand, he was able to replicate bunches of copies of the models (and insane amount of geometry) and the Mac Studio still worked, where he said his PC would have crashed.
You can't tell, though, if you don't have the choice. This is similar to Apple making iPhone screens bigger and concluding it's fine because people keep buying iPhones. My wife loved her wedge MBA, and hates the new form-factor... but she also wanted a new computer. So, she's dealing with a form-factor she doesn't like... the market didn't speak.
That would be great for Mac Pro users, but kind of suck for the rest of us. Hopefully there is some plan to give some more GPU power to Studio and laptop users, too.
Blender opendata
The M2 Ultra is on page 2 and in the same performance range as AMD's 7900XTX as well as all the mid-range Nvidia GPUs.
Nvidia's RT means that anything above a desktop 3070/4060 is faster and the $1600 4090 is 3-4x faster. Nvidia is doing a great job getting optimal software and hardware working together:
https://www.pcgamer.com/nvidia-rtx-40-series-ray-tracing-performance/
https://www.nvidia.com/en-us/geforce/news/geforce-gtx-dxr-ray-tracing-available-now/
https://developer.nvidia.com/blog/accelerating-inference-up-to-6x-faster-in-pytorch-with-torch-tensorrt/
For standard rendering, the GFXBench 4K Aztec 3D test has the 4090 highest again at 507FPS, M2 Ultra 299FPS, about 60% of the performance. M2 Ultra is pretty much even with AMD's highest end GPUs.
gfx bench
Given that Apple wouldn't support Nvidia, a Mac Pro would have offered those high-end AMD GPUs and allowed two of them. The M2 Ultra is equivalent to an Intel Mac Pro with 24-core Xeon W-3345 (250W) plus an AMD W7900 (295W), excluding RT cores.
Hardware RT takes time to implement properly and should allow for motion blur and shader execution order. Apple is doing the software side just now. If it can make it into M3, the M3 Ultra will have the ability to beat all the other GPUs in the Studio form factor but the M2 Ultra looks like it's performing well.
It does seem this MacPro is an inter-mediate step it where it could go in the future.