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 2023 = Another Mac Pro 2013 crisis and failure.
- Thread starter sunny5
- Start date
- Sort by reaction score
Nothing went wrong with the M extreme - based on information shared by this very website - they just thought it wasn’t worthy the investment to such a small niche.
Mac Pro will get naturally better and definitely ready in the next years as ram and graphical performance improves.
1. Even if it was true, for the end user it's a distinction without a difference.
2. "only how fast you can do it". Only? Is not it hugely important? Netbook with 192G of RAM (if it existed) could be able to do everything Mar Pro does only way way slower.
6. "True, but why do we want to promote NVIDIA vendor lock in?". Well, Apple is the king of lock ins so it's illogical for the Apple customers to be concerned about lock ins.
Ok I think there is a communication problem: I wrote that the difference between the Mac Studio and the Mac Pro lies in the thermal limitation of the Mac Studio COMPARED the Mac Pro, NOT that the Mac Studio has a thermal problem.
The Mac Studio has the additional components external, it's necessary that it doesn't heat up, otherwise it would be a big problem!
Otherwise make it simple: call Apple and tell her that it has installed 2 unnecessary fans in the Mac Pro!
This must be a forum full of engineers, otherwise it is not explained how a simple concept can be equivocal 22,000 times in half a day.
The Mac Studio has the additional components external, it's necessary that it doesn't heat up, otherwise it would be a big problem!
Otherwise make it simple: call Apple and tell her that it has installed 2 unnecessary fans in the Mac Pro!
This must be a forum full of engineers, otherwise it is not explained how a simple concept can be equivocal 22,000 times in half a day.
1. Same thing. They CANT just make it. Mac Pro 2023 proves it.
2. Compare to RTX 30 series, M2 Ultra is just a joke. Beside, RTX 30 series aren't even TSMC 5nm.
3. Face falm
4. Which is Intel Mac based, not Apple Silicon Mac.
5. You are proving yourself wrong.
6. Because Nvidia has the best graphic card and it's an industry standard.
7. I doubt that.
Like the OP stated it’s lacking in GPU performance in desktops vs the competition.
Anything related to GPU task. M series are only good for power by watt, not performance itself especially for GPU. They also limit the power consumption too much as well.
Because they are based on A series or mobile chip. Yet still proving that Apple GPU performance is poor. Beside, that's a poor excuse for Mac desktop.
The reason why Apple has the efficiency edge is mainly due to Apple having an edge in terms of the node they use, the PDN (power delivery network) tech, and packaging.
Their ARM cores are actually more complex than the x86 competitors; significantly wider and with larger resources for out of order and speculation. Most people assume there is some kind of "magic" that makes ARM better that x86, but that is not the case. The ISA has little impact on overall power consumption given the same microarchitectural resources.
Apple uses their larger/more complex cores to their advantage, by running them at a slower clock rate. While allowing them to do more work per clock cycle. This allows them to operate on the frequency/power sweet spot for their process. One has to note that power consumption increases significantly (way higher than linear) the higher the frequency.
Here is where the PDN technology comes into play. Apple uses the most advanced technology to distribute power to keep all the functional units feed, which requires the ability to supply a lot of instantaneous power. To do so, Apple uses a 3D stacked architecture of 2 dies; one for the logic, and another one on top (or bottom depending where you look at it) to distribute the power. In contrast, almost every one else has to use the same die to do logic and distribute power.
The irony is that a simpler/smaller ARM core would have to be clocked faster in order to compete with Intel/AMD cores. And it would end up consuming the same high power.
Apple also has a very good SoC design. Meaning that they integrate most of the system on a single die; the CPUs, the GPU, the NPU (AI accelerator), the Codec (video processing), the camera block, I/O (USB, WiFi, ethernet, PCIe/TB, etc), and the memory controller.
For some stuff like AI and video encoding, having custom silicon handling it is far far more efficient than running it on a general purpose code.
Lastly, it also comes to packaging. Apple not only integrates the SoC in a single die, but it has the memory chips on the same package. This allows them to use low power mobile DDR chips, and since they are on package it also reduces significantly all the power that having the memory transactions run through the system's PCB externally would consume.
So it's a combination of Apple using a single package where Intel/AMD laptops require multiple through their PCBs to support the same functionality. As well as Apple having access to better overall fabrication technology for that single package that AMD/Intel have for theirs.
The trend seems to be that it is becoming more efficient for mobile vendors to scale up their products into laptops, than it is for desktop vendors to scale down their products into laptops.
There is also a key difference in business models: Apple is a system's vendor. Meaning that they sell the finished product, not just the processors. So they can use several parts from the vertical process to subsidize others. In this case, Apple can afford to make very good SoCs because they don't sell those chips elsewhere, meaning that they are not as pressured to make them "cheap" in terms of area for example. Since they're going to recoup the profit from elsewhere in the product.
In contrast; AMD and Intel sell their processors to OEMs, so they only get profit from the processor not the finished system. So they have to prioritize cost, by optimizing their designs for Area first and then focus on power. This is why both AMD and Intel use smaller cores, which allows them for smaller dies. But which have to be clocked faster in order to compete in performance, unfortunately that also increases power.
This is probably they ke difference; Apple can afford the larger design that is more power efficient for the same performance. Whereas AMD/Intel have to aim for the smaller design that is less power efficient for the same performance.
Disappointed and confused... and we hit the limit of Apple Silicon as it seems. I was expecting much more processing power and certainly much more than 192GB of memory. It's a big Studio with much emptiness inside. Yes it saves buying external PCIe enclosures and that might be worth it for some. People doing video/audio work will be happy, the rest probably not so much. Was looking forward to better graphics and ML performance and here we are with a rusty old 1080Ti still running circles around it in something like PyTorch. It's not a trash can, more like a trash container and while not thinking about it for too long, the Studio + external enclosures might actually seem like a better solution considering the enclosures can be reused after upgrading the Studio to a new model while the whole "PCIe block" of the MP is thrown away as well with an upgrade. Maybe time to give up on the idea of a full workstation from Apple and just stick to Dell Precision, Lenovo or Nvidia DGX boxes for that purpose. 🤷♂️
Doesn't matter. Results tell everything. Since Apple Silicon is also being used for desktop, it's not a good excuse to make that Apple GPU is poor for purpose or something else. Beside, Mac also use software that PC has except very few software like Logic Pro and FCPX which also can be replaced with other software. Even if SoC is good, it will keep disappoint us because of performance and limitation just like Mac Pro 2023 proves it.
Like I said, this is why Nvidia is dominating the GPU market by 80~90% and they are the industry standard. It's hard to ignore it especially since it was Apple who compared their chips to Nvidia GPU when they were so proudly announced M1 Max and Ultra. And it turns out, they never come closer.
1. Not the same thing, putting CANT in big bold letters doesn’t make it true.
2. Don’t know that yet… the M2 Max is putting up a pretty good showing in compute limited workloads - in Blender Open Data it looks like the M2 Max is comparable in compute to the desktop 3070 or mobile 3080 Ti which is pretty amazing. If you include the RT cores sure the NVIDIA GPUs pull ahead but by that logic NVIDIA is way way way behind because they don’t have any ProRes hardware on their chips.
3. Again, citation needed
4. Most 3D rendering software has been ported or is likely to be ported to apple silicon, there is a whole thread on this forum talking about it.
5. You asked why we even have the Mac Pro line, there are reasons, as many many others have pointed out, ignoring those reasons because you don’t like the answer doesn’t make the answer go away.
6. So Intel and AMD should just give up on fighting for the data centre? Apple should give up on Metal Compute And CoreML?
7. The best selling Mac has an M2 and has twice the GPU compute muscle of the PS4, Apple should want game studios to target that because it would open up AAA games to the majority of their users. I think you are wildly off base if you think Apple wants game studios to target the M2 Ultra for a good mac gaming experience.
Which is laptops, not desktop especially who needs high GPU performance. Clearly, you dont get the point. I kept saying same thing over and over again.
Considering the raw compute performance of the M2 Ultra that sounds like a PyTorch problem not an apple silicon problem.
That's a limitation of scaling A series core architecture. They meant for mobile devices, not desktop or workstation. Otherwise, they wouldn't mess up Mac Pro 2023 like that.
I'm still holding to the belief that there is at least one scaling level possible beyond Ultra (M3 Quadra/Extreme), and we didn't see it yet due to supply chain / fabrication process constraints.
They released the Pro with only 1 SoC choice because they couldn't make it work yet, and couldn't carry selling the Intel Mac Pro forever.
But still, the pricing is outrageous and out of every logic.
$1000 upselling compared to the Studio would probably have made them just as much money by increasing the volume.
Apple could actually do that considering 13 months ago 240W USB PD standard was released. This would be sufficient to power a MBP 16" M2 Ultra.
Although it would be very very loud like Intel's Core i9-13980HX laptop but probably not as hot.
There is actually a market for it which many MR users refuse to acknowledge even when Apple used to sell Intel laptops with this characteristic.
Still, there are people who prefer desktops... even those with PCIe slots.
1. You are refusing the fact: They cant make it. Can and can not is totally different.
2. We pretty know that since M2 series are just upgraded version of M1 series.
3. Again, you dont know what you are talking about.
4. And how many of them are truly available? Beside, they are still Nvidia based software which can not take advantage with CUDA as Apple ditched them.
5. Only one chip on Mac Pro clearly proves that they dont know what they are doing.
6. Do they even trying? None of them are. The truth is, Nvidia invested a lot of time and money on GPU for several decades unlike other companies.
7. Mac platform itself is extremely hostile to gaming industry so I highly doubt it. It has nothing to do with the performance and even then, Apple GPU is still poor and yet much more expensive than what PC can do.
Apple ditched Intel and AMD because they dont wanna pay for their components and save money. Guess what? Apple can just increase the price whatever they want.
Since Apple proves themselves as they can not make chips for Mac Pro, I dont think they can even solve it for now.