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 should use NVIDIA cards
- Thread starter Zest28
- Start date
- Sort by reaction score
Nvidia is just another Apple. They make their own ecosystem and not allow open up their system to people.
That's why Apple ditched Nvidia over AMD since they can modify/upgrade their GPU for Apple to use Metal. Nvidia's CUDA is simply not allowed that.
NVidia didn't want Apple's business enough, if at all. I got the impression their beta driver program was one or two people in a cubicle in some distant unloved corner. Huang expected Apple to 'throw them a bone' (he tweeted about it) without actually making much of an effort. He could have picked up the phone and actually asked about winning some business. Apple certainly is anything but blameless in this, but ATI/AMD were certainly more amenable. Both of them were in Steve Jobs' doghouse at different times, only one got out of it…
Apple is trying to attract machine learning researchers.
www.swift.org
On-device ML research with MLX and Swift
The Swift programming language has a lot of potential to be used for machine learning research because it combines the ease of use and high-level syntax of a language like Python with the speed of a compiled language like C++.
Nvidia is already dominating in machine learning and AI for a while and Apple is just a latecomer. Beside, Mac is using their own library and language which is a huge disadvantage and who even use Mac? Mac is limited only for 2D markets such as video, music, photo, illustration, and more. If you gonna use 3D or other stuffs, you better use PC instead. Too bad that Apple themselves are limiting their own market.Apple is trying to attract machine learning researchers.
On-device ML research with MLX and Swift
The Swift programming language has a lot of potential to be used for machine learning research because it combines the ease of use and high-level syntax of a language like Python with the speed of a compiled language like C++.
Last edited:
I disagree as my M3 Max laptop easily trounces my 2019 Mac Pro in After Effects; and actually it trounces most PCs as well if you believe the validity of Pugent System's AE benchmark.
And everything to do with video editing is way faster on the M3 also, due to the hardware encoders/decoders. Basically anyone using Intel Macs will see immediate faster results on M3. I say M3 because in my eye it is the first system that checked 90% of the boxes for me.
It's kind of sad how much better this laptop is at After Effects; just goes to show how much Intel's Xeon chips from that era sucked.
And yet, you only mentioning the video side which only proves my point. Beside benchmarks proves nothing compared to real life working. Currently, Mac is only limited to a few markets and even then, the hardware specs especially RAM is a huge problem.
Umm, back when nVidia cards were still supported, You could use OpenCL, Metal or CUDA for compute, at least all those options were available in Adobe Creative Cloud apps.
Metal has gotten a lot better since then.
As far as 3D goes, Apple still commands the largest GPU Ram pool for a single card; that is very beneficial for modeling and such.
The fact that Blender, Octane and RedShift are all embracing Metal on macOS for 3D GPU rendering is a good thing and goes to show that Apple must be serious, otherwise they would not bother.
I am telling you right now in real world usage, the M3 Max is faster than my Mac Pro at everything. Have not found one thing that the Mac Pro is faster at. The only thing I could beat my M3 Max on would be GPU rendering, but then I would need to add another W6800X card and why waste money on something like that.
If a M3 Ultra or Extreme is ever going to be a thing; it will be very good.
Because AS Mac Pro sucks and you are comparing M3 Max to M2 Ultra or Intel Mac Pro which is not a fair comparison. Even then, it will be limited for a few market usages and it will never be great on 3D or other stuffs where PC is good at.
Why do I get the feeling Apple would create an AI Afterburner card before supporting Nvidia.
Why isn't a fair comparison, in my opinion the 2019 Mac Pro was the last "workstation" class tower Apple released; pretty much why I bought one only to be stabbed in the back by Apple with the Apple Silicon announcement.
If one *must* be in the Apple ecosystem or prefers macOS over Windows, then it is every bit of a fair comparison.
And since you are kind of getting snotty, please name one nVidia card that has more VRAM than the top spec M3 Max's GPU can address?
You are comparing with 3~4 years old tech. Such a poor comparison after all.
I am not in charge of Apple's upgrade cycle, the 2019 should have been updated a month after new Xeon's were announced by Intel. But Apple Silicon, so that was a no-go for obvious Apple reasons.
Ummm, the last time I checked the Pugent site on After Effects benchmarks (last month); the M3 Max was beating everything.
After Effects is a huge part of my workflow, so performance in it is very important to me; but I also kind of hate Windows. But I have a Windows box anyway.
Everything is old tech the minute it gets purchased. So what's your point?
You really don't think there's any validity in the idea that a modern computer shouldn't be compared to a 4 year old one? I think the spirit of this thread is looking at Apple through the lens of modern x86 (Windows/Linux) offerings.
Wow, you really think that's how it works? Your logic already failed.
I think it would only be valid if macOS ran on those systems *easily* (it can be done, but for the most part, too much of a headache.)
If we digress from this forum being primarily a Mac focused forum and we are purely just comparing tech to tech, then yes it is valid to compare everything to everything.
CPU wise; Apple still holds it's own unless you want to do something stupid like compare it to a 96-core thread ripper and complain because the thread ripper is faster at multi-threaded workflows --- DUH!
GPU wise, Apple still has a lot of work to do, but they also command the largest VRAM pool that can be allocated. Based on the timeline of 3 years, I think they are doing pretty good. And Apple has not even attempted to make a power slurping massive GPU. I doubt they ever will.
Nobody knows what Apple will do because they also never release roadmaps, which is also something that bugs the crap out of me. But at the end of the day, I am very impressed with the M3 Max.
That is how it works, there is always something better right around the corner.
No it does not. Stop trolling especially since you are comparing Intel Mac vs Apple Silicon Mac.
I am the one trolling? This is a Mac forum last I checked. Y'all are beating a dead horse that has been beat to death so many times here; even by me.
Found the guy who thought NVDA was too expensive at 300, now he's doubling down.
"Nvidia's largest US customers are Amazon, Microsoft, Google, Meta, and Dell."
You see they are selling to real companies, not SPAC funded AI companies.
Specifics are few and far between, but I believe that R1 is the component responsible for managing the spatial aspects of Vision Pro in conjunction with the cameras and sensors built into the device. The graphics side is still left to the M2.
Because they are petty and more concerned with egos than doing what would be best for their actual customers.
It's really silly