I'm guessing you don't work in content creation. You know, that burgeoning industry that every tech and media company is getting into? It is impossible to be too fast in rendering video. Time is money.
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.
M1 Ultra CPU is 60% Faster Than 28-core Mac Pro, GPU is 80% Faster Than Highest-End Radeon Pro W6900X Graphics Card
- Thread starter MacRumors
- Start date
- Sort by reaction score
And how much power will that use? 2X? Good God man the machine you describe will be a space heater.
Does anyone have any idea how was the maxed out MACBOOK PRO M1 MAX
10-Core CPU / 32-Core GPU / 64GB Unified Memory
Compares to:
MAXED OUT STUDIO MAC
Apple M1 Ultra with 20-core CPU, 64-core GPU, 32-core Neural Engine / 128GB Unified Memory
How much faster is the Studio Mac?
Are there any benchmarks or comparisons anywhere?
10-Core CPU / 32-Core GPU / 64GB Unified Memory
Compares to:
MAXED OUT STUDIO MAC
Apple M1 Ultra with 20-core CPU, 64-core GPU, 32-core Neural Engine / 128GB Unified Memory
How much faster is the Studio Mac?
Are there any benchmarks or comparisons anywhere?
Last edited:
Similar to my experience when I bought an $1100 M1 Mini that seemed way faster than my Intel MBP16 that cost about 3 times as much and ran hot and noisy... I was delighted with the advances that Apple had made, even though there was a mild sense of loss at how quickly my expensive MBP had been superceded.
Yeah, I didn't miss the Mac Pro bit. Very interesting.
The other thing that grabbed my attention was the 2.5 TB/s (UltraFusion!) interconnect bandwidth (at 27:55). Johny then said that it's "more than 4 times the bandwidth of the leading multi-chip interconnect," which I'd think refers to AMD's Infinity Fabric 3.0 at 400 GB/s bidirectional. Oddly, though, he could have also said "more than 6 times the bandwidth," so I'm not sure what to make of that.
Anyway, the M1 Ultra has 800 GB/s memory bandwidth, 400 GB/s per die. So, 2.5 TB/s die-to-die seems... excessive. I guess there's some kind of cache unification that requires enormous throughput. Or perhaps the interconnect is designed to handle more dies in a different configuration.
Edit: After ruminating a bit, I think that Johny’s “more than 4 times the bandwidth” statement may be a hint at the next-gen interconnect/interposer.
Grabbing a napkin… Since each die can do 2.5 TB/s, and four dies would require six direct interconnects to be fully connected (3 on each die), each interconnect would handle 2.5/3 TB/s = 833 GB/s. Infinity Fabric can do 400 GB/s switched, so 800 GB/s total in a quad setup, and thus an average of 800/4 GB/s = 200 GB/s overall. 4 * 200 < 833 < 5 * 200, QED… Yeah, that’s the best I could massage the numbers.
Edit 2: I looked into Infinity Fabric some more and while it’s not switched, it’s unclear to me what the actual bandwidth would be in different configurations and my numbers above might be correct… ish. It seems like there are a lot of rules for different configurations.
Anyway, I saw a twitter post on March 8 by a Japanese reverse engineering firm TechanaLye indicating that they’d analyzed the UltraFusion region on the M1 Max. Nice die shots, but, sadly, the details would be in one of their paid reports.
Last edited:
The M1 chips scale pretty linearly... it should be twice the performance give or take a few percent.
I know it's stupid, GET SOME THING like a "Shovel" and start digging! Then in 3-5 years get a "Backhoe"
When you can actually DO SOMETHING with it! You gotta dive in a get started, don't worry about...
Excuses to keep sitting on their ass, hey I admit I do too [butt] still...
That's what's so hilarious about these posts, FOLKS now calm down calm down, guess what?
This is the M(1)NE family line of SoC hehehehe, what comes after 1? a... 2 (4nm)? exactly hehehehe just 2? what about 3 also! and 4?
Like LeBron said: 'Not 2, not 3, not 4, not 5, not 6, not 7' titles!!
LOL
I really am surprised most people still trash this (pun sorta intended?) but praise the Xbox Series X (which is basically the same concept with boxy instead of cylindrical design).
The biggest downside for Apple computing products has been them not adjusting thermal needs of their partner products from Intel/AMD/etc. If Apple controls the silicon, their designs seem to work brilliantly.
Power usage might become a more relevant metric if energy prices continue to rise due to global geo-political turmoil, or if we simply need to ration usage due to limited supply from renewable energy sources. Energy efficiency could turn in the dominant specification.
After World-War III, this might be your "battery" and "generator". Use the wind to pump water into the tank, and then release it to generate an electric current in a scaled-down hydro-electric generator.
Got a trashcan - it did well until the MacbookproM1Max. - which is of course, way faster.
The macbook is "just" a laptop. The trashcan was great design.
If the Studio had been a half height trashcan, I would have been v happy.
New Studio looks like a sandwich box.
I'll have a go at designing a stand for it - if only to get it out of the dust.
The macbook is "just" a laptop. The trashcan was great design.
If the Studio had been a half height trashcan, I would have been v happy.
New Studio looks like a sandwich box.
I'll have a go at designing a stand for it - if only to get it out of the dust.
I'd love to see someone design a clear plastic shell to lift it off the desk like the old G4 cube. It really feels like the G4 Cube reborn, and that was my favorite Mac ever.
that's my 3080ti right at the end of the power consumption acting like a unwanted space heater. I cannot wait to thrash the studio ultra and see how much [or little] heat and noise it pumps out in comparison.
That would just give you some other bottleneck (might even be the software that can't keep up) that would make that not ungodly.
My only beef with the companions is like when they compared it to the 27" iMac - they aren't exactly regularly updated with the latest and greatest internals.
It's like when the M1 Mac mini came out - well the previous Mac mini update was from like 2014 or something.
It's like when the M1 Mac mini came out - well the previous Mac mini update was from like 2014 or something.
No idea about Maya, but Blender should have Apple Silicon GPU support by way of a Metal backend quite soon.
What's super odd is that you need to be so aggressive in your response. My comments were just questions/opinions and I'm fully open to discussion. But don't really appreciate the agro of your comment. I never said this Mac would have a thermal issue. I know how much hell the trash can caused me. Apple publicly said they backed themselves into having problems. This is a super small form factor. I would worry about that happening again. Of course it's like comparing chalk and cheese as the apply silicon is the big difference, and I said I have no experience of this.
Not sure how you decided that it doesn't fit my needs, given you have no idea what my needs are. But I can tell you I utilise 192gb of ram right now in my system. Also no comparison of speeds with the pro Vega card I have in the system now.
Maybe don't make a comment at all, I was merely asking questions and voicing a concern. If there is nothing to it then fair enough, but what Is the deal with the shi**ty attitude of people commenting here. We're all allowed to speak freely.
Oh it will be ridiculous. That’s one thing Apple has over everybody else by a long shot is power consumption or power per watt. The Lovelace cards are going to be 400 W at the high-end.
M1 Ultra is super powerful. The replacement for Mac Pro with Apple Silicon should be even more powerful.