Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Interesting. But there is also this from WCCFTECH:

MacBook Pro With M1 Max Is Barely Faster Than Microsoft’s Surface Laptop Studio in 8K Video Render Test

That's about the 8K video rendering test which was performed on Adobe Premiere Pro (it's available as AS native).

Does Premiere use the GPU to render, though, or is rendering just GPU accelerated? Not a rhetorical question, I've never used it. But from what I've read, Premiere isn't a GPU renderer.
 
So basically the only way to get the best results is to use apps optimized for the chip.
Give or take 2-3 years and everything should be like that.
It is pretty impressive what apple did here.
 
Does Premiere use the GPU to render, though, or is rendering just GPU accelerated? Not a rhetorical question, I've never used it. But from what I've read, Premiere isn't a GPU renderer.
I find it amusing that the person who sent the tweet thought Apple was over hyping the GPU, while other video editors perform significantly better. lol.
 
So basically the only way to get the best results is to use apps optimized for the chip.
Give or take 2-3 years and everything should be like that.
It is pretty impressive what apple did here.
This is unfortunately the truth yes. My worry is that we may never actually get there.
 
  • Like
  • Love
Reactions: g75d3 and Mr.PT
This is unfortunately the truth yes. My worry is that we may never actually get there.
You can already replace Adobe Photoshop, InDesign and Illustrator with Affinity Photo, Publisher and Designer.

If you need to do video editing there is DaVinci Resolve and Final Cut Pro.

That covers a lot of content creators if they are willing to learn new applications.
 
Imagine 2-4 M1 Max 64GB in an iMac Pro with 500W PSU.
  • 20-40 core CPU
  • 16-32 performance cores
  • 4-8 efficiency cores
  • 64-128 core GPU
  • 32-64 core Neural Engine
  • 800-1,600GB/s memory bandwidth
  • 128-256GB unified memory
 
Last edited:
What I personally find impressive that Overwatch is getting 120FPS while inside of a VM AND going through Microsoft's bad translation layer lmao. Imagine if we could bring this down one layer and/or just get native.

Not bad but doesn't tell the full story without resolution and quality settings. Could be 720p low for all we know plus it's one of those games like CSGO that gets crazy high fps.
 
No you can’t unless you don’t get files from clients.

Unfortunately you can’t really import those files into the Affinity apps.
I guess it depends on the complexity. I’ve had good luck importing into Affinity in the past, but exporting back to PSD is usually where things go haywire. That said, it was more for sharing among a community effort rather than as a paying job (one reason I went with Affinity instead of Photoshop, I use it for hobby, not work).
 
No you can’t unless you don’t get files from clients.

Unfortunately you can’t really import those files into the Affinity apps.

Yep. It's like Google. Even if YOU decide to move away from Gmail (for example) that doesn't stop the rest of the world from using it which means a lot of what you write ends up on Google's servers anyway.

Likewise, moving to Affinity doesn't stop people from sending you PSD files, etc.
 
Not bad but doesn't tell the full story without resolution and quality settings. Could be 720p low for all we know plus it's one of those games like CSGO that gets crazy high fps.
Yep definitely true and its partly why I don't watch too many videos from Dave2D. Just too sparse on the details.
 
I'm not sure how much it'll matter for native ports. I'm rather doubtful that the industry will suddenly switch to Apple and what not. What I am more excited for is something like Crossover/Parallels. I really would like to see how those work. The M1 with either of those works incredibly well for what the M1 is.
The only real advantage the MacOS market has for game ports right now is the lower competition. Sure the overall marketshare is smaller, but you don't have to compete with as many titles to get sales.

Crossover/Parallel gaming is not a thing unless it is super casual gaming (like Tetris or Minesweeper) that does not depend on GPU/CPU. The M1 would have to translate not only x86 code to ARM, but also manage Windows APIs to Mac APIs. That kind of performance hit would be 4x slower at least than a native port of that same game to MacOS.
 
  • Like
Reactions: luvmango
The only real advantage the MacOS market has for game ports right now is the lower competition. Sure the overall marketshare is smaller, but you don't have to compete with as many titles to get sales.

Crossover/Parallel gaming is not a thing unless it is super casual gaming (like Tetris or Minesweeper) that does not depend on GPU/CPU. The M1 would have to translate not only x86 code to ARM, but also manage Windows APIs to Mac APIs. That kind of performance hit would be 4x slower at least than a native port of that same game to MacOS.
Uh. Crossover and Parallels actually worked really quite well for the M1! Here's a good video for the performance of some games. Also, you can do Steam via Crossover now which is really neat!
 
  • Like
Reactions: altaic
That I agree with. If Apple want gaming on the Mac, they need a big push to get those developers to see the potential. Maybe they will. Who knows.
Personally I could not care less about gaming, but I desperately need a lot of software I would like to work with to be put out of the CUDA spell. That will only happen when Apple quits its aside posing and Really put some serious effort im making things happen…until then overpowered MBP for some use cases?
 
So basically the only way to get the best results is to use apps optimized for the chip.
Yes. This actually takes me back to the 1980s when every computer was different. You had a game for Commodore Amiga but wanted it on the Atari ST? Well, rewrite it because the hardware is different.

The scene has changed quite a lot, but under the hood, this remained partly true. To run the same game on Windows and iOS is a rewrite. But even on Windows, using the same APIs, you'll notice some graphics cards are just not as well supported in some situations as others. I'm thinking of AMD under AVID or Premiere versus NVIDIA.

With this move to Apple Silicon, we're basically taking a step back for cross-everything compatibility. But for a computer geek like me, it's exiting times. I love diversity of architectures!
 
  • Like
Reactions: JMacHack
Does Premiere use the GPU to render, though, or is rendering just GPU accelerated? Not a rhetorical question, I've never used it. But from what I've read, Premiere isn't a GPU renderer.

Who knows. It wouldn't surprised me if they exported to a codec that is GPU accelerated on Windows but not working for Metal on Apple Silicon (or working very poorly).

Another thing to keep in mind: If you are using "old" Intel-complied plugins for Premiere Pro, you will probably need to run it via Rosetta 2. At least that is the case for InDesign which is also ready for Apple Silicon. Devil in the details, as always. ;)
 
Imagine 2-4 M1 Max 64GB in an iMac Pro with 500W PSU.
  • 20-40 core CPU
  • 16-32 performance cores
  • 4-8 efficiency cores
  • 64-128 core GPU
  • 32-64 core Neural Engine
  • 800-1,600GB/s memory bandwidth
  • 128-256GB unified memory
dont know there were rumors about that...the imac pro could be a similar case like the 24" that will take the m1 pro and m1 max from the laptops
 
dont know there were rumors about that...the imac pro could be a similar case like the 24" that will take the m1 pro and m1 max from the laptops
That originates from my imagination. Aren't all rumors just a figment of the mind?

My idea is grounded on supply chain & economies of scale.

Like, would it be profitable enough for Apple to create a 1 or 2 Apple Silicon chip SKUs for Pro Desktops that ship less than 1 million units per year?
 
That originates from my imagination. Aren't all rumors just a figment of the mind?

My idea is grounded on supply chain & economies of scale.

Like, would it be profitable enough for Apple to create a 1 or 2 Apple Silicon chip SKUs for Pro Desktops that ship less than 1 million units per year?
Apple will always seeling you the same chip for multiple products...see M1
The only mac that will scale more than 1xM1 pro or 1x M1 max will be the mac pro i think
The bigger imac and the mac mini next year will came with the same config as the new macbook pro...
dont get me wrong,,i want in a bigger chassis that is not limited by battery, with an 250W drawing..to have what you said, i would take 64 core gpu in a heartbeat
 
Apple will always seeling you the same chip for multiple products...see M1
The only mac that will scale more than 1xM1 pro or 1x M1 max will be the mac pro i think
The bigger imac and the mac mini next year will came with the same config as the new macbook pro...
dont get me wrong,,i want in a bigger chassis that is not limited by battery, with an 250W drawing..to have what you said, i would take 64 core gpu in a heartbeat
That's the reason why I was thinking of multiple M1 Max 64GB die on a multi SoC package

Or

Multi M1 Max 64GB SoC onto a logic board that accepts multi SoC.

Reusing current designs for Pro Desktops with PSUs historically at 300W (iMac 27") 500W (iMac Pro) & 1,400W (Mac Pro).

We will know of this in 52 weeks time.
 
I think Affinity devs said it very nicely: "the 'old rules' no longer apply". When developing for Apple Silicon you don't have to worry about CPU/GPU communication latency or bandwidth, GPU memory limitations, GPU memory management... everything is radically simplified, virtual memory works as expected, you don't need to jump though weird hoops with texture streaming...

Expect much faster software and games, fewer bugs, quicker and cheaper development. Right now the situation is still suboptimal since one has to develop for both Intel and ARM Macs, but few years from now when the Intel Mac support can be safely dropped, Mac will be the simplest platform to develop GPU software for.
 
Crossover/Parallel gaming is not a thing unless it is super casual gaming (like Tetris or Minesweeper) that does not depend on GPU/CPU. The M1 would have to translate not only x86 code to ARM, but also manage Windows APIs to Mac APIs.
That's just not accurate. While virtualization gaming is inherently more limited, Parallels does run a few sophisticated games:

You'll have to sacrifice some of the graphical quality though with the original M1. I don't know how it is yet for the M1 Pro and the M1 Pro Max.

Crossover has much better speed, but lower compatibility.
 
I think Affinity devs said it very nicely: "the 'old rules' no longer apply". When developing for Apple Silicon you don't have to worry about CPU/GPU communication latency or bandwidth, GPU memory limitations, GPU memory management... everything is radically simplified, virtual memory works as expected, you don't need to jump though weird hoops with texture streaming...

Expect much faster software and games, fewer bugs, quicker and cheaper development. Right now the situation is still suboptimal since one has to develop for both Intel and ARM Macs, but few years from now when the Intel Mac support can be safely dropped, Mac will be the simplest platform to develop GPU software for.
yep, since Affinity chief is Jony Ives brother ...they really know how to do it and this is the pure example
 
I just got my 16" M1 Max, and I've recently been coding a simple little dissolve / Thanos-snap dust style particle system in Metal so I airdropped the project from my M1 MacBook Air and compiled it in Xcode on the M1 Max....

It runs at 60fps on the M1 air, and .... uh ... 52 fps on the M1 Max. What!?

Ok so I'm probably bottlenecked by single core, I'll try putting it onto my Pro Display XDR at full 6016x3384 resolution..... 40 fps now on the M1 Max, so I guess now it's taxing the GPU a little bit. Plug it back into my 8gb M1 Air..... err 60 fps solid at 6k.

These projects are absolutely identical, and both projects are running in release with the same optimisations.

I've tried enabling high performance mode too, still getting 20 fps less on the M1 Max at 6k than on the little MacBook Air.

This is incredibly confusing....
 
I just got my 16" M1 Max, and I've recently been coding a simple little dissolve / Thanos-snap dust style particle system in Metal so I airdropped the project from my M1 MacBook Air and compiled it in Xcode on the M1 Max....

It runs at 60fps on the M1 air, and .... uh ... 52 fps on the M1 Max. What!?

Ok so I'm probably bottlenecked by single core, I'll try putting it onto my Pro Display XDR at full 6016x3384 resolution..... 40 fps now on the M1 Max, so I guess now it's taxing the GPU a little bit. Plug it back into my 8gb M1 Air..... err 60 fps solid at 6k.

These projects are absolutely identical, and both projects are running in release with the same optimisations.

I've tried enabling high performance mode too, still getting 20 fps less on the M1 Max at 6k than on the little MacBook Air.

This is incredibly confusing....

This does not bode well. Could the M1 Max be thermal throttling too early (which limits performance)?
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.