Can’t say have not been using them for a bit, but I do check the daily build changes from time to time and I would guess no.
Good question - my Mac studio work buys.
The max mini I have only suffers from 8 gb ram.
If the rumors are right and they will offer an M2 pro max mini that then would be quite nice
Sorry, but I have to take offence here. The only reason you get to use Blender on MacOS is because of the Open Source movement. MacOS in turn is a fork of BSD, which is open source and to which Apple no longer contributes back to in any meaningful way. You would be surprised how much of the CG industry runs, develops and renders on open source OSs, certainly a much larger proportion than is reflected by Linux desktop usage.
Apple has a specific goal - and thus supports Blender in actually the most effective way.
I don’t see where the issue here is or am I missing something ?
And they mostly do that to look user into a secure linux world, where the user can’t do much outside the apps he is allowed to use. Keep in mimd no one at the larger faculties has a webbrowser or even is allowed to bring his phone.
just a nice anecdotal experience:
I’m rendering a large environment on my M1 Max. its a preview so its a low sample count. I fired up my 2080 tower to help out. rendering the exact same project/settings, the M1 Max is about 25% faster. I think the difference is in the bvh build time. likely on higher sample counts the 2080 would overtake, but its still a nice example of a place where the MacBook Pro shines. also this was using cuda on the 2080 because optix was straight out of memory.
Sure is interesting because theoretically or going by blender open data the 2080 should be twice as fast even on Cuda.
Also should Optic not have worked, I thought Blender would support out of core rendering?
Ah found it official doc has this:
With CUDA, OptiX, HIP and Metal devices, if the GPU memory is full Blender will automatically try to use system memory. This has a performance impact, but will usually still result in a faster render than using CPU rendering.
Could also explain why it is slower.
I had this effect on the M1 Macs too.
A scene that was very slow on the air (that I used to have) render with way less issues on the M1 Pro with 32 GB and I am fairly certain it was the memory difference and not the GPU power.
Definitely. It would allow MagicaCSG to run on macOS. ![]()
Now keyshot is also ready for apple silicon!
Thanks. I’ll ask the MagicaCSG developer if this would make a macOS port easier.
Nice. Keyshot is a fine renderer, but I really dislike their recent subscription-only switch.
In don’t allow my students to use crapshot
Das ist verboten!
But is shows how the M1 adoption in the industry grows
it makes me really wish the 64gb model had been in stock. even my max will render happily for a couple hundred frames then suddenly give an out of memory error in the console. I can see in my usage monitors its already going out of core, using 50gb when I only have 32gb.
Yes wish Marmoset would support Apple Silicon but they have been fairly quiet about it.
Makes me wonder if they have a bug, should it not free up some space after finishing the frames ?
Or does complexity change after the 100 frames?
its a random issue that seems to always happens over time. I wouldn’t be surprised if its related to persistent data.
Good idea ![]()
By the way have you seen this?
I did not test it yet.
I have asked them more than once in their Facebook Beta Tester group and they always tell me they are not working on a native port because the rosetta version runs fine. Such a frustrating answer.
Yes I have seen similar answers which I think sucks. Good thing not every company thinks like that.
Perhaps their Mac user base is to small or they just don’t care and the day when there is no more Rosetta there is no more Marmoset.