Replies: 2 comments 1 reply
-
I've heard performance on macs decreases a lot when memory is limited. You could try a mix of the low ram/vram command line arguments listed on the wiki. |
Beta Was this translation helpful? Give feedback.
-
Yes, Mac is an issue as it is totally different hardware from everything else these days and very few actually consider Macs seriously when developing. The thing is that the GPU can only be used for limited tasks and Mac will always load everything to memory. It's not very good at handling this. There's also the issue that on new Macs, you will never be able to use CUDA and if there would ever be an implementation, it can only come from Apple since others have very limited access to building anything connecting hardware with software. So the answer for Mac users who want to use this seriously is, get compatible equipment. Honestly, and it's not that I hate Apple, but, nothing much good ever came out of that company. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
When trying to generate a batch of 6 large (1024X1536) img2img, it takes me up to 6 hours on my Mac Pro Max 32GB vs 10 minutes on free google colab for the same job.
Is this normal?
What types of performance should I expect on the mac?
Beta Was this translation helpful? Give feedback.
All reactions