There are certain image generation tasks that I cannot try out without installing Sonoma — for example, use the SDXL model natively via CoreML in a Swift app, at least as far as I’m aware. (Though I must confess that I didn’t look into this too deeply …)

So I’ve been wondering about installing the Sonoma beta. But since this would be on my regular production machine (and I do development for regular work), I wasn’t really keen on the possible downtime should something go wrong or some major app I relied on didn’t work correctly on the Sonoma beta.

Beta 5 was the target that I’d set for myself. By beta 5, we should be close enough to release, and hopefully fairly stable, that I should be OK was my thinking.

And beta 5 is here now …

But I’m not sure I want to switch. Just too much going on at the moment …

So I guess I’ll keep dithering? 😛

#Apple #macOS #Sonoma #Beta

@f You can run sonoma in a VM with a virtualized GPU. UTM or virtual buddy are good options. Sure it would be a bit slower but might get you a preview of the software.

@trode I’m already doing that 🙂 But it doesn’t work too well for GPU intensive stuff like running Stable Diffusion, unfortunately …

@f Just slow or explody? I havn't tried it yet. Been trying not to burn time playing with Apple Betas this cycle.

@f I just rolled Sonoma beta on work machine. Stable enough. YOLO!

@trode Just slow — it doesn’t crash, but Stable Diffusion processing can be agonisingly slow without native GPU support …

@natik Good to know 🙂 Maybe I’ll take the plunge over the weekend … Thanks!

@f Ah the virtual GPU doesnt work sounds like