For some reason our PCs bake occlusion very slowly. Our Macs that have inferior specs bake them a lot faster.
Looking at our processes in Windows it looks like the umbra process doesn’t use over 25% of the CPU at any time… is there any way to just let it run fully?
EDIT: (Beast uses 99%)
on osx the osx own thread handling splits its over the cores.
on windows it must be manually done or it will end on a single core (-> 25% on a quad core as you see it) which seems to be the problem, no parallel implementation present, just pure os fallback
I can’t figure out how to set it manually though. Was browsing around a little and it mentioned setting affinity but it’s already set so I have no idea 
you can not do it.
its something that ut has to do on the implementation, otherwise nothing will happen as a single thread on windows will run on a single core only
Oooh, I see.
Sigh… guess I will have to live with it. Heh.
Maybe there’s some vote thing going on for it? Perhaps it could be changed for 3.1 to make use of all available cores.
Can’t find any… this together with queued bakes is kind of essential imo… time not spent designing is very very bad.