On par performance with a gtx 580 on windows? Really? Is that tested!? That’s surprising as it has many more cuda cores, is the core clock speed really an issue? I’m surprised!
WOW, 250 days before Christmas! A lot on my wish list here!
One question though, do any of the improvements help with HSL selection while working in log space? I find the HSL keyed to be optimised for linear work.
The method still works though as long as you
Only do step one, to get linking back you can use groups instead. So batch copy the old conform, import new conform and everything should have the correct grade version but alas won’t be linked to the same takes.