FYI: many online tests and test of my own work in FCP have shown that when exporting rendered ProRes or HEVC even in 8K the Studio with the ultra chip (48C in my case) is no faster than my 32 core M1 max. The Ultra definitely has a damages with GPU heavy effects and multiple clips (mostly in 8K; it is able to play multiple 8K HEVC. clips with effects in best performance at a... usable speed, the M1 max can't). This means timelines with many 3D effects and so forth by I see a 20-30% increase in rendering speed and exports of unnrendeed timeline. But as far as exporting already rendered ProRes video… there's really no difference. The transcode speeds are about the same... maybe 10% faster in some cases?
Practically speaking this means the $2000 Max studio is nearly as fast as the $4000 version. And I've tested both versions. Practically speaking for FCP use currently they're about identical unless you use a lot of effects. And I mean heavy like 3D work and so on that lean on the GPU.
The ultra is also much faster with GPU heavy loads like raw video decoding from red and black magic. But for most compressed workflows… It's only like one set of media engines are being used from one of the two dies.
"Maxtech" on YouTube indicated that Apple said they're going to improve this transcoding speed with Ventura to allow both sets of media engines in the Ultra to work. It's pretty clear empirically that only one set of media engines is working now on the ultra chip. Well I tried the latest public beta of Ventura, and exported a bunch of things from Final Cut Pro. The export times were virtually identical between Ventura and Monterey, which means they were very close to the same as the M1 Max 32 core.
Perhaps this will change in later betas but right now… There's no difference.
Practically speaking this means the $2000 Max studio is nearly as fast as the $4000 version. And I've tested both versions. Practically speaking for FCP use currently they're about identical unless you use a lot of effects. And I mean heavy like 3D work and so on that lean on the GPU.
The ultra is also much faster with GPU heavy loads like raw video decoding from red and black magic. But for most compressed workflows… It's only like one set of media engines are being used from one of the two dies.
"Maxtech" on YouTube indicated that Apple said they're going to improve this transcoding speed with Ventura to allow both sets of media engines in the Ultra to work. It's pretty clear empirically that only one set of media engines is working now on the ultra chip. Well I tried the latest public beta of Ventura, and exported a bunch of things from Final Cut Pro. The export times were virtually identical between Ventura and Monterey, which means they were very close to the same as the M1 Max 32 core.
Perhaps this will change in later betas but right now… There's no difference.
