Jake Blackstone
Forum Replies Created
-
Unfortunately no. Smoke doesn’t really use cache. It’s actually a render, just like FCP. And Scratch implementation of cache is as bad as Resolve. I’d like for BM to strive to be the best of the bread, not just “good enough”.
-
No Mike. What you’re describing is optical timing. I was talking about Log grading used in color grading and DI. Traditional Log grading ALWAYS was done with TWO trackballs- Brightness/Exposure and Contrast. See all other grading platforms, like Baselight, Lustre and FilmMaster. The third trackball is always assignable to one of the traditional three way grading controls. By your logic, why not use FIVE trackballs? More is better, right? And no, I don’t have a problem with four trackballs. My issue is with the whole Log grading implementation. Affected preset ranges are way to narrow, adjusting ranges until they intersect causes one of the controls to stop working altogether and, if one is not careful, it is very easy to induce noise and create “solarization” effect.
-
Again, BM tries, unsuccessfully I might add, reinvent the wheel. Traditionally LOG grading is done with TWO trackballs. I’m not aware of any other grading systems, that uses FOUR trackballs. Existing THREE trackballs are plenty enough for proper LOG operation. Needless to say, number of trackballs has nothing to do with BM ‘s desire NOT to allow LOG operation on any panel beside the BM panel.
-
I actually want nothing to do with the way cache is implemented in Resolve, even if did work. What I’d like is the implementation of cache in FilmMaster and to a lesser degree. Baselight, where you can cache debayered material, as well as apply REAL noise reduction, REAL sharpening etc. Once those operations performed, there is almost never a need to re-do those operations. So, when I operate FilmMaster, I simply apply it all and go for a walk, a coffee etc. When i’m back, it’s done and I’m on my way grading Red material in FIULL resolution with no Red Rocket! And if that wasn’t enough. I can cache pretty much any layer, so only the new layer needs to be calculated. Doing it this way dramatically increases the speed of render (in this case it is pretty much real time and it is always in the background). If that is not enough, you can use distributed renderer, based on cheap Linux computers. And the best part, i don’t have to worry about fragmenting my drives. I can set each session the time frame for the cache expiration. All I do is once a week or so I run defrag on my drives.
And that’s how it’s done! -
Unfortunately, it has always been this way from day 1 of support MC Color. And yeah, it’s painful…
-
Thanks Peter for a quick reply. Right now I’m in the middle of a feature, but once it’s done I’ll give it a try.
-
Oh yeah, it’s Alexa Prores.
-
I checked both ways of setting “Copy Grade: Preserve Source Settings” and they both work exactly the same.
-
Yes, it does work the way I had asked. You need to follow Peter’s suggestion and uncheck the “Copy Grade: Preserve Source Settings” option in the PREFERENCES:-) I believe, that unchecking “Copy Grade: Preserve Source Settings” option it in the gallery affects only that event.
-
Get GT120 and GTX 570