2020-11-18, 11:57 PM
(2020-11-18, 11:27 PM)zoidberg Wrote: Without a screening of the print for reference it's just guessing really, I don't think the LUT is 100% accurate
Yeah, that is true. It's why I tried doing that sort of universal grade for the Open-Matte Grindhouse release but it led to it looking really strange in some places, and I think I made it look too green and used an old scan RGB channel adjustment process which has been superseded.
For the record btw, no LUTs are involved. What I ended up doing this time (out of using the new scan color adjustment) is essentially doing a reel-by-reel white balance using little spots (that are cropped out on the open-matte) on the side of the reel that have consistently stayed the same color throughout. I just made an averaged out value from them per reel, set that as the white-point, and then adjusted the gain to a certain number so that the brightness stays consistent throughout the whole scan. With these grade changes done, I skimmed through it all, and, again, suddenly the movie feels like it has a somewhat consistent color tone. I also checked it against the few bits of white text I have in the movies and they all seemed okay.
I do plan on doing some more testing with some RAW DNG files for Big Trouble In Little China sometime in the future to be sure this process works since I have yet to test how it looks with the colors from a different movie.
Although I dearly wish I could see how the movie looked when it was projected back in the day when it came fresh from a lab, let alone a print of the movie projected in general, that opportunity won't likely come for me for a while. So until that day comes (or the day another print shows up that can be scanned for cheap), this is the best we got to determine the intended color timing.
![[Image: ivwz24G.jpg]](https://i.imgur.com/ivwz24G.jpg)