Posts: 393
Threads: 23
Joined: 2019 Oct
Thanks: 385
Given 525 thank(s) in 185 post(s)
Country:
(2020-06-08, 12:54 AM)LucasGodzilla Wrote: Out of curiosity, do you recall any notable color differences between 35mm and the LD master from off the top of your head? I've always wondered exactly how blue the movie's supposed to look as a whole given the director's penchant for it (though we'll all find out soon enough when the proper scans come in, and that teased frame of the trucks is already a sign).
I can't say I noticed any dramatic differences, but then I'll admit that I don't know the film well enough to spot things like that. I seem to remember the colours looking pretty natural.
Program material is recorded on the other side of this disc...
Posts: 44
Threads: 1
Joined: 2020 Jun
Thanks: 62
Given 36 thank(s) in 20 post(s)
Country:
Preview image looks good!
deleted user
Unregistered
Thanks:
Given thank(s) in post(s)
Crazy what a difference in colors there is. They completely ditched the grading on the LD it seems.
Keep in mind these are likely uncalibrated colors. This will get even better, color-wise. Noise, we'll have to see. But in the earlier scans there was also issues with flare coming from the sides which luckily seems to have been pretty much solved.
Posts: 2,049
Threads: 56
Joined: 2016 Dec
Thanks: 161
Given 1008 thank(s) in 612 post(s)
If the scan is uncalibrated how do you go about recovering the actual colours on the print? Do you get a LUT from the scanner
deleted user
Unregistered
Thanks:
Given thank(s) in post(s)
@
zoidberg
You scan a color target, which has reference colors. You then just have a tool calculate your calibration matrix and apply that to the data. Gotta scan it with exactly the same settings on the scanner as the film itself, meaning white balance, light, etc. Davinci can do it directly I think, at least for the Color Checker target, you just drag the target values mask over the target and let it calculate the correction. This pretty much does everything that used to be a problem with the initial scans - increase saturation, while simultaneously correcting reds and greens so they dont all look orange-y (depends on light source too ofc), while at the same time not leading to excessive saturation in other places.
A bit of trivia, many digital sensors seem to have a strong overlap of the green and red channels, which means that in the uncorrected raw data red and green bleed together and thus everything is orange-y and low saturation, while the blues tend to still look somewhat okay.
Posts: 393
Threads: 23
Joined: 2019 Oct
Thanks: 385
Given 525 thank(s) in 185 post(s)
Country:
Looks great! A really nice clean print. Can't wait to see this in motion!
Program material is recorded on the other side of this disc...
Posts: 2,049
Threads: 56
Joined: 2016 Dec
Thanks: 161
Given 1008 thank(s) in 612 post(s)
Thanks for your explanation Tom. So what colour target is the scanner using?