White (dis)balance and Darktable

[sgotti (Simone Gotti)]

@bastibe I also own a fuji. And the green cast when using daylight is due to a non perfect matrix (the adobe one).

A good fix is to use what is described in the manual darktable 4.2 user manual - color calibration.

@bastibe

This is easy enough to work around with a CC preset for “as shot”, which solves the issue for me.

When you use as shot you are extracting hue and chroma of the illuminant from the image exif. You can also fix them with custom. Basically you’re removing the greenish caused by the not perfect d65 matrix.

@bastibe

Frankly, though, it is simply highlighting the need for an old-school temp/tint slider. I know temp/tint not “correct” because tint is a lie, but it is a useful lie, at least for daylight-adjacent illuminants.

This will restart the discussion of the original thread where AP explained quite well the reasons of not using tint. Just use hue and chroma with custom if you want to manually set the illuminant color. That thread is a really good read.

As a side note one of the more interesting comment in the original thread is this:

Introducing color calibration module (formerly known as channel mixer rgb)

All I can say is your problematic sample is very peculiar. The light source is suspicious in the first place, it’s around 2400 K with seemingly low color reproduction index. Since I don’t know anything about the conditions of the shot, it’s hard for me to conclude where the failure is. For cases like this, the white balance module is more robust. As I said above, more robust usually means less accurate, but predictably inaccurate. To solve this kind of issue, I guess we will have to perform c…

Using spectral sensitivites instead of matrixes that are also non perfect will help to obtain really good color reproduction. The dual illuminant DCP profile are a step forward but with many issues (see dcamprof neutral operator implementation to obtain a real neutral dcp profile, adobe dcp profiles, that like the matrices extracted by it, have some subjective look and aren’t neutral profiles. For example I heard somewhere that in lightroom Fuji cameras has too much magenta. Obviously it’s not caused by the camera (raw files don’t have colors…). it’s just how they are profiled…).

Vkdt has some support for spectral profiles in the colour module but looks like you have to manually set the CCT, so I’m not sure how it works and how it could be ported to darktable.