There is a (small) difference in the light around the rising sun between the log-tonemapper of ART and the filmic and Sigmoid module of Darktable. Can the yellow rings around the rising sun also be solved in ART?
Hallo, try to precompress the highlights (Voorcompressie hoge lichten), does that change anything?
Unfortunately, that only reduces the highlights, but the yellow rings remain (saturation of the red pixels causing green and blue to appear yellow)
Most likely yes. I’m not at the computer right now, but try enabling the tone curve in neutral mode and make sure it’s not an identity (e.g. just set the contrast to 1) and see if that helps. If it doesn’t, I’ll give it a try later today or tomorrow.
Ps. In general, it might be useful to consider that art is not darktable, they work in different ways and there’s no 1:1 correspondence between modules
20230904-170-logtone mapper.jpg.out.arp (11.0 KB)
20230904-170-tonecurve.jpg.out.arp (11.0 KB)
20230904-170.rw2 (18.6 MB)
Apparently the tone curve should always be used. But there is a clear difference between the log tone mapper and filmic or Sigmoid.
But the result is great!
They are different indeed.
BTW, in this case you can try to change the highlight reconstruction algorithm to “Color Propagation” and see if that helps with avoiding the yellow rings
Color Propagation certainly gives the desired result. I did some more experiments with bright light from a beautiful sunrise. The (raw) photos are still unedited, only a comparison between the Log tonemapper (ART) and Simoid/Filmic (DT).
The left two are with the log tone mapper, the right one with Sigmoid. The middle one is with color recovery and is close to Sigmoid, with a little more yellow. I think the middle one with the log tone mapper most closely resembles what the eye can see (but that is not possible with such a bright sun 
)
1 Like