Wrong red color x-trans (X-E1)???

(Antonio López) #1

Hi everybody,

I have noticed that when I process RAF pictures (X-E1) with red areas/things/clothes/etc. the final JPEG pictured shows a red color but with purple/pink hues.

I have been checking different options in Rawtherapee but I haven’t found the right one to achieve the correct red color.
I have upload a picture picture processed with RT 5.3 (stable): DSCF1552_RT
I have also upload a JPEG obtained directly from the camera: DSCF1552_Camera
And finally I have upload the original RAF and the ppt3 file for your review and checking if necessary:

It seems that the colored highlight areas are filled with a wrong color?
I noticed that issue on christmas when I was shooting children disguised in Santa Claus. If you unselect the “base table” at DCP Iluminat area it seems that the problem it solved, however that affects the rest of the colors of the picture and they look desaturated.

RT is a very powerful tool but sometimes is complex for amateurs as me :slightly_smiling_face:, your help will be much appreciated.

Kind regards,

(Alberto) #2

I agree that the colours are different, but I don’t see any particular issue with the RT output – is just a different shade of red. can you be more specific about what you mean?
if you want to match the camera, you can do so by using an appropriate tone curve and then shift the red towards yellow a bit, e.g. with the HH equaliser

(Morgan Hardwood) #3

A DCP pulls and stretches colors to get them as close to reference values as possible, to minimize errors and to make transitions smooth. Sometimes the functioning of our human vision and human expectations is not consistent with a color math implementation. With this DCP, there seems to be a tendency for reds to shift more toward magenta than toward yellow, which is not wrong but our human expectations notice that as being odd. This is something I will look into more in the long run. For now, if you find that you don’t like a shade of red, you can change it very simply using the L*a*b* HH curve:

(Antonio López) #4

Hi Alberto and Morgan,
Thank you very much for your comments and explanations.
I understand this issue is more a “feeling” than a real trouble. As Morgan has explained I expected a “pure red” colors in my Santa Claus/Christmas pictures, in fact I had the same sensation when I shooted red buses and phone booths during my last trip to London.
Anyway the solution proposed (Lab* HH curve) it is really easy and very quick, I definitely will use this tool in the future.
Thanks again for your support!