Thank you for your help @ggbutcher. I read your article and a few others. I understand your point about RAW files. However I have only access to TIFF files. My problem can be separated in two questions:
- How can we import a TIFF file with correct L* a* b* values knowing the color profile, either embedded or not?
- In the particular case of my optical microscope, I need to create an icc or dcp profile and I suspect there are some errors in my process, especially due to glare. Could you check?
Update: I just upgraded from RawTherapee 5.9 to 5.10 under Windows 11 and L* a* b* values are very different, using the same process. Has color management changed under the hood? Nevertheless using Adobe abstract profile still degrades the color accuracy obtained with none or sRGB, which seems unexpected for me.
- A1 dark skin: RT_5.9 gives 36.1/11.3/13.4 (ΔE_2000 = 2.6) → RT_5.10 gives 37.5/14.7/14.6 (ΔE_2000 = 0.4), expected values are 37.54/14.37/14.92;
- F3 cyan: RT_5.9 gives 52.0/-18.3/-24.6 (ΔE_2000 = 5.8) → RT_5.10 gives 49.7/-29.8/-28.2 (ΔE_2000 = 0.2), expected values are 49.57/-29.71/-28.32
@Lawrence37 @martbetz @jdc, your insight will be appreciated.
Here are the screenshots.
Regards,
