Deliberately difficult gamut colors in macro shot of hex nut

Makes me wonder if there is a comprehensive ‘textbook’ for in to the weeds color theory.

@Waveluke Google Scholar? Found a few good books that way. Trouble is finding a physical or digital copy to read.

Yeah, the strange thing is any books I find just by googling tend to seem more geared for casual photographers and artists than what I want.

I have the opposite experience. I find advanced books that no one carries and when I look at the prices direct from the publishers I see $$$ which is a nonstarter.

Any recommendations? I am looking to gain a solid grasp of how colorspace transforms are done to the mathematical details, understand color management pipelines from raw, and be able to understand what is going on inside a DCP/ICC profile and have all the more advanced color theory explained in a more comprehensive manner than various articles and discussions here on PIXLS.

I have seen some recommendations scattered across the forum. Perhaps start a thread to collect them in one place.

1 Like

They came out before digital so they don’t deal with raw pipelines (see the DNG spec for that) but for understanding color science I can recommend two somewhat overlapping books:

Measuring Color - RWG Hunt
Visual Perception - TN Cornsweet

This is fairly timeless stuff, so I typically buy such books used. Plus the older they are, the easier they are to understand because the science wasn’t as evolved :slight_smile: If I had to choose just one I would go for Hunt, I have the third edition.

1 Like

I have just exported the hex nut from DT using 3 different intents: relative colorimetric, absolute and perceptual.

Workflow was
Input profile: standard color matrix
Working space: linear rec 2020
Output profile: srgb

Only other modules used, same settings for all, were Raw black/white points, white balance, orientation and demosaic.

All three look exactly the same. That’s not what I was expecting…

Edit: Tried use Lcms2 to apply output profile in ‘processing’ tab of settings both ticked and unticked.
And in case this is a build thing instead of a processing thing, I’m on darktable 3.2.1, mac 10.12.6. (And according to color sync my display profile is ever so slightly bigger than srgb, including in the blues, so that shouldn’t be the issue). Can’t wait to get Linux…

Gamut indicator for rec 2020 working space is showing OOG background (with fringe in gamut).

Edit: For good measure I downloaded a different srgb profile with perceptual intent and put it as output… same result. I even try to use that profile as working space… same result.

Whoa!

image

2 Likes

This exercise will help with your understanding or maybe make it worse :slight_smile: Developing a RAW photo file 'by hand' - Part 1

Do you have little cms enabled in preferences?? Otherwise I don’t think rendering intent works… you must though or it would not be an option…Might be worth checking

As mentioned earlier, perceptual intents are uber-subjective. In fact many editors (including PS) skip them altogether and, if selected, give you relative or absolute instead. Or at least that’s the way it was the last time I looked.

1 Like

Your input profile has to support the intended rendering intent. Your ‘standard color matrix’ won’t do perceptual. That’s what the LUT profile is for… Go back and read the Agyll gamut mapping page…

LittleCMS will fall back to a rendering intent appropriate to the input profile; checking and enforcing something is up to the application developer. I recently had to put that code into rawproc…

1 Like

Try duplicating color balance 1, which produces CB3, and then deleting CB1.

DSC08363_05.ARW.xmp (43.4 KB) dt3,3

Interesting! Thanks!
Yes, I see it works, CB3 behaves properly. But why??!
Have you found duplicating modules sorts out gremlins previously, maybe?
Seems like a bug surely?

Found it by accident. I was going to move CB1 above Filmic, but instead of moving it, I copied it so I could turn off CB1 and move the new one. When I turned off CB1 SHAZAM!!! Fixed.

Yes, some kind of bug because as far as I could see when I started investigating, you had it set up correctly to do what you wanted.

Wow:

What is new under the sun?! Not so nice people hating nice people.

1 Like

I just bought it anyway :slight_smile:

Did you get any further with this? I’ve also noticed the lack of difference between the 3 rendering intents, at least using sRGB (websafe). I started using the sRGB v4 Preference Perceptual Intent (with LCMS in preferences) from here -

(International Color Consortium site)

Here is a checker shot with the saturation ramped up showing a difference between perceptual and relative colorimetric as set in Output Color Profile module.