Why not absolute colorimetric rendering?

Discussion in 'Digital Darkroom' started by raczoliver, Dec 17, 2021.

  1. I just started making my own prints, and have been reading up on color management and rendering intents. I'm certainly a beginner at this (not so much the picture-taking part), but the topic is perhaps not as mysterious as it appeared a few weeks ago.

    When I read about rendering intents, all the sources seem to say that photographers should really only be concerned with relative colorimetric and perceptual. Some online posts on the topic don't even bother describing the absolute colorimetric rendering, they simply say you should never choose that as a photographer.

    My question is why? That seems to be the most straightforward, "you get what you see" rendering. I know there's a good chance many parts will be clipped, but let me elaborate: I use capture one for my editing, and I made an export recipe with the paper's ICC profile selected as color space. If I enable soft proofing while that recipe is selected, I can edit the image while directly viewing a soft proof, with the rendering intent selected in the color settings.

    If I edit an image in Adobe RGB color space for instance, then enable soft proofing, often I feel it is easiest to modify the edit to get satisfactory results with the absolute colorimetric rendering intent selected. It usually involves lowering the white point so the highlights are not clipped, and very little adjustment apart from that. With relative colorimetric, highlights do not get clipped out of the box, but the whole image seems to get darker, and it is hard to adjust for that without reducing contrast, particularly in the highlights. Saturation rendering intent is just way over-saturated, so I wouldn't use that, but I feel that the mostly recommended perceptual rendering often looks out of whack as well, almost no color looks like it's supposed to, and highlight separation also suffers. So I fail to understand why absolute colorimetric rendering is never recommended for photography (obviously with the necessary adjustments), while perceptual seems to be the go-to selection, even though it changes hue, saturation and brightness values that we carefully set during the editing process.
     
  2. digitaldog

    digitaldog Andrew Rodney

    The absolute colorimetric rendering intent reproduces the exact color that existed in the source—absolutely. If the source was light color on the dingy yellow-white of newsprint, the resulting color on your brilliant coated ink jet paper will be dingy yellow. This intent is really designed for making one device simulate the appearance of another device for use in proofing (make an Epson simulate a press sheet as an example). You'd want that for proofing, then trim away anything that is paper white that wasn't printed on. Absolute and Relative Colorimetric are the same expect for this mapping of white. That is the only difference.
     
    Ed_Ingold likes this.
  3. digitaldog

    digitaldog Andrew Rodney

    There is absolutely no rules in how a profile making product conducts either the Saturation or more importantly, Perceptual Rendering. Just as there are no rules in how an E6 film manufacturer produces a rendering of their color transparencies. They build them hoping their customers prefer one rendering over another. They are not equal! One person may prefer Velvia, the other Agfachrome. The same is true with a perceptual rendering. You should soft proof with all available rendering intents EXCEPT Absolute Colorimetric unless you are proofing, then only use that. Profiles and rendering intents know nothing about images in context. They simply examine every individual pixel, one at a time to provide a conversion from Lab (they don't even 'know' the original color space, like Adobe RGB (1998), by the time an output color space enters the picture, they are handed Lab through the PCS). Pick the RI you visually prefer for your image based on the entire image shown to you.
     
    raczoliver likes this.
  4. Thanks for the reply, Andrew! It is becoming somewhat clearer through reading and printing lots of photographs. I bought some smaller size paper just for experimenting with different settings too. I must say the differences between different rendering intents are smaller after printing than they appear during soft proofing. Either way, I'm quite glad that I can get prints that are close enough in appearance to what I see on screen. Perhaps with more experience my requirements will get more stringent too.
     
    digitaldog likes this.
  5. digitaldog

    digitaldog Andrew Rodney

    There are many colors in a print output color space that fall out of display color gamut (wide gamut displays or otherwise). Meanwhile, no printer can print the entire sRGB color gamut. Soft proofing can't over come that. But it's a useful tool with limitations.
    The simulate paper white/ink black are important to use in full screen mode in some instances.
     
    raczoliver likes this.
  6. Nice read. As a non-professional, using my prints as decoration at home and in office I really only have me (and clients who sit in the office) to please. I most commonly use absolute colorimetric, although I flip thru the other options to see what looks best. "Looks best" is my criterion. Usually the print is close to the monitor given good color management. Sometimes the print looks off and I realize that the image on screen is off......
     
  7. digitaldog

    digitaldog Andrew Rodney

    Depending on source color space to destination color space white, there will be no difference in Abs vs Relcol conversions.
     

Share This Page

1111