This is really a continuation of my previous thread on it, which has drifted a bit from glop/tissue issues to pigment set selection: https://www.photrio.com/forum/forums/alternative-processes.88/
However, I think the pigment selection issue is more than complex and important enough to have its own thread, hence the new one. The basic premise of this thread is that when you undertake color printing with a pigment process, you have to select your pigments. It's something I'm currently involved in and it's sort of fun, but also a bit of a (very) deep hole. I wrote a blog post about it that outlines my rather rudimentary approach to selecting a pigment set, and translates that into an actual shopping list: https://tinker.koraks.nl/photography/couleur-locale-the-search-for-color-in-carbon/
In the post above, I take my first ever color carbon print, point out the most glaring problems and then use that as a starting point to embark upon a pigment selection process, involving a very quick & dirty look at a color wheel and some (dangerously shaky) conclusions concerning the resulting gamut of a pigment selection.
There are many loose ends however, and I'm sure some of you have dealt with these. For instance:
Estimating the gamut?
If you select e.g. 3 primaries for a pigment set, you can make a gamut by connecting them with straight lines. But the edges of a gamut are never straight lines. To what extent can you determine the resulting gamut of a 3-pigment selection on a theoretical basis? Or is the only feasible route an empirical one, i.e. print the gamut with the actual pigments and measure the outcome?
Is cyan really cyan, or is it blue?
If you look at existing pigment sets, something odd occurs to me. The de-facto subtractive standard is a CMY(K) set, suggesting we work with cyan, magenta and yellow. And yet, if you look at for instance inkjet inks, they generally use something like PB15:3 for the cyan pigment and PR122 or so for the magenta pigment. Now, correct me if I'm wrong, but PB15:3 is pretty blue and not all that cyan, and PR122 is rather red and not all that magenta. So in reality, a set of PB15:3, PR122 and some yellow pigment (these are grouped pretty close together in the cadmium yellow section of the color space and also all pretty perfectly yellow) will in my eyes be more like a blue, red and yellow set than a cyan, magenta and yellow set. Mind you, the inks in my inkjet printer look pretty CMY to my eyes, although I know that the C is nearly certainly a PB15:3 pigment, which really is rather blue (look at Calvin Grier's cyan color paste, which is decidedly blue as well). What's up with that?
How much is enough?
Also, how do you determine pigment load in a glop (for carbon) or a gum mix (for gum bichromate)? Surely, you want to use enough of it to be able to print a pure primary hue. But how much is enough? Especially for carbon tissue, you don't want to overdo it, as excess pigment is a waste, limits relief and more importantly creates problems with delicate highlights as the contrast goes up too far. Is 'enough' the point where adding more pigment does not make a visible difference anymore? Or is that already too far?
How do we know what we'll print?
This one is perhaps more for the hybrid forum, but the issue of soft-proofing is probably inescapable. Given the nature of the pigments used, you can't just assume that if you decompose an sRGB color image into CMYK components, you'll end up with a reasonably looking color print. A translation to the destination color space must be performed, and in my case, that color space would be the gamut of the carbon color printing process and ink set I use. How do you manage that - doing it well probably involves a color profiling tool of some sort (ColorMunki etc. ) - is there a haphazard, seat-of-the-pants way as well that gets you in the ballpark?
So, many more questions still...
Btw, this print started it all, it's my first color carbon photograph, and evidently it needs some work (right half is the digital original, left half is the actual print. Yeah it's the same image!):
However, I think the pigment selection issue is more than complex and important enough to have its own thread, hence the new one. The basic premise of this thread is that when you undertake color printing with a pigment process, you have to select your pigments. It's something I'm currently involved in and it's sort of fun, but also a bit of a (very) deep hole. I wrote a blog post about it that outlines my rather rudimentary approach to selecting a pigment set, and translates that into an actual shopping list: https://tinker.koraks.nl/photography/couleur-locale-the-search-for-color-in-carbon/
In the post above, I take my first ever color carbon print, point out the most glaring problems and then use that as a starting point to embark upon a pigment selection process, involving a very quick & dirty look at a color wheel and some (dangerously shaky) conclusions concerning the resulting gamut of a pigment selection.
There are many loose ends however, and I'm sure some of you have dealt with these. For instance:
Estimating the gamut?
If you select e.g. 3 primaries for a pigment set, you can make a gamut by connecting them with straight lines. But the edges of a gamut are never straight lines. To what extent can you determine the resulting gamut of a 3-pigment selection on a theoretical basis? Or is the only feasible route an empirical one, i.e. print the gamut with the actual pigments and measure the outcome?
Is cyan really cyan, or is it blue?
If you look at existing pigment sets, something odd occurs to me. The de-facto subtractive standard is a CMY(K) set, suggesting we work with cyan, magenta and yellow. And yet, if you look at for instance inkjet inks, they generally use something like PB15:3 for the cyan pigment and PR122 or so for the magenta pigment. Now, correct me if I'm wrong, but PB15:3 is pretty blue and not all that cyan, and PR122 is rather red and not all that magenta. So in reality, a set of PB15:3, PR122 and some yellow pigment (these are grouped pretty close together in the cadmium yellow section of the color space and also all pretty perfectly yellow) will in my eyes be more like a blue, red and yellow set than a cyan, magenta and yellow set. Mind you, the inks in my inkjet printer look pretty CMY to my eyes, although I know that the C is nearly certainly a PB15:3 pigment, which really is rather blue (look at Calvin Grier's cyan color paste, which is decidedly blue as well). What's up with that?
How much is enough?
Also, how do you determine pigment load in a glop (for carbon) or a gum mix (for gum bichromate)? Surely, you want to use enough of it to be able to print a pure primary hue. But how much is enough? Especially for carbon tissue, you don't want to overdo it, as excess pigment is a waste, limits relief and more importantly creates problems with delicate highlights as the contrast goes up too far. Is 'enough' the point where adding more pigment does not make a visible difference anymore? Or is that already too far?
How do we know what we'll print?
This one is perhaps more for the hybrid forum, but the issue of soft-proofing is probably inescapable. Given the nature of the pigments used, you can't just assume that if you decompose an sRGB color image into CMYK components, you'll end up with a reasonably looking color print. A translation to the destination color space must be performed, and in my case, that color space would be the gamut of the carbon color printing process and ink set I use. How do you manage that - doing it well probably involves a color profiling tool of some sort (ColorMunki etc. ) - is there a haphazard, seat-of-the-pants way as well that gets you in the ballpark?
So, many more questions still...
Btw, this print started it all, it's my first color carbon photograph, and evidently it needs some work (right half is the digital original, left half is the actual print. Yeah it's the same image!):
