Yves,
Thank you for your reply. I’d be interested in taking a look at those papers
you say have a larger volume than sRGB. You say that “no paper can faithfully
reproduce the gamut of sRGB”, that’s the sad truth. It has to do with the
nature of the colorants found in current inkjet inks, maybe in a few years,
things will improve? But I suppose the real killer is the substrate itself,
compared with the “Luminance range” of a monitor.
Keep in mind, that “most imagery” falls quite short of “filling the entire sRGB
gamut”. If my work involves a lot of flowers and other saturated objects, I
agree, even the best “paper gamut” falls short of being to transport those
colors. If my work involves a lot of panting, same idea. If my work involves a
lot of portraiture then sRGB offers all the required gamut, and some.
A digital camera may be considered as “having a gamut of its own” but you and I
and everybody else know that this is “misnomer”, a camera does not have a gamut
per se, it is the scene that makes the gamut of colors being captured. Which
makes profiling digital cameras difficult in an absolute sense. Please correct
me if I am wrong.
I’d be curious to statistically evaluate the range of imagery you process in
your work? How is it distributed? I agree with you – but I have yet to realize
this, in practice – the idea of “image-based” gamut mapping is a marked
improvement over standard ICC profile conversion practice.
Subjectivity is very important, your taste matters a lot 😊
I remember experimenting with sRGB_v4_ICC_preference a while ago. The idea was
“sound” but I can’t remember what the results were? I’ll have to revisit it
soon…
/ Roger
From: argyllcms-bounce@xxxxxxxxxxxxx <argyllcms-bounce@xxxxxxxxxxxxx> On Behalf
Of Yves Gauvreau
Sent: December 7, 2019 8:01 AM
To: argyllcms@xxxxxxxxxxxxx
Subject: [argyllcms] Re: Image dependent gamut mapping using device links
Roger,
I try to use as large gamut papers as possible and even if their numerical
volume is larger sometimes then the sRGB gamut, I haven't found yet a gamut for
a paper whose boolean intersection with the gamut of sRGB that is not smaller
then sRGB. That means for the moment no paper can reproduce faithfully all the
color of sRGB. We can also say that many papers can "print" colors outside of
the sRGB gamut and a few can do the same for Adobe RGB.
We all know I'm sure that cameras have a gamut that is much larger then even
Adobe RGB. If we trow them out the door from the start by using an input
profile like sRGB it is certain they'll never end up on the prints. I using the
"Image dependent gamut mapping using device links" scenario more and more, I've
even made a batch file to simplify my life. All this so as to print as much of
the captured colors as possible and now that I use it, I find using the
"Luminance matched Appearance" give me a softproof image that I like better
looking then using the paper profile as usual. The difference is small but
still I like it better, that's subjective I know.
My question is could we incorporate the same idea in the paper profile itself?
I understand that if I want a truly image dependent mapping, I should stay with
the device link method but I wonder if something like is done with the
sRGB_v4_ICC_preference.icc, just the gamut part of course would improve print
just enough to warrant the effort?
Just thinking
Yves
On 12/6/2019 3:52 PM, graxx@xxxxxxxxxxxx <mailto:graxx@xxxxxxxxxxxx> wrote:
Yves,
Congratulate on your first paper profile.
I was comparing the full gamut of my monitor (NEC PA271W) to the gamut of my
“paper” profile, yesterday, and there is no way that my Epson printer (SC5000
on glossy) is able to render the saturation extent of my monitor – physically
impossible. The yellows, yes, but forget the saturated blues and greens. It’s
an exercise in compromise.
/ Roger
From: argyllcms-bounce@xxxxxxxxxxxxx <mailto:argyllcms-bounce@xxxxxxxxxxxxx>
<mailto:argyllcms-bounce@xxxxxxxxxxxxx> <argyllcms-bounce@xxxxxxxxxxxxx> On
Behalf Of Yves Gauvreau
Sent: December 6, 2019 3:48 PM
To: argyllcms@xxxxxxxxxxxxx <mailto:argyllcms@xxxxxxxxxxxxx>
Subject: [argyllcms] Image dependent gamut mapping using device links
I did my first paper (ink, printer) profile and it is suggested to use an input
profile like sRGB for the gamut mapping and I wonder if for those who have a
wide-gamut display, almost Adobe RGB in my case, wouldn't it be better to use
Adobe RGB, something else?
I see also that colprof seems to have the same abilities as devlink as is use
in the image dependent gamut mapping scenario. Does this mean we could
incorporate into a regular profile a gamut representative of a few images or
even just one image. Would that be useful or just a waste of time?
I'm curious about the "la - Luminance matched Appearance" since we could only
apply this using Argyl tools. Maybe we can trick Photoshop into using this
rendering intent, possible?
Yves
