Color gamut of an Epson 3880

[Mark D Segal]

I generally roll my eyes when I come across yet another technical article about how to identify and then automatically tweak "key memory colors" like sky, grass, skin etc. in images to make the result "more pleasing". But then this sort of tweaking has gone on forever - none of the photographic processes reproduced color faithfully, they all tweaked it (using chemical "algorithms") to enhance saturation, contrast etc. etc. so that people liked what they saw, and make it better corresponded to what they remembered. Much of the digital workflow has emulated all that.

In the final analysis photographers, unless they are practicing forensic, scientific or commercial photography where "colour accuracy" may matter, generally want their photographs to suit artistic purposes for which faithful colour reproduction may be either a rather low priority or even irrelevant. Processes aren't ends in their own right - they are a means to an end and the ends will differ depending.

[Lundberg02]

I believe all color space conversions are made in RC....no matter what you select.

Yes for the monitor, but you can soft proof or do what I do, let the Epson give you perceptual and make a few prints til it's right.

[hjulenissen]

Probably not - because you haven't mentioned creating a device link or ICC profile with a gamut mapping tailored specifically to your source space or images. I'd guess you are using the generic gamut mapping created by whatever software created the ICC profiles you are using.

*I have profiled my camera using a colorchecker Passport/xRite software and a couple of different illuminations.
*I semi-periodically profile/calibrate my displays using the xRite i1 display pro and i1 profiler software
*I download printer profiles for my 3880 from my paper suppliers.
*I print from Adobe Lightroom after using softproofing.

All of these assume that the color characteristics of my devices are of a somewhat static nature (measure in one or a few cases, reuse afterwards).

It's a requirement from the definition of (real) gamut mapping - mapping one gamut into another. That implies two gamuts, a source and a destination, if the mapping is going to be accurate in this task. As I was alluding to, while you can get away with assuming your images completely fill a smaller gamut space like sRGB or AdobeRGB, assuming this with a large gamut space (like ProPhoto, scRGB, L*a*b* etc.) will give extreme compression, and it probably won't look very good if mapped to your printer space.

So if what I am doing (which seems to be a pretty common normal way of doing things) is not (real) gamut mapping, then what am I doing, and what am I missing? If a patch of an image is numerically represented as rgb [12,42,255] and the camera profile describe how the camera maps (perceptual-models of) color to 3-channel readings and the display/printer profile describes how the (post calibration) display/printer maps 3-channel input to (perceptual-models-of) color, then what is left?

Makes sense to me. But that's not how things have evolved. Under the pressure of people demanding that "it just work", everyone in the chain has tried to make things look "really nice" without needing to judge or adjust or choose anything.

I always thought that sRGB was the "always just works" method, and that color management was an attempt to make things "right" at the expense of endless hair-pulling.

I generally roll my eyes when I come across yet another technical article about how to identify and then automatically tweak "key memory colors" like sky, grass, skin etc. in images to make the result "more pleasing". But then this sort of tweaking has gone on forever - none of the photographic processes reproduced color faithfully, they all tweaked it (using chemical "algorithms") to enhance saturation, contrast etc. etc. so that people liked what they saw, and make it better corresponded to what they remembered. Much of the digital workflow has emulated all that.

I want my pipeline to be able to do perceptually transparent reproductions of reality (or as much so as 2-d static technology and my time/economy allows). Then I take the liberty to bluntly ignore the "scientifically correct" answer whenever I feel like it based on my conscious choice, not some faceless product developer.

-h

[Mark D Segal]

I always thought that sRGB was the "always just works" method, .................

-h

Actually, if we are talking about today's professional inkjet printers you can be sure that sRGB will be the "always DOES NOT work method" because it's colour space is far too narrow relative to the colour reproduction capabilities of these printers. It is safest to use ProPhoto as your colour working space and softproof to your printer profile. I expect you know this, so I'm wondering what you meant by that comment.

[GWGill]

Well call it a calculation/function/mapping/transformation,  ..

Conflating related but different terms together, doesn't make for clear communications.

[GWGill]

In the final analysis photographers, unless they are practicing forensic, scientific or commercial photography where "colour accuracy" may matter, generally want their photographs to suit artistic purposes for which faithful colour reproduction may be either a rather low priority or even irrelevant. Processes aren't ends in their own right - they are a means to an end and the ends will differ depending.

Sure, and this works well if it's a closed loop processes, where one party (i.e. Kodak, Fuji) has made it all work.

But in the modern world where photography has many players in the chain, it doesn't work so well when each player adds their own "special subjective sauce" to the mix. To untangle it all requires a clear understanding of each step in the chain, preferably with just a single step where the subjective/artistic manipulations take place, either automatically or under control of the user, with all the other steps striving for faithful color reproduction.

[GWGill]

So if what I am doing (which seems to be a pretty common normal way of doing things) is not (real) gamut mapping, then what am I doing, and what am I missing?

You are relying on the generic (i.e. not specific to the gamut you want to reproduce) gamut mapping/clipping in your printer profiles. Your paper suppliers certainly don't know what your image colorspaces are, nor what gamut your images occupy out of that space.

If a patch of an image is numerically represented as rgb [12,42,255] and the camera profile describe how the camera maps (perceptual-models of) color to 3-channel readings and the display/printer profile describes how the (post calibration) display/printer maps 3-channel input to (perceptual-models-of) color, then what is left?

Making one fit into the other (i.e. gamut mapping).

[Mark D Segal]

Sure, and this works well if it's a closed loop processes, where one party (i.e. Kodak, Fuji) has made it all work.

But in the modern world where photography has many players in the chain, it doesn't work so well when each player adds their own "special subjective sauce" to the mix. To untangle it all requires a clear understanding of each step in the chain, preferably with just a single step where the subjective/artistic manipulations take place, either automatically or under control of the user, with all the other steps striving for faithful color reproduction.

Yes, put in that way it makes sense.

[elolaugesen]

Hi.   First,  Thank you to all the contributors to this post.  It was very educational and I found out why I was having problems with some colors when printing on my 3880.  I went back to the original Raw files set Camera Raw to Prophoto rgb    then printed our a sample print on art paper.  Also printed a sample of the same original raw file but using Adobe RGB.  What a difference.  My other half who is a working artist and uses very vibrant colors asked why I had not done this before.  Her prints just popped out. and the details could now be seen etc....  I did not make any other changes to the settings in Camera Raw....

Background:   I am a one trick pony and only work with original art/paintings etc.  No photographs of lakes, mountains etc etc.  Only Original Art.

I then ran a test with some of the standard traditional paintings from the lake district in England.  Paintings of scenes in Ontario, Terra Cotta up to Georgian Bay.  There was no obvious difference in the results when comparing Prophoto to AdobeRGB.  (all using the original Raw data).  So it seems that only where you have very vibrant scenes may Prophoto rgb be of visible benefit.

Therefore while I will now use Prophoto all the time,  I think I know when it will enhance the results.

One Question --   Camera Calibration?   when shooting a sample of my Color Checker Passport, I check it out in Camera Raw to see if light is ok etc...
Then I save the image in DNG and go in to create my camera Profile.  
Will the use of the Prophoto have any effect on this process.  That is will all the patches be within the adobe rgb gamut only??

Thank you   Elo

[BobShaw]

So here's how you can test this and see the results, ink on paper:
High Resolution Video: http://digitaldog.net/files/WideGamutPrintVideo.mov
Low Resolution (YouTube): https://www.youtube.com/watch?v=vLlr7wpAZKs&feature=youtu.be[/i]

Thanks. That was a very effective demonstration.