Problem: Copying Artwork & Matching Colours

[Mark D Segal]

Reviewing the discussion to date on this thread, and based on gamut testing I've done in the context of my various printer and paper reviews for this website, I am landing on a high probability that Tim Lookingbill nailed it correctly in his post #37. I do know that between an Epson 4900 (very wide gamut capability) and an Epson P800 or Canon iPF Pro-1000 (narrower range of gamut capabilities compared with the 4900), there is a reduction of gamut potential and the colours that most bore the brunt of that were in a narrow band of bright green to yellow-ish. This seems pretty close to some variants of jade, and therefore support's Tim's observation.

So, the path to a solution that occurs to me is to try to eek-out the widest possible gamut rendition you can from that iPF8400, which of course raises questions about the capability of the iPF8400, the gamut rendering potential of the paper and the "reach" of the profile. So I had a look at all this for you. Because Canon's website is a train-wreck for finding such things, I went to Ilford's instead, where I easily downloaded their profiles for Gold Fibre Gloss (a very wide gamut paper) for both the Epson 4900 and Canon iPF 8400. Then I compared these profiles in ColorThink Pro, and here is what I found: for the 4900, gamut volume is very close to 978K. For the iPF8400, gamut volume is a tad over 880K - hence considerably less gamut on the IPF8400 for the same paper. Then the question is, where does the gamut take a hit on the iPF8400. For that, a three dimensional diagram generated in ColorThink Pro helps a lot. Please see the illustration, where the multi-colour is the Epson 4900 and the solid red is the Canon iPF8400. You will see that in the full range where that jade hue could fall, it may just be out of gamut for that printer/paper combination.

I have also determined from my prior research that making custom profiles - with high quality profiling equipment - can often expand gamut relative to OEM profiles. So perhaps your next step would be to either make or order a high quality profile for the widest gamut paper you can use and see whether it eeks out enough to reproduce the jade more correctly. Wide gamut papers would include such as Ilford Gold Fibre Silk, Ilford Gold Fibre Gloss, Hahn FA Baryta Satin, Canon Pro Luster, Epson Legacy Baryta, Canson or Epson Legacy Platine, Canson Baryta Photographique, and there are others. The custom profiling should be done with better than a Colormunki. If you don't have an i1Pro and i1Profiler, I would recommend using a good custom service that does. If you do not have ColorThink Pro for analyzing the profiles you are now using or the one you eventually order, you can send them to me and I shall read them for you. I do suspect we are closing-in on a gamut/profiling issue that just happens to hit that part of the spectrum a bit hard. The only other thing I can think of is that the spectral properties of the paint are such that conventional ICC profiling targets are not well-adapted to the subject matter at hand. I have also seen this issue some years ago. But I suggest first seeing what is the best that can be produced from that printer and a really good profile on paper that has very wide gamut rendering potential.

[Doug Gray]

The only other thing I can think of is that the spectral properties of the paint are such that conventional ICC profiling targets are not well-adapted to the subject matter at hand. I have also seen this issue some years ago. But I suggest first seeing what is the best that can be produced from that printer and a really good profile on paper that has very wide gamut rendering potential.

Good observation and a definite possibility. Some pigments are spectrally not very smooth in their transitions. This can interact with spectrally spiky illuminants tor produce significant metameric error. I encountered this on products ostensibly designed with spectrally smooth pigments. GMB had a color hue test they sold with carefully pigmented tiles in adjacent hues which you would line up and score based on the number of out-of-sequence tiles. I could normally arrange them perfectly under a North facing window but had a couple out of sequence when using a high CRI (93)  CFL color booth. Out of curiosity I recorded the spectrum of the illuminant and mismatched tiles. Sure enough there was a slight, narrow, bump up in the reflectance spectrum where a major illuminant spectral spike presented.

Artists, unlike GMB's tile specifiers, aren't looking at the spectral smoothness of their pigments and there could be some sort of amplification of color shift due to the relative position of peaks both in the pigments and illuminant.

Might be worth an investment in Solux 5000K lamps or higher CRI studio lamps. Flash strobes are also very good.

On another point, since the OP has a colormunki I would recommend creating several patches in Photoshop around the colors that are not reproducing correctly. Print them using Absolute Colorimetric. Then measure the Lab values of those patches with Argyll's free software. If there's a gamut issue or if supposedly in gamut colors are being printed with a large hue shift then it can be quantified and any new profile, customer, manufacturer, or otherwise, can be easily compared for accuracy in those colors by doing the same.

[xpatUSA]

Have you printed a crop of the jade color in the OP's posted jpeg of the source file on your printer using similar hot press type paper/profile, or just printing with "Printer Manages Color"? This is a tagged image that should communicate to the printer what ink amount combination gets dropped on the paper to reproduce that jade color. The numbers don't matter of course.

Sorry, I don't print. My post was offered for it's worth, not as a rebuttal of anything said.

I only made that point to indicate that particular hue of jade may or may not be right on the edge of the printer's color gamut capabilities to where there may now be a very narrow space on the target to get it or not get it on paper. Can't recall in this thread if someone mapped that jade to see if it fits within that particular printer/paper combo reproduction capability.

Can't do that kind of mapping; could open each image in ColorThink but don't have the profiles compare, sorry.

[Tim Lookingbill]

Something basic, rather than esoteric is wrong.  Those colors might be (probably are) out of gamut, but bringing them back into gamut would be much, much closer to the original compared with what your results are.   in this case, those colors in your two images are not off a little and are unacceptable.....they are a million miles apart.   
Brad

You got me to think about that and upon further examination of the OP's original shot of the print, not only is the jade way off but all the rest of the colors as well which looks kind of familiar. So familiar that with some rummaging around in my printer profiles in Photoshop I found ONLY ONE (Epson Proofing Paper) that mimics those colors exactly (or close enough) by assigning it to the OP's original sRGB image after first converting it to AdobeRGB. See screengrab below.

Decided to print a crop of the OP's original sRGB image on my Epson NX330 "All In One" using "Printer Manages Color" and at first try got a white balance mismatch as I've always gotten and to correct reduce the blue channel middle slider in Photoshop Levels. Pretty good match.

See Raw shot of the print lit under daylight fluorescent T8 tubes taken with my Pentax K100D and processed to match in ACR 6.7.

It's not hairsplitting exact but at least it's not a million miles off compared to the OP's original shot of the print. The jade is within gamut of an inkjet.

And just FYI the Epson NX330 print of this jade lit under my daylight flotubes was difficult to get the same intensity viewed on my sRGB display. Sliding the Aqua Hue/Sat in ACR barely made a dent in changing the jade. It just got lighter so I just made look as close to the Epson print as possible.

[Mark D Segal]

You got me to think about that and upon further examination of the OP's original shot of the print, not only is the jade way off but all the rest of the colors as well which looks kind of familiar. So familiar that with some rummaging around in my printer profiles in Photoshop I found ONLY ONE (Epson Proofing Paper) that mimics those colors exactly (or close enough) by assigning it to the OP's original sRGB image after first converting it to AdobeRGB. See screengrab below.

Decided to print a crop of the OP's original sRGB image on my Epson NX330 "All In One" using "Printer Manages Color" and at first try got a white balance mismatch as I've always gotten and to correct reduce the blue channel middle slider in Photoshop Levels. Pretty good match.

See Raw shot of the print lit under daylight fluorescent T8 tubes taken with my Pentax K100D and processed to match in ACR 6.7.

It's not hairsplitting exact but at least it's not a million miles off compared to the OP's original shot of the print. The jade is within gamut of an inkjet.

And just FYI the Epson NX330 print of this jade lit under my daylight flotubes was difficult to get the same intensity viewed on my sRGB display. Sliding the Aqua Hue/Sat in ACR barely made a dent in changing the jade. It just got lighter so I just made look as close to the Epson print as possible.

You managed in quite a methodical way to get some convergence, but it's a fair bit of work-about which perhaps shouldn't be necessary in a normal workflow producing acceptable output, so reverting to Square-1: His original photo is a raw file in ProPhoto space, so it is unlikely that he clipped the jade at the capture stage. He used Museum Etching paper, which is a relatively lower gamut paper (being matte). He also says this colour is the only one he sees as "off". So I am still thinking he has an OOG problem with that particular colour range, which may be resolved using a luster paper and a good custom profile. .

[Tim Lookingbill]

You managed in quite a methodical way to get some convergence, but it's a fair bit of work-about which perhaps shouldn't be necessary in a normal workflow producing acceptable output, so reverting to Square-1: His original photo is a raw file in ProPhoto space, so it is unlikely that he clipped the jade at the capture stage. He used Museum Etching paper, which is a relatively lower gamut paper (being matte). He also says this colour is the only one he sees as "off". So I am still thinking he has an OOG problem with that particular colour range, which may be resolved using a luster paper and a good custom profile. .

I didn't say the OP clipped the jade at the capture stage.

He also says this colour is the only one he sees as "off".

He also said the original ProPhotoRGB image which he posted in sRGB in his OP looks very close to the painting. That sRGB image shows skin tone a somewhat saturated pinkish orange while the print shows it as dull yellow.

How can only the jade be off.

But even more puzzling is how can you explain the close match of the assigned Epson printer profile to the AdobeRGB. In color management either it's WAY off or it's slightly off. But to have some random box of colors that represents the print shot with a digital camera match up by just assigning ONE out of a long list of printer profiles is more than a coincidence wouldn't you say?

[Mark D Segal]

Correct - it's not you, it's me saying he didn't clip at the capture stage - I'm just trying to rehash from "SQ-1" where we have headed from....to!

I'm taking his word about the other colours because he sees what's actually on paper relative to what is on his display, but we don't. Who knows what happened going from ProPhoto raw to sRGB JPEGs posted here. It can get very confusing. I'm not sure what to make of one profile amongst many working reasonably well - could be a fluke, especially as it's a profile that isn't related to the paper he is using. I don't have a view one way or another on that happy coincidence.

So knitting together eveything he told us and assuming he has good visual perception, as seems reasonable based on what he reports about what he is doing, I would prefer to take his word that he is having one CM issue with the reproduction of the "jade range" and if that's the case, I think the cause is one of four things: (1) the paper gamut is inadequate (quite likely); (2) the printer gamut is inadequate (less likely with wide gamut potential paper); (3) the profile is inadequate (more likely), (4) the spectral response of that colour range is causing an issue (quite possible). To deal with items (1), (2) and (3) he should get a high quality custom profile made by a reputable provider and test it. If that solves the problem, he's home free and has his colour-managed work flow back in order. If it doesn't solve the problem, the issue is then likely item (4) in which case he probably needs a profile based on a target that can handle those paint hues.

[Mark D Segal]

Continuing further from Reply 46, I went to the Hahnemuhle site and downloaded/analyzed their Museum Etching profile for the iPF 8400. The gamut volume is a slender 479K, while in that printer for IGFG it is 880K. When you compare it with the gamut volume of IGFG in the same printer, the result is what you see in the attached illustration, where IGFG is multi-colour and Hahn Museum Etching is red. There is no red to be seen in the 3D anywhere near where the jade range would sit. I think William you should try printing this photo with IGFG or a paper similar to IGFG using the manufacturer's profile and see what happens to the rendition of jade. Perhaps do this before going to the trouble of getting a custom profile. The whole problem may be gamut constraint in that particular colour range for that paper.

[William Walker]

As I said in my post yesterday, the last ditch print I did using a 180gsm "Bond Paper" with Canon's Bond Paper Canned Profile produced a colour "close enough". I mean, really close.

Here is a picture taken with my iPhone showing the difference, which is: the larger print on the left would be the colour blue that appeared on just about everything I tried (from matt to glossy) - including canned profiles.

The smaller print on the right is the Bond paper print (no soft-proofing) - the difference you see there is what the problem has been from the start. The small print is fine as far as I am concerned - which means the iFP 8400 can print that particular colour.

Does this not narrow this down to the profile then?

William

PS. Mark, I am blown away by the trouble that you have gone to! Thank you.

[Tim Lookingbill]

As I said in my post yesterday, the last ditch print I did using a 180gsm "Bond Paper" with Canon's Bond Paper Canned Profile produced a colour "close enough". I mean, really close.

Here is a picture taken with my iPhone showing the difference, which is: the larger print on the left would be the colour blue that appeared on just about everything I tried (from matt to glossy) - including canned profiles.

The smaller print on the right is the Bond paper print (no soft-proofing) - the difference you see there is what the problem has been from the start. The small print is fine as far as I am concerned - which means the iFP 8400 can print that particular colour.

Does this not narrow this down to the profile then?

William

PS. Mark, I am blown away by the trouble that you have gone to! Thank you.

So the one on the left is Hahn Museum Etching Paper? This is what you indicated in your OP. Or is that Bond Paper too?

The Bond Paper version on the right is REALLY close and that iPhone is pretty accurate with color rendering which is surprising. Don't know if you had to edit it.

[William Walker]

So the one on the left is Hahn Museum Etching Paper? This is what you indicated in your OP. Or is that Bond Paper too?

The Bond Paper version on the right is REALLY close and that iPhone is pretty accurate with color rendering which is surprising. Don't know if you had to edit it.

No Tim, that particular print - on the left - was the Canon Glossy Photo paper with their canned profile. However, it is representative/typical of all the prints I made regardless of paper, texture or profile.

Thanks to you too for your interest!

[Czornyj]

William - the whole discussion is totally over engeneered, while there's some basic issue in your workflow/application settings.

Printed on matte fine art paper with Fine Art Heavyweight Photo stock Canon profile, with iPF8300:

[Czornyj]

The whole problem may be gamut constraint in that particular colour range for that paper.

Nonsense. This phtalo green doesn't get even close to the gamut of iPF8300 on matte FA media:

[Mark D Segal]

No, it's not *necessarily* nonsense because I've seen this kind of thing before. Starting from his original file in ProPhoto space, it depends on the specific Lab recipe of the jade range relative to the specific gamut size and more important perhaps gamut shape of the profile(s) that seem(s) to be causing the problem. You claim there is a *basic issue* in William's workflow and application settings. I think William has provided enough information putting together the content of several posts above to indicate that he knows his way around colour management and printer settings; not to say mistakes can't still happen, but for avoidance of doubt William, perhaps you can provide screen grabs of all the colour management and printer settings you used that gave you the wrong outcome, so we can see whether Marcin's suggestion turns out to be correct.

I wanted to add, William, as Tim has found a profile that he thinks seems to work for your image, it may be an idea to use that same profile on the Museum Etching paper in your iPF 8400 (I know, a mismatch of profile and paper) to see whether *by any chance* it contributes to a solution.

[Czornyj]

William provided enough information to conclude there's something terribly wrong, so that the colours on his prints are completely distorted.

[Mark D Segal]

William provided enough information to conclude there's something terribly wrong, so that the colours on his prints are completely distorted.

As I said, not impossible, but precisely WHAT? That's why I suggested we see all his settings and that he try several experiments, rather than be contented with general statements to the effect that "something is wrong" or there is a "basic issue". William needs a solution and as you no doubt know, the right solution depends on a clear and specific understanding of the problem's cause. Some of us are trying to be helpful by pursuing a number of approaches that may lead to an operationally significant diagnosis.

[Czornyj]

As I said, not impossible, but precisely WHAT?

It looks exactly like an sRGB image printed with no color conversion (like a target):

[Mark D Segal]

Well, this isn't my understanding of what he is doing.

William, at the risk of getting tedious, grateful if you could gather in one place the details and screen grabs of ALL the capture, colour management and print settings you used that generated the original problem you observed and let's see if we can detect more or different than what has been said so far.

[BobShaw]

No Bob, what I meant to say was that my monitor is properly calibrated, the tweaks I made were in Lightroom, so that the the colours on the (properly calibrated) display matched the colours of the painting.
The print is where I am battling to match the colours on both the display and the painting itself.
Keep in mind that it is only with this particular image that I am having a problem. My "normal' day-to-day printing of my landscape photography and for other people for who I print are fine.
I digitised 22 pieces of artwork for another artist and had a similar problem with 2 of those, the rest were fine. The problem seems to be in the green/blue areas. It just seems crazy that my prints normally get pretty close to what I see on the display, but not in this instance...

OK, so how did you "tweak" these colours to be the same? The only way is to measure them, not by eye. The values displayed in the application will always be different to the values on the monitor seen using Digital Color Meter.
If the colours that you are photographing are out of gamut of the monitor then it will never look the same. Ignore the display.
Ideally you shoot a ColorCheckR or similar to adjust the colours recorded. Phocus does this almost automatically for Hasselblad.

[Tim Lookingbill]

Well, this isn't my understanding of what he is doing.

William, at the risk of getting tedious, grateful if you could gather in one place the details and screen grabs of ALL the capture, colour management and print settings you used that generated the original problem you observed and let's see if we can detect more or different than what has been said so far.

Mark, William made it clear this odd color mismatch marcin points out looks like a "color target" rendering which also resembles my Epson Proof Paper profile assign to AdobeRGB data experiment with this quote...

No Tim, that particular print - on the left - was the Canon Glossy Photo paper with their canned profile. However, it is representative/typical of all the prints I made regardless of paper, texture or profile.

So this "color target" look as marcin described is even happening with canned profiles, not just the custom profile. So it appears there could be some transform getting turned on/off within the printer driver/profile hand off that might be caused by the way those particular profiles are written, maybe some line of code or tag causes the driver to flip a setting off/on that needs to be thoroughly and carefully checked out.

If my $50 Epson NX330 can print a reasonable facsimile of his sRGB original using "Printer Manages Color" then William shouldn't be getting such poor results using canned and custom profiles.

And the one on the right in William's iPhone shot of the two prints was also done with a canned profile on glossy Bond Paper and it is a very close match.