Green fleshtone prints.

[W. Fisher]

Calibrated screen, printer, etc. done with i1 PhotoPro2 hardware and software on HP Advanced Glossy paper with Epson 3880. Even tried with Basiccolor software (dropRGB, Catch 5, etc.) but yet my prints still have a sickly green/cyan tint in the skin tone. Landscapes are fine, but skin color is a bit off to my liking with a pucky looking cyan green tint. Even same results with the ColorMunki Photo too.

What I do for a fix is toss a color Photo Filter in Photoshop CS6 onto the image with a RED setting of 5-10% and maybe a GREEN of 3% to help it out, but still don't know why the LCD screen and the printer do not match going through all the calibration steps. Density seems fine, but the damn cyan/green fleshtone is maddening!

Ideas on how to fix it since calibration isn't working? Close, but no cigar and there must be a way to refine it better in the printer's ICM somehow that I'm missing. Basiccolor sells some ICM tuning software called IMProve for $950, or MatchPatch for $400, that may tune the profiles better, but not crazy about buying it ($$$) or which is suited to the task.

I'm using the D50 too under a 5,000K Fiilex examine light. Nozzle checks are fine too prior to calibration and final printing.

Ideas?

W.F.

 

[Ink stained Fingers]

Which program are you using for your printouts ? What are the settings for color management ? Could it be that something with the program/driver settings is off in regard to color management - like double profiling ? If you are using Photoshop - how does the print look in proof mode - as well that greenish ? Do have a chance to use a profile by somebody else - a genuine Epson profile if you use their inks and paper ? Or a profile by the ink and/or the paper supplier for your ink/paper combination ? We should try to understand whether it's a profiling issue or something related to the system/software/driver/settings causing this effect

 

[W. Fisher]

ISF, I got this late last night and was thinking about it - and most of the night too!

The paper and printer profile was made with the Colormunki Photo or possibly the i1 PhotoPro 2 setup. I almost always use Qimage Ultimate for the print as I like its control and memory of settings.

That said, X-rite uses its own printer setup and output for its targets so if any software is later involved like PS, LR, Qimage, then who knows what that is compared to X-rite's printout? I know PS and PaintShop Pro X8 differ in outputs and PSP X8 even changes my screen color when it starts, but some say Corel never has gotten color management under control verses PS. So I stick with Qimage for now.

X-rite does have a secondary means of re-tuning the profile made with some other program like PS, etc where you drop an image made from PS, etc. into its software to re-analyze and hopefully improve the profile first made with it. Might need to do this re-analysis with each new X-rite profile made too with your editor/printer software preference?

Thinking back, BasICColor uses whatever printing software you use to print its test targets where X-rite does that within its own software. Maybe BasICColor has a better plan? Bad part is it is expensive, and the tuning/refining part is another more expensive program add-on. Their support isn't as great either (Like no response!) compared to X-rite.

I may try another profile with BasICColor dropRGB with the HP paper and see how that goes. If my X-rite software is leading to the greenish tint, maybe a new profile made with BasICColor dropRGB might help - maybe.

I can see where BasICColor Display 5 takes over on computer startup for the D50 profile as the LCD dims a bit for my dim workroom (80cd/m) and also turns a greenish/yellow for that Kelvin too vs. the D65. Same ugly green/yellow tint I see on the skin color in the prints too. Odd, but maybe that BasICColor Display 5 for the screen and the X-rite paper profile are in conflict? Also, BasICColor uses the X-rite i1 Display Pro head for the screen, and the i1 PhotoPro 2 head for the paper targets so maybe the two heads do not agree either?

I really hate color management! (


W.F.

 

[Ink stained Fingers]

I must admit I'm getting lost, there are quite some programs etc in the game, Colormunki, I1PhotoPro, Qimage, Photoshop, PaintShop Pro BasiCColor and what else ? I don't really know yet where to start in this environment. We somehow need to find out whether it's a program/settings/driver issue or a 'plain' fault of the profile before we consider 'tuning' the profile.
'X-rite uses its own printer setup and output for its targets', I don't relly believe that, can't you access the printer driver when the program starts the target printout ? Can't you save the target into a .tiff file instead ? If you can't access the driver settings at this time anymore the program most likely uses the default settings which you could change via Windows ? Printing the target via Qimage should give you easy access to the driver settings, I cannot imagine that x-Rite has a myriad of their owns printer setups for all printers around the world. I'm still using the previous version Profile Maker of X-Rite, I can save the target file to .tiff without problem, as well with Colorport which I use to read the color data with the I1 robot arm. And Colorport let me easily modify targets, adding and deleting patches, changing patch sizes, making them smaller for the robot scan. But I'm not familiar with the I1PhotoPro etc software by X-Rite.
OK., back to the starting point - testing the situation with a 3rd party profile made by some other company, e.g. PrecisionColors offers lots of profiles for the inks and papers in their offering. That 3rd party profile should eliminate any possible issues with your profile generation. And then printing the same picture with Qimage and the other programs to see whether the color cast is still there, or even differs by program.

 

[The Hat]

@W. Fisher, It also seems to me like you got to many cattle in the pen at once, try starting at the very beginning again and don’t change a thing.

Use one of the many test prints that are available free from say “Northlight” or any other test image your familiar with, and print that image in Qimage with no setting changes or profiling, just let the printer handle the colour settings on a glossy paper.

Then print the same image again, but this time set the paper and its corresponding profile up and allow Qimage to handle the colour setting, this will give you two good known quality prints to study because now you’ll have a basis to begin figuring out where your problems all began.

 

[W. Fisher]

You're lost? Where do you think I am? ;o)

Sort of odd that I also got a yellow/green skin image print back from a pro lab too. I know they use a CMYK printer for aluminum metal plate (dye-sub) prints so might be something there. Now I wonder if my notebook's screen calibration software or hardware is at fault at my end?

If my notebook's LCD screen calibration is at fault, it might explain some of it. But that is a result of the X-rite hardware too and maybe not the printer or the print profiles themselves? I don't know if the X-rite hardware (i.e. i1 PhotoPro 2, ColorMunki Photo, or i1 Display Pro) is at fault with regards to the screen calibration with the X-rite software. The BasICColor Display 5 D50 does look more green/yellow (The Windows 10 background blue window shifts to a warmer tint when the color management takes over for the LCD screen on booting up.) than the X-rite i1 Profiler software so something is askew there between the two companies.

For fun, I also put the Eizo CG monitor into emulation mode of the notebook using the X-rite hardware and it turned out looking just like my notebook screen; i.e. yellow/green skin on it too.

For what it's worth, I just switched papers to one that I reversed engineered by eye using the Eizo Color Navigator 6 software using the X-rite ColorChecker Passport. Once I printed the ColorChecker Passport via Qimage, I manually dialed in the Eizo Color Navigator 6 to match the print. I had to move both the red and green sliders way down in CN6 as they were much stronger on the Eizo screen than the print - almost a red and green neon!

Once done, I just printed from the Eizo screen via Qimage and the skin tones are now accurate - for that paper. No more green tint skin so Yay! I don't know what is going on, but maybe the notebook's LCD has some gamut issue that X-rite calibrators, or their software, are not addressing quite right. It is sort of a pain to go through the reverse engineering in CN 6 to pull the green and red sliders down in a ColorChecker Passport test print for anything I have already profiled for previously - but if that is what I have to do then I guess that's the plan.

Puzzling, and maddening both!

W.F.

 

[Ink stained Fingers]

- Don't add more complexity to the case ..... notebook displays - many of them are just not usable for decent color critical work - even calibrated as long as the viewing angle changes the colors on the screen. You wouldn't necessarily need an expensive Eizo etc, but there are some screen types - I think it's the IPS panels - with the colors less angle dependant than with other types. It's pretty critical , you move around in front of the disyplay, and you look to the sides of a screen under a different angle than looking to the center. Do you know somebody who could print your test images on his printer - for comparison with your setup. Or the other way around - you create a profile for somebody else - his computer and printer, ink and paper, and he prints your test image or his images with your profile on his equipment. This could show whether the green cast is showing up as well - or not, or only in your images.... Some of the debugging needs to be done by exclusion - it's not this or that.