General Question on Black Inks for Epson 2880

[Robert Boire]

My understanding is that the reason that we have black inks on ink jet printers is because we can never get deep enough blacks with the normal cyan, magenta and yellow inks.  So my question is, in an image with no blacks or greys (ie fully color) do these inks get used at all? Do they have an effect on the color rendition of say green, yellow and blue?

(this is sort of related to another pos that I continue to struggle with (http://forum.luminous-landscape.com/index.php?topic=110239.0)

Robert

[Mark D Segal]

My understanding is that the reason that we have black inks on ink jet printers is because we can never get deep enough blacks with the normal cyan, magenta and yellow inks.  So my question is, in an image with no blacks or greys (ie fully color) do these inks get used at all? Do they have an effect on the color rendition of say green, yellow and blue?

(this is sort of related to another pos that I continue to struggle with (http://forum.luminous-landscape.com/index.php?topic=110239.0)

Robert

Yes, my understanding is that they do get mixed with other colours.

As far as your other problem, if you are still experiencing incorrect results after insuring that there is no double colour management, the correct profile is selected, the nozzle check is fine and the right media is selected in Media Type relative to the Media type you are using and for which the profile is keyed, then perhaps it is time for a new printer. A 2880 is VERY long in the tooth by now.

[Robert Boire]

Thanks. I doubt its the printer. It may be old, but I do not use it that much. And buying a new printer is an expensive option until I have exhausted everything else.

But maybe this is a color management problem after all that I do not quite understand.

I printed two sets of test images. Image 1 is virtually completely blue, Image 2 is the one that is driving me crazy (the one with the shift to purple).

In the first set, I printed with color management by LR, turned off color management in the printer, selected the right profile for the Epson paper (as supplied by Epson) in LR etc

In the second set, I printed with color management by printer, turn on auto color management in the printer.

In both sets Image 1 is virtually identical and no blue to purple shift.

In the case of Image 2, the second set (CM by printer) does NOT show the purple shift and is pretty close to what I expect. Yet I would expect both cases are using the same Epson profile.

So I am a bit lost.

[Mark D Segal]

If "printer manages colour" is giving you about correct results, then you are most probably right there is a settings issue in the workflow with colour management through LR.

In the one printed with colour management through LR, not the printer, did you check in Printer Settings (in the Epson driver interface) whether you have the correct Media Type setting for the paper and profile you are using?

[Robert Boire]

In the one printed with colour management through LR, not the printer, did you check in Printer Settings (in the Epson driver interface) whether you have the correct Media Type setting for the paper and profile you are using?

Yes, a million (or almost) times. The Epson support person suggested that it could be a LR bug or some software corruption. But I doubt that. Its very repeatable.

[Mark D Segal]

Bugs can be, and usually are, repeatable. While it is unlikely to an LR bug, however, perhaps worthwhile doing a test by printing it through Photoshop with Photoshop managing colour to see whether you get a better result. If you do, it could be some stray LR setting that needs detection, though there aren't that many. Are there any other photos that do the same purple shift - i.e. is it systemic or just this one photo?

[Robert Boire]

Yes, there are others, they can be fixed by printing with Printer Manages color and are very repeatable.  What I find wierd is the fact that Image 1 (which is virtually 100% blue) looks fine in both cases.

[Mark D Segal]

Blue is for the most part a mixture of Cyan and Magenta. A purplish cast in the Blue sounds as if there is an over-emphasis of Magenta in the mix. If it works fine with Printer Manages Color, this suggests a problem with the ICC profile that becomes apparent with the need for hue mixtures that are not nearly as saturated as 100% Blue. You should get a hold of Bill Atkinson's printer test target or the Outback version as these have known reference colour values and print it using your ICC profile colour managed workflow through both Photoshop and Lightroom and see how they look. This may help detect what's going on or rule out what isn't.

[Robert Boire]

Just as a lark, I printed to a completely different paper on a different printer. Selected LR does CM. Choose the Epson 2880 profile for the paper (its all I have) but printed on a really cheap canon ip4300 printer (which does not have a profile for the Epson paper). Choose the closest Canon media type on the printer. Result - pretty crappy overall  (as expected) but no color shift.

Then realized I forgot to turn off CM on the Ip4300. Repeated above with CM turned off on the printer side. Result - pretty crappy overall but got the blue to purple color shift.

Printed again with printer manages color. Result - no blue to purple shift.

So I am getting more or less the same results on two different printers. This lead me to conclude (I think) that when I print from LR and use the paper profile, LR is doing something wierd with the profile...regardless of what printer I use.

[Robert Boire]

You should get a hold of Bill Atkinson's printer test target or the Outback version as these have known reference colour values and print it using your ICC profile colour managed workflow through both Photoshop and Lightroom and see how they look. This may help detect what's going on or rule out what isn't.

Thanks, our posts crossed each other. I will try what you suggested. Stay tuned...

[Mark D Segal]

I don't think LR is doing anything weird with the profile. There is probably something weird about the profile itself. OK, let's see what happens next.

[Robert Boire]

The reason I suspected the profile is that when LR does the CM, I am using the Epson profile for that paper which comes with the printer.  When I print with the printer doing CM I assume that internally the printer is also using the same profile...which gives decent results...or did I get that wrong.

[Mark D Segal]

The reason I suspected the profile is that when LR does the CM, I am using the Epson profile for that paper which comes with the printer.  When I print with the printer doing CM I assume that internally the printer is also using the same profile...which gives decent results...or did I get that wrong.

OK we're both suspecting the profile, which so far seems a reasonable hypothesis; all I'm saying is that this is most likely not the fault of Lightroom because colour management is handled in the Color Management Module of the computer's operating system. Lightroom is being used to call-up the profile you want to work with.

Now what is being used internally to concoct the colour recipe with "Printer Manages Color" I do not know. Perhaps others on this Forum do. It may not necessarily be the profile used in an ICC color-managed workflow.

[Robert Boire]

OK we're both suspecting the profile, which so far seems a reasonable hypothesis; all I'm saying is that this is most likely not the fault of Lightroom because colour management is handled in the Color Management Module of the computer's operating system. Lightroom is being used to call-up the profile you want to work with.

 workflow.

So LR is simply telling the CMM what profile to use during the printing, but is not manipulating the profile itself?

The thing that makes me suspect a software issue is because I have tried this with three different Epson papers (and their profiles) and get the same results in terms of the color shift.

[Mark D Segal]

So LR is simply telling the CMM what profile to use during the printing, but is not manipulating the profile itself?

The thing that makes me suspect a software issue is because I have tried this with three different Epson papers (and their profiles) and get the same results in terms of the color shift.

LR does not manipulate profiles. Nor does it tell the CMM what profile to use - you do that using LR as the platform. LR does handle the processing of image data; the reason I suggested that you try printing the same photos through Photoshop was to try to isolate whether the problem could be something in LR itself, unusual as that would be.

The more information you put out raises more questions, but perhaps not all the essential information yet. What computer operating system are you using, what version of LR are you using, and can you post screen grab samples of the photos that are affected by the colour shift along with low-res JPG scans of the printed results? We're talking in the abstract without those basics on the table. Given the age of the printer I'm wondering whether there could be some kind of incompatibility that others could point out if the whole image preparation and processing  environment and the visual character of the issue you're having were known.

[Robert Boire]

Mark,

First of all thanks for your help so far. 

FYI I am using LR 6.5.1 (though I have the same problem with 6.4) and WIndows 7, though I do not think it matters anymore.

It looks you are right, that it appears to be the profile(s). I do not have Photoshop, but I was able to print to the Epson from Canon's DPP, again with CM by DPP and then CM by printer. Same result. The original image is attached. The problem area is at the bottom. I do not have a scanner, but I took a picture of the printouts. The results are not very accurate, but I think you can see what I am talking about, especially if you zoom in to the bottom of each image in the second attachment. The top image is with CM by DPP (bottom is decidedly purple) and the bottom with CM by the printer.

What I find disturbing is that I got similar results with three different papers, which implies a consistent problem with all the profiles.  I guess I am at a dead end now, short of creating custom profiles.  BTW all colors are in gamut, at least according to LR's soft proof.

[Mark D Segal]

OK, this is useful. By replicating the same problem with two different printing applications, we can assume the problem is not Lightroom, much as I had expected. BUT, that three separate profiles should be causing the same problem is very puzzling, especially if these are Epson profiles for their own papers. It leads one to suspect that something else is causing the problem that is neither in the profiling or in the application. If all colours are in gamut, the color space and the rendering intent would not seem to be the locus of the issue either. Somewhere somehow the composition of the blue hue is being compromised by either a lack of Cyan or an excess of Magenta. The one variable that's left is the printer. I'm still wondering whether given the age difference between the printer and the operating system there is some kind of disconnect between the CMM in the OS and the printer driver/firmware. One more thing I can think of is to be sure you have the latest driver and firmware installed that is specified for your OS version and available for that printer.

[Robert Boire]

The driver is up to date. And anyway as I mentioned somewhere up above, I actually tried printing using another printer, with more or less the same results, in terms of the color shift.

I should mention that one of the papers I tried is Exhibition Fibre, which is interesting because the profile does not even come from Epson but from PixelGenius.

Its a real conundrum.

[Mark D Segal]

I kind of discounted the test with the other printer because you didn't have a correct profile for the printer/paper combination, so it's hard to be sure that anything of reliable indicative value comes from that test, and as you mentioned, it is a very basic printer. It could be that both printers cannot manage to reproduce the particular colour value you need from that photo because of gamut limitations. The gamut warnings aren't necessarily all that accurate in borderline cases. There is one more thing we can try - I shall send you a PM through this website.