Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: Robert Boire on June 21, 2014, 01:52:58 pm
-
Hi,
Not sure if this is really a color management problem but here goes....
I printed an image from LR that has significant blue and purple highlights on the screen. The printed image has shifted strongly to purple so that the blue highlights are largely lost. Any thoughts or suggestions? I have seen a few other similar posts but cannot recall the resolution and I cant seem to find them now.
A few notes:
- Soft proofing does not show the purple shift
- Monitor has been profiled and there is no great variation in the image on different monitors
- EPSON R2880 printer
- color management in printer is turned off
- I am using the right paper profile
Thanks
R
-
Sounds like the profile. It was more common years ago and due in part to the warts in Lab color space. And profiles have two tables, one that affect the output, one the soft proof so having a magenta shift that isn't seen again sounds like a profile issue. Can you provide more info about the profile itself?
-
Are you using the same rendering intent when soft proofing as you do when printing?
Brian A
-
Sounds like the profile. It was more common years ago and due in part to the warts in Lab color space. And profiles have two tables, one that affect the output, one the soft proof so having a magenta shift that isn't seen again sounds like a profile issue. Can you provide more info about the profile itself?
Sure....though I don`t know what exactly you are looking for.
I used Epson Ultra Premium Photo Paper Lustre and the profile for that paper and my printer downloaded from the Epson web site.
I also tried Epson Exhibition FIbre Paper and the profile down loaded from Pixel Genius.
Both gave very similar results.
Are you using the same rendering intent when soft proofing as you do when printing?
Yes, though I don`t think it makes much of a difference since everything is pretty much in gamut.
Thanks
-
> Are you using the same rendering intent when soft proofing as you do when printing?
Yes, though I don`t think it makes much of a difference since everything is pretty much in gamut.
Well perceptual will shift some colors whether they are in gamut or not.
Brian A
-
go to the Bruce Lindbloom website, look under Calc and then Munsell Display Calculator. He has an explanation for the 'blue turns purple' issue.
Bottom line seems to be that it's a printer/profile/capability issue wehre the blues are out of gamut so the math of color substitution turns that OOG blue into a in-gamut purple. Try maybe desat'ing the blue a bit
-
I use a epson4900 and suffered a problem like yours.turned out to be the ink which got too hot, over 50 ' centigrad for a few days in a row. Could it be the same for you?
Or old inks ?
I reckon the blue and yellow.
-
Bottom line seems to be that it's a printer/profile/capability issue wehre the blues are out of gamut so the math of color substitution turns that OOG blue into a in-gamut purple. Try maybe desat'ing the blue a bit
Well that and googling "blue turns purple" (24Million hits) indicates I have some reading to do.
But if I understand correctly this should only be problem of out-of-gamut blue right? My blues are all in gamut..at least according to Lightroom soft proof for the paper.
Nevertheless I will try you suggestion... once I get more ink.
R
-
I use a epson4900 and suffered a problem like yours.turned out to be the ink which got too hot, over 50 ' centigrad for a few days in a row. Could it be the same for you?
Or old inks ?
I reckon the blue and yellow.
Well possibly, I do not use the printer that often. Is there anyway to check short of putting in new ink?
Thanks
-
But if I understand correctly this should only be problem of out-of-gamut blue right?
No, in gamut colors could be affected.
My blues are all in gamut..at least according to Lightroom soft proof for the paper.
The OOG gamut indicators are not accurate.
-
No, in gamut colors could be affected.The OOG gamut indicators are not accurate.
So I thought the whole point of relative rendering intent (i was using relative) is that in-gamut colors are preserved. Did I get that wrong?
Also if the OOG gamut indicators in LR are not accurate, is there an accurate measure? This kind of defeats the purpose of soft-proofing.
-
So I thought the whole point of relative rendering intent (i was using relative) is that in-gamut colors are preserved. Did I get that wrong?
With RelCol, OOG is clipped to the nearest boundary but Perceptual will affect in gamut colors too.
Also if the OOG gamut indicators in LR are not accurate, is there an accurate measure? This kind of defeats the purpose of soft-proofing.
Why is that? OOG overlays and soft proofing are two different functions. OOG overlay predates soft proofing first provided in Photoshop 5. And yes, try taking a few sRGB images into PS or LR and ask to see OOG overlay. Should be none but some overlay shows up. It's not accurate.
-
Why is that? OOG overlays and soft proofing are two different functions.
Ok, maybe they are different functions, I seem them as related because they are both accessed from the soft proof capability. I guess what I am saying is that if the OOG were accurate it would help me determine if the color shift is caused by OOG colors or something else. Well I suppose it is what it is.
And yes, try taking a few sRGB images into PS or LR and ask to see OOG overlay. Should be none but some overlay shows up. It's not accurate.
Dont quite understand. Even the sRGB can be legitmately out of gamut?
-
Print a grey step wedge on a paper for which you have a profile.
If that shows the purplish color cast, probably most visible in the middle greys, it wasfor me the pointer. Even after replacement it will take some prints to replace the old ink in the feed tubes with fresh ink. I do not know how much ink you have left in the printer, but if someone could help you out with a custom paper profile, you can sort of bridge the period to fresh ink?
-
Even the sRGB can be legitmately out of gamut?
Are there colors in sRGB that can fall out of printer gamut? Sure.
-
So problem is resolved..at least partially by 1) changing ink and 2) nozzle clearning. What is curious for me (and exasperating) is the process that took me there.
I consulted Epson who in addition to providing a few useless (and unlikely) suggestions (check your USB cable, print with Windows Photoview...with or without color management I asked..the guy did not seem to know what I meant) suggested cleaning my nozzles. I told him that it was not necessary since the AutoClearn function tests whether this is necessary or not and that the autoclear showed everything was ok. (He did not seem to be aware of the auto clean process and told me to do it anyway. I did the nozzle check everything was fine.)
I decided to print a calibration target for which I have a comparison print. Predictably most colors were ok, except deep blues which were too purple. The printer driver told me I was running low in two colors (light light black and light cyan) and I had assumed that even if I was running low I should be getting the right pigments in the right proportions until the driver told me the cartridges were in fact empty. Well..maybe not.
I replaced that cartridges in question and did another test print. Disaster! Blues were ok but purple (which I had had too much of) and red, among other colors, were gone. The whole result was worse. After a few other test prints, with color management off and on, which were pretty consistent, I checked the nozzle patern again. This time the nozzles were blocked and I needed to clean twice. After that the test print was ok...though the clearning process had pretty much emptied all the other inks.
So my question is this (sorry for being so long winded, I am getting to the point).. Does it make sense that the nozzles need clearning after inserting brand new cartridges, but not before. Is it possible that when a cartridge is running low (but not empty) that I am not getting the right amount of pigment?
Needless to say figuring this out has used a lot of very expensive pigment.
-
So my question is this (sorry for being so long winded, I am getting to the point).. Does it make sense that the nozzles need clearning after inserting brand new cartridges, but not before. Is it possible that when a cartridge is running low (but not empty) that I am not getting the right amount of pigment?
Needless to say figuring this out has used a lot of very expensive pigment.
Unfortunately yes. I used an Epson 4000 and a 4880 for years and I really do feel your pain :'(. What can happen when you change inks (it shouldn't but sometimes does) is that air gets into the ink line. I don't know about the 2880, but with the printers I used what can happen when the ink goes very low in the cartridge is that it causes suction in the line. When the cartridge is removed (or when the new one is inserted, air gets sucked in. The best thing is to do a visual check of your ink lines to see if there is air in them (some air won't cause a problem, but a lot will).
I don't know if the 2880 can do a selective clean of individual inks, but the printers I used could not, so a clean is very wasteful of ink. An alternative is to do a print using just the offending color - that will usually sort things out.
Robert
-
Unfortunately yes. ...
Robert,
Thanks for the useful... if somewhat depressing... insight. Needless to say the Epson tech never mentioned this possibility.
Unfortunately yes. ... An alternative is to do a print using just the offending color - that will usually sort things out.
And a good suggestion. I cannot even see the ink lines let alone flush them selectively. Hopefully I will not need to try your suggestion for some time.
Robert