Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: Legolas on May 27, 2015, 06:27:32 am
-
I encoutered two likely related problems for which i don't have an answer. I hope that someone can help me.
A new icc profile that was created for me prints too dark (about 2 stops) but softproofing in LR6 does not show this issue.
What have i done so far.
I printed three i1profiler test chart pages (1600 patches) using Adobe Color Print Ultility (ACPU) in order to let a company create a new Epson PGPP profile for me.
For this, i set the Epson 3800 to PGPP and 'no color adjustments' (similar to what i use in LR for printer presets). When i got the new icc-profile it look much better on the (calibrated screen) with softproofing. Several issues i had with the original epson PGPP profile were gone. However, printing with LR6 using the new profile is about two stop darker than what i see on the screen with softproofing.
That is the first thing i don't understand. Why does softproofing not show a darker output? Note that printing and softproofing in LR6 by using the standard Epson profile does not have this issue at all.
I reviewed the profile with GamutVision but could not find an issue (not that i am not an expert on this so i am not sure what i need to look for).
My next thought was that possibly the patch file prints were wrong. To test this, i printed the patch files with LR6 by using Profile: 'managed by printer' and used the normal PGPP printer settings (no color management, etc). In my view that should give the same result as printing with ACPU but it doesn't. The print results are very different.
How can ACPU and LR6 with option 'profile managed by printer' with printer setting identical, result in different prints for the patch files. Would this be the cause of the darker prints?
Am i missing something?
Many thanks in advance,
Alexander
I am using a Epson 3800 printer, PGPP paper, LR6, Windows 7
Now it gets
-
A new icc profile that was created for me prints too dark (about 2 stops) but softproofing in LR6 does not show this issue.
Do you see a big two stop difference when you turn on and off the soft proof?
If so, it's the profile. Profiles have two tables. One affects output, the other the soft proof. It's possible the two are out of sync if it's not a good profile.
But if the soft proof on or off looks about the same (you don't see a two stop difference toggling), here's what you need to read:
http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml
-
The one point you didn't mention is whether in the Epson driver you have checked to make sure you have the correct Epson reference paper selected for which your profile was made. If for some reason the wrong reference paper were in place in the Epson driver, there would be an inconsistency of ink lay down between what the profile wants and what the driver is doing.
-
Hi Digitaldog & Mark,
Many thanks for your answers.
If i turn softproofing on and off, there is surely no 2 stops difference on the screen. There obviously differences in color, but not in luminance.
When printing the patch file, i made sure that the medium type in the printer settings is PGPP, the paper i am using. I also checked every other setting in the 3800 driver and that look fine to me.
The two stops darker print _only_ happens with this new profile, not with many other standard profiles i use. This is also the first time i want to have a custom profile for my printer/paper combination.
I have read the contents of the url, i can not find any issue with my setup since and note all the other standard profiles i use are fine (no difference). Note that a friend of mine has used the same profile on his Epson 3800 and he gets the same issue.
What i also do not understand why patch chart prints with ACPU and LR6 using 'profile managed by printer' produce significantly different result. I would expect the same result. I made sure the 3800 setting were the same in both cases. I have the feeling that the clue might be somewhere in this area but is don't see it.
The profile is based on the ACPU chart prints while i normally print my photos with LR6.
-
If i turn softproofing on and off, there is surely no 2 stops difference on the screen. There obviously differences in color, but not in luminance.
Sounds like the profile is fine and your display calibration is way off.
I have read the contents of the url, i can not find any issue with my setup since and note all the other standard profiles i use are fine (no difference). Note that a friend of mine has used the same profile on his Epson 3800 and he gets the same issue.
Odd. Triple check all the print settings. As Mark points out, I suppose there's the possibility the targets for the profile were not printed correctly. I'd suspect both tables would be way off.
What i also do not understand why patch chart prints with ACPU and LR6 using 'profile managed by printer' produce significantly different result. I would expect the same result. I made sure the 3800 setting were the same in both cases. I have the feeling that the clue might be somewhere in this area but is don't see it.
LR can't print without color management like ACPU can, that's why you use ACPU to print the targets. The targets have to be printed in ACPU and will not look anything like the same target printed in LR.
-
Hi Andrew,
Actually my comment was more directed at the driver paper setting for making prints of image files using the new profile, but as you suggest here, it could also be a problem reaching back to the printing of the profiling targets - so in both circumstances a factor to just double-check and be sure the problem is not as simple as that.
Now one thing more - Andrew - the OP is on Windows, so that raises another question in my mind. Do we have the same colour-management stickiness issue with Windows as we do on Mac? I seem to recall ACPU was originally developed to overcome the problem Adobe encountered back in 2010 when Apple discontinued the ability to turn off colour management at the OSX level. Did Microsoft follow suit? i.o.w. I'm wondering whether he needs ACPU for Windows, and from there, whether in some weird and wonderful way the combination of ACPU+Windows has caused his profiling target to be off?
-
Now one thing more - Andrew - the OP is on Windows, so that raises another question in my mind. Do we have the same colour-management stickiness issue with Windows as we do on Mac? I seem to recall ACPU was originally developed to overcome the problem Adobe encountered back in 2010 when Apple discontinued the ability to turn off colour management at the OSX level. Did Microsoft follow suit? i.o.w. I'm wondering whether he needs ACPU for Windows, and from there, whether in some weird and wonderful way the combination of ACPU+Windows has caused his profiling target to be off?
ACPU was created because Adobe yanked the No color management option (and the null profile trick later) from Photoshop.
ACPU seems to work on both platforms, it's how all my customers are told to print targets for a custom profile. Aside from the occasionally scaling bug.
-
ACPU was created because Adobe yanked the No color management option (and the null profile trick later) from Photoshop.
ACPU seems to work on both platforms, it's how all my customers are told to print targets for a custom profile. Aside from the occasionally scaling bug.
OK - good to know.
On the history of it - perhaps my memory is failing me, but I was under the impression Adobe yanked the No Color Management Option because Apple made it impossible for it to work, so they decided to pull it and offer ACPU as the "fix" for those wanting to profile their printers properly.
-
On the history of it - perhaps my memory is failing me, but I was under the impression Adobe yanked the No Color Management Option because Apple made it impossible for it to work, so they decided to pull it and offer ACPU as the "fix" for those wanting to profile their printers properly.
I think that's accurate. They (Adobe) had all kinds of printing issues with Apple and removed this option.
-
Hi Mark and Andrew,
Thanks for all the background on ACPU and Mac color management.
I used ACPU and LR6 on Windows 7.
@Andrew, you mentioned that LR can not print without color management. There is an option in LR that looks like that in the print module: 'profile' in the color management section. This can be set to a profile or set to 'managed by printer'. I assumed the latter would indicate that LR would not apply any color management. That is what i used to compare the ACPU target prints with the LR target prints. If you are right, it would explain why LR and ACPU results are different.
I conclude it was right to use ACPU to print the targets.
I am not sure what to do next...
I can ask the company to create a new printer profile based on a new set of profile targets to make 100% sure all driver settings are right.
Any other suggestions are welcome.
Alexander
-
@Andrew, you mentioned that LR can not print without color management. There is an option in LR that looks like that in the print module: 'profile' in the color management section. This can be set to a profile or set to 'managed by printer'. I assumed the latter would indicate that LR would not apply any color management.
It does indeed. Again, there's no way to send the target values 'as is' to the printer in LR, you must print the targets in ACPU or the product that created the targets (which presumably and usually prints without color management). If you got a profile made for you remotely, you don't have access to that software.
http://digitaldog.net/files/DDPrintingTargetsVideo.mov
-
I'll will get the new profile based on new target prints in a few days and i will report on the results. Many thanks for you help.
Alexander
-
I'll will get the new profile based on new target prints in a few days and i will report on the results. Many thanks for you help.
Alexander
Alexander, I assume you know the drill now: only use ACPU for printing the targets, make sure the choice of paper is correct in Print>Printer Settings>Media Type, and then use that same group of settings when you get to making prints with the new profile.
-
Hi Mark,
Yes, i made 100% sure all the setting were correct when i printed the new targets. Media Type is also right :-). I made scrrenshots of the printer settings so i can always refer to these. Let's what for the results. That you so much for your support.
Alexander
-
I'll will get the new profile based on new target prints in a few days and i will report on the results. Many thanks for you help.
Once you do, it will be possible with the right tools to see what was different, if anything between the two. That could point out what went wrong (if anything).
-
OK - good to know.
On the history of it - perhaps my memory is failing me, but I was under the impression Adobe yanked the No Color Management Option because Apple made it impossible for it to work, so they decided to pull it and offer ACPU as the "fix" for those wanting to profile their printers properly.
Doesn’t it seem a little odd that they can write a separate utility that works, and yet can’t put the same concept inside PS so it also works?
-
Doesn’t it seem a little odd that they can write a separate utility that works, and yet can’t put the same concept inside PS so it also works?
Maybe it's because ACPU is free, but Photoshop is not, and they make it available to all and sundry, not only Photoshop users.
-
Doesn’t it seem a little odd that they can write a separate utility that works, and yet can’t put the same concept inside PS so it also works?
They probably could but based on the back and forth between Adobe and Apple to fix each other's bugs, I suspect it was vastly less expensive to remove a 'feature' no one needed** and build that into a standalone that will likely never be updated.
** printing without color management to output targets for building profiles is the task of the product that created the targets.
-
Doesn’t it seem a little odd that they can write a separate utility that works, and yet can’t put the same concept inside PS so it also works?
Nope, not at all. Adobe was essentially FORCED to remove the No Color Management by Apple. And yes, Apple had a nasty habit of breaking the pint pipeline as well as color management that the Photoshop engineers were forced to work around.
And, I agree with Andrew that the app building the profiles should be the app to print from. Course that's kinda hard for remote profiling serves but ACPU solves that problem.
-
** printing without color management to output targets for building profiles is the task of the product that created the targets.
And so if Adobe, with all their resources and clout can't get this straightened out, the very much smaller with no pull at all with apple color management vendors are supposed to do it ?
-
Doesn’t it seem a little odd that they can write a separate utility that works, and yet can’t put the same concept inside PS so it also works?
Adobe's states reason was that they wanted a separate utility so that they could release fixes at a faster pace than Photoshop and their other products. (Photoshop needs a lot more QA before release, because people yell and scream if things get broken.)
-
And so if Adobe, with all their resources and clout can't get this straightened out, the very much smaller with no pull at all with apple color management vendors are supposed to do it ?
Yes! Other than printing targets for a profile, please tell us why anyone using Photoshop would need to send data out without color management?
I have one, it's not at all a critical function, you want to send out an image on one page using a profile to test two rendering intents. But it's again, a function the vast majority of PS users would need to do AND they can do it with ACPU.
-
Yes! Other than printing targets for a profile, please tell us why anyone using Photoshop would need to send data out without color management?
That is reason enough. There is no point in using Photoshop if your prints look terrible because no-one can profile the printer, because there is no way to print without color management.
It's about as sensible as a car manufacturer saying "no user will change the oil in the engine, so we'll omit the drain plug, oil filler and dip stick", or an MP3 player with no way of loading songs on to it "because 99% of the time people only play, not record".
-
And so if Adobe, with all their resources and clout can't get this straightened out, the very much smaller with no pull at all with apple color management vendors are supposed to do it ?
X-Rite can print using No Color Management and other CM companies as well. Again, you forget that Apple is in control of the Mac OSX print pipeline...
-
Again, you forget that Apple is in control of the Mac OSX print pipeline...
Given the years it took to sort out the color problems between Adobe and Apple, I don't thing "in control" is an appropriate term...
-
Given the years it took to sort out the color problems between Adobe and Apple, I don't "think" "in control" is an appropriate term...
I don't "think" you understand the relationship between; the OS level print pipeline, the print driver & the print application. It's a very complex relationship. If Apple dictates a certain behavior, both the print driver (Epson or Canon) and the application must play along. Apple has dictated that the print driver and the application cease allowing a behavior that allows a No Color Management behavior. The applications that are dedicated to creating printer ICC profiles get a pass, for several reasons, most notably the needs of the user base. But not general apps.
But for mainline apps, the No Color Management option became unsustainable. Hence the removal of that option from the print delog of Photoshop under Mac. Note, this is NOT an issue under Windows because in this case. MSFT doesn't impose this requirement.
It's ironic that in this case, MSFT is more "free" than Mac, but there you have it. It sucks (in this small case) to be using the Mac OS.
-
........... It sucks (in this small case) to be using the Mac OS.
Does it really Jeff? - ever since we've had ACPU we've at least been able to generate our profiling targets correctly, and I'm finding the print pipeline works rather seamlessly on OSX 10.x, especially printing from LR, but even in PS; one would wish for a more user-friendly layout of the driver settings, but something we get accustomed to sooner or later.
-
OSX's ColorSync Utility has similar functionality..........
-
That is reason enough. There is no point in using Photoshop if your prints look terrible because no-one can profile the printer, because there is no way to print without color management.
I'm not following you. You're suggesting no one can produce a good print in Photoshop because they don't have profiles because Photoshop can't print an untagged target?
It's about as sensible as a car manufacturer saying "no user will change the oil in the engine, so we'll omit the drain plug, oil filler and dip stick", or an MP3 player with no way of loading songs on to it "because 99% of the time people only play, not record".
The analogy doesn't wish IMHO. We do need to change our oil. I asked you to produce a scenario other than printing an untagged target for a profile the Photoshop user would need. I even provided one, albeit something few people would ever do. There are dozens of products that can print an untagged target for a profile, many of them the products that created the target in the first place as those products SHOULD provide. So short of that, can you tell us why Photoshop needs to supply a non color managed path for untagged data, especially when they proved a free cross platform utility to do so, aiding folks (no names) who's products must output untagged profile targets.
-
I don't "think" you understand the relationship between; the OS level print pipeline, the print driver & the print application. It's a very complex relationship.
Having implemented several such workflows and pipelines, I think I have a better understanding of it than most people, and some insights that Apple don't appear to have.
But for mainline apps, the No Color Management option became unsustainable.
"Unsustainable" - what on earth does that mean ? It's using too many electrons ?
There are elegant ways of handling a calibration workflow, but Apple have chosen to do it badly, and disenfranchise anyone who wants to have a robust path to achieve accurate color.
It's ironic that in this case, MSFT is more "free" than Mac, but there you have it. It sucks (in this small case) to be using the Mac OS.
It seems you have brought yourself to terms with it, but as a "Maker", it just seems completely asinine to me.
-
There are dozens of products that can print an untagged target for a profile, many of them the products that created the target in the first place as those products SHOULD provide.
No doubt there are workarounds. But since you have to switch to a different application to print the test chart, how do you know that the processing is the same, i.e. that the profile thus generated, will work as intended with PhotoShop ? The answer is you don't - you either take it on trust (and we know how often color goes wrong because there are too many cooks, and because it's seen as the "right thing" to hide everything from the user), or you need to check it in some way.
One upon a time, Photoshop was the tool you would use to do these checks, because you could see exactly what was going on. You knew which colorspace your raster file was in, and you knew what processing (if any) was going to be applied when you printed it.
-
"Unsustainable" - what on earth does that mean ?
I think it means it got very expensive to support a 'feature' so few needed after the back and forth bug fixing and then reintroduction of new bugs between Apple and Adobe.
Clearly Adobe can print without color management. They did it in the past, they coded ACPU (scaling bugs not withstanding). They saw an expensive proposition keeping a 'feature' so few needed and felt I believe, as I do, printing untagged data for the creation of profiling was the task for others. So they said, screw it. I can't blame them. X-rite has a current product that prints targets without color management. The CS utility can do it. While I haven't checked in years and years, presumably DataColor can do this with their Spyder-whatever product. QTR-Print-Tool can do this task too.
-
No doubt there are workarounds. But since you have to switch to a different application to print the test chart, how do you know that the processing is the same, i.e. that the profile thus generated, will work as intended with PhotoShop ?
Me personally? I test it. Measure both targets from each products, load the measurement data into ColorThink and produce a dE report.
I've seen no difference in recent versions of i1Profiler, the Copra app and ACPU in printing targets and I've seen zero issues using the generated profile in Lightroom or Photoshop. As we've seen in the past, that could change. But in no way do we have to print the profile targets IN Photoshop! Unless you have some additional data points I'm not aware of, multiple products that say they can print an untagged target do so correctly and consistently and the generated profile works in all the product I and my clients use when printing with the generated profile. Again, that could change, but it's been bankable for a number of years now. So I see no problem here.
-
I agree - my own profiling using ACPU for printing the targets has achieved very satisfactory results.
-
I received a new icc profile for the Epson PGPP paper. I can report that the issue is resolved. The prints from LR6 are now very good (colors are much better now compared to the standard epson icc profile) and the 2 stop darker problem as gone. I conclude that i have made a mistake in generating the target file using ACPU. Likely i forgot to set Media Type correctly.
I would like to thank you for all your help, very much appreciated!
Alexander
-
Very good! Glad to see it resolved.
-
X-Rite can print using No Color Management and other CM companies as well.
Just for the sake of clarity this is not entirely true with X-Rite’s ColorMunki software. On the Mac you get a choice of either accurate printer profiles (software v1.2.1) or accurate display profiles (software v1.1.1) – but not both with either software versions. See ColorMunki postings in this forum.
-
Just for the sake of clarity this is not entirely true with X-Rite’s ColorMunki software. On the Mac you get a choice of either accurate printer profiles (software v1.2.1) or accurate display profiles (software v1.1.1) – but not both with either software versions. See ColorMunki postings in this forum.
I'm sure Jeff's point is, all the X-rite bugs aside, the product is supposed to provide both functionality.