A classics revisited: My prints turn out too dark!

[digitaldog]

But are you saying that even in the case of hardware calibration it is preferable to remain close to the generic white poit. BTW what do you mean by generic white point of the monitor.
From what I understand, the best for me would be to set the monitor to the best possible paper independent state. This should be where I can get the lowest DeltaE. Only, the next step involves soft proofing and print to paper match, right?

First off, while the native white point may produce the least banding, the question is, do you get good print to display matching with that native WP? If not, I suspect most of us would take a bit of banding to counter a display that is too warm or cool compared to the print.

Second, the deltaE values are kind of meaningless unless they are very high, indicating an issue somewhere (you are after all, using the same instrument to report on itself that you used to build the calibration). Again, do you have a good screen to print match? I’d take a higher deltaE value considering the above use of the same instrument if I got a match than a lower deltaE value that didn’t match.

Lastly, the paper profiles are in the chain here when you soft proof so I’m not sure what you mean by set the monitor to the best possible paper independent state. I’d get a good reference image and print it out with a good paper profile. I’d pop that into my viewing booth and load the soft proof correctly (paper and ink simulation on, full screen mode, no palettes) and my goal would be screen to print matching. I’d calibrate for that aim. Switch paper profiles and load the soft proof, all should be fine (again assuming good profiles who’s tables for preview and output are working as they should).

[tho_mas]

Here we go again :-)

From what I understand, the best for me would be to set the monitor to the best possible paper independent state.

if I understand you correctly "paper indepent" means a state, where the monitor matches paper white visually but without further tweaking of color saturation (either in the monitor hardware - actually not recommended unless you really know what you are doing - or on a color layer in Photoshop).
If so: yes, that's fine.

This should be where I can get the lowest DeltaE. Only, the next step involves soft proofing and print to paper match, right?

yes, correct.

From here on it's up to you... the thread is full of talk about the visual match of monitor and paper white and Andrew ("digitaldog") and I agree basically on most things - except for the white reference.
Andrew suggests to calibrate to a paper white with softproof enabled (and paper simultion activated) whereas I suggest to calibrate to paper white with softproof disabled (so if you want so, to calibrate for an average white or your preferred paper ... or, if you can get one, for the white of a proof paper). Consequently I do not enable SPC when softproofing - I only enable "simulate black ink".
So Andrew's display is probably a bit more accurate for that one particular paper with SPC enabled but off for anything else.
My display is neutral to the ambient light I am working under (all D50 here) ... but the white point of some particular papers might be a bit colder or a bit warmer in some cases. In my case they all match quite fine as long as we talk about coated offset printing papers and the range of photographic papers I am using. If I'd also use papers with warmer tones I'd probably simply make a second monitor profile for those...

Both ways will work fine - your choice ;-)

[nilo]

First off, while the native white point may produce the least banding, the question is, do you get good print to display matching with that native WP? If not, I suspect most of us would take a bit of banding to counter a display that is too warm or cool compared to the print.

Second, the deltaE values are kind of meaningless unless they are very high, indicating an issue somewhere (you are after all, using the same instrument to report on itself that you used to build the calibration). Again, do you have a good screen to print match? I’d take a higher deltaE value considering the above use of the same instrument if I got a match than a lower deltaE value that didn’t match.

Lastly, the paper profiles are in the chain here when you soft proof so I’m not sure what you mean by set the monitor to the best possible paper independent state. I’d get a good reference image and print it out with a good paper profile. I’d pop that into my viewing booth and load the soft proof correctly (paper and ink simulation on, full screen mode, no palettes) and my goal would be screen to print matching. I’d calibrate for that aim. Switch paper profiles and load the soft proof, all should be fine (again assuming good profiles who’s tables for preview and output are working as they should).

On this last point: If I do as you said, popping a print into the viewing booth and trying to get the match, which is what I did, I'll get a good color temperature match for that paper, but not for another. So, when I switch paper profiles and load the soft proof, that should not be that fine, no?

By paper independent state, I meant, if that makes sens, the state that someone who does not print profiles/calibrates for.

Finally how can I tell if I have a good profile who's tables for preview and output are working as they should?

[nilo]

Here we go again :-)
 if I understand you correctly "paper indepent" means a state, where the monitor matches paper white visually but without further tweaking of color saturation (either in the monitor hardware - actually not recommended unless you really know what you are doing - or on a color layer in Photoshop).
If so: yes, that's fine.
yes, correct.

From here on it's up to you... the thread is full of talk about the visual match of monitor and paper white and Andrew ("digitaldog") and I agree basically on most things - except for the white reference.
Andrew suggests to calibrate to a paper white with softproof enabled (and paper simultion activated) whereas I suggest to calibrate to paper white with softproof disabled (so if you want so, to calibrate for an average white or your preferred paper ... or, if you can get one, for the white of a proof paper). Consequently I do not enable SPC when softproofing - I only enable "simulate black ink".
So Andrew's display is probably a bit more accurate for that one particular paper with SPC enabled but off for anything else.
My display is neutral to the ambient light I am working under (all D50 here) ... but the white point of some particular papers might be a bit colder or a bit warmer in some cases. In my case they all match quite fine as long as we talk about coated offset printing papers and the range of photographic papers I am using. If I'd also use papers with warmer tones I'd probably simply make a second monitor profile for those...

Both ways will work fine - your choice ;-)

Ok. I got confused again, but now after your post in which you briefly recap the thread, it's clear again.

Thanks once more to both of you for clearing this issue up. It was, once again, very instructive, helpful and interesting.

[digitaldog]

On this last point: If I do as you said, popping a print into the viewing booth and trying to get the match, which is what I did, I'll get a good color temperature match for that paper, but not for another. So, when I switch paper profiles and load the soft proof, that should not be that fine, no?

That’s not been my experience, but I’ll admit that differing papers could be an issue. When you switch paper profiles however, the information about the paper white should of course be updated in the soft proof.

Finally how can I tell if I have a good profile who's tables for preview and output are working as they should?

You get good matches <g>

[tho_mas]

On this last point: If I do as you said, popping a print into the viewing booth and trying to get the match, which is what I did, I'll get a good color temperature match for that paper, but not for another. So, when I switch paper profiles and load the soft proof, that should not be that fine, no?

Nino, when you look at the four paperas you are using frequently do have the impression that they are significantly different?
So one is cleary blueish/greyish (i.e. dark), one is blueish and bright, one is magentaish and one is yellowish?
Or are we talking about very subtile differences... so one is slighty warmer/colder/darker than the other but actually they are all quite similar?

Now make a white 800x800 pixel wide square in Photoshop (either way how your display is calibrated for the time being) and enable softproof with SPC enabled for the respective four papers.
Do you have the impression that the simulation of paper white in Photoshops represents the real differences the papers show visually or do you have the impression the differences in Photoshop are more emphasised... or maybe even less?

[digitaldog]

So Andrew's display is probably a bit more accurate for that one particular paper with SPC enabled but off for anything else.

I don’t know about that. But what I can do with the SpectraView software is build multiple calibration target aim points (and associated ICC profiles) and switch on the fly which is really useful. I’m also building these various calibration targets at differing contrast ratio’s depending on the papers.

[nilo]

That’s not been my experience, but I’ll admit that differing papers could be an issue. When you switch paper profiles however, the information about the paper white should of course be updated in the soft proof.

You get good matches <g>

I'll try to calibrate for each paper white in a separate monitor profile, as it is very to switch between them (and to forget to switch between them  )

Regarding good profiles, I would like to add, that from my recent experience with the manufacturer's profile for CIFA Baryta Photographique on 3880/k3VM, what ever I did and tried, no satisfying  match was achievable.

[nilo]

Nino, when you look at the four paperas you are using frequently do have the impression that they are significantly different?
So one is cleary blueish/greyish (i.e. dark), one is blueish and bright, one is magentaish and one is yellowish?
Or are we talking about very subtile differences... so one is slighty warmer/colder/darker than the other but actually they are all quite similar?

Now make a white 800x800 pixel wide square in Photoshop (either way how your display is calibrated for the time being) and enable softproof with SPC enabled for the respective four papers.
Do you have the impression that the simulation of paper white in Photoshops represents the real differences the papers show visually or do you have the impression the differences in Photoshop are more emphasised... or maybe even less?

I am impatient to try this out! I'll report back on that.

[nilo]

I don’t know about that. But what I can do with the SpectraView software is build multiple calibration target aim points (and associated ICC profiles) and switch on the fly which is really useful. I’m also building these various calibration targets at differing contrast ratio’s depending on the papers.

With the new PA series, one could also just load the profile into the monitor's 3D-Lut?! Nothing else should be necessary?! Maybe fine tuning?!

[tho_mas]

I don’t know about that. But what I can do with the SpectraView software is build multiple calibration target aim points (and associated ICC profiles) and switch on the fly which is really useful. I’m also building these various calibration targets at differing contrast ratio’s depending on the papers.

my issue with your workflow is that Photoshop's palettes and anything white on the monitor that is not affected by Photoshop's softproof settings is actually not a neutral white anymore. I prefer a monitor that is neutral to the ambient light I am working with.
I can handle the response curves of the papers I am working with much, much better with layers in Photoshop... and these layers only contain very subtile adjustments.
The monitor is my "default" ... the respective papers are only certain targets that need some (minor) adjustments.
But I certainly do not doubt that you personally get very good results with your workflow - not at all!

[tho_mas]

I am impatient to try this out! I'll report back on that.

do it!
I bet Photoshop will show you something like this (from page 1 of this thread): http://www.luminous-landscape.com/forum/index.php?action=dlattach;topic=46010.0;attach=23650;image
Have you EVER seen a printing paper that looks like wrapping paper?
Now someone will tell you these profiles are "wrong" or "not good" or contain "bad reverse tables".
Hell, these are most likely very, very good profiles. But at least that's what you get in 99% of the cases unless someone takes the time to tweak these profiles for really accurate softproofing.
As long as Photoshop tells me that one of these fancy candy colors is a "paper white" I simply ignore the simulation of paper white... it won't work with any of the papers shown in the image linked above.

[nilo]

do it!
I bet Photoshop will show you something like this (from page 1 of this thread): http://www.luminous-landscape.com/forum/index.php?action=dlattach;topic=46010.0;attach=23650;image
Have you EVER seen a printing paper that looks like wrapping paper?
Now someone will tell you these profiles are "wrong" or "not good" or contain "bad reverse tables".
Hell, these are most likely very, very good profiles. But at least that's what you get in 99% of the cases unless someone takes the time to tweak these profiles for really accurate softproofing.
As long as Photoshop tells me that one of these fancy candy colors is a "paper white" I simply ignore the simulation of paper white... it won't work with any of the papers shown in the image linked above.

I get your point and I remember this post of yours now (It was a long break).

[tho_mas]

I get your point and I remember this post of yours now (It was a long break).

no trouble at all.
Maybe your profiles are fine! Who knows. If so: fine!
But if not... use your eyes... rather than following "rules" derived from colormetric theory.

[nilo]

I think you both noticed this new thread. "Simulate Paper Color:Don't know anymore" http://www.luminous-landscape.com/forum/index.php?topic=47185.msg393058#new

regards nino

[tho_mas]

I think you both noticed this new thread.

yes, but this has been an ongoing topic over years and this or the other thread won't solve anything.
You really have to find out what works best for you personally.

What I am honestly finding totally unsatisfying is a statement like: "set the sofproof to rel.col or perceptual and active SPC - it must work."
Well, it must (should). But it doesn't ...mostly.

[digitaldog]

my issue with your workflow is that Photoshop's palettes and anything white on the monitor that is not affected by Photoshop's softproof settings is actually not a neutral white anymore.

Correct and it is an issue. I hope the Lightroom team when they get around to soft proofing fix this. But I find I get much higher degree of print to screen matching, and the previews with the soft proof on are closer with the simulation so I only turn it on when I want to do output specific tweaks on an adjustment layer using dual displays OR when I want to evaluate that the screen and print do match. IOW, I use the soft proof with and without the simulation based on the task at hand.

[tho_mas]

Correct and it is an issue. I hope the Lightroom team when they get around to soft proofing fix this. But I find I get much higher degree of print to screen matching, and the previews with the soft proof on are closer with the simulation so I only turn it on when I want to do output specific tweaks on an adjustment layer using dual displays OR when I want to evaluate that the screen and print do match. IOW, I use the soft proof with and without the simulation based on the task at hand.

yes, I believe this a good workflow. But I also think that the papers (and profiles) you are using allow you to work like that. Unfortunately this is not always possible.

[nilo]

Anyhow good profiles are a prerequisite. I find it very disturbing to having to tweak a bad profile with layers in PS. Till now I did not encounter a relay bad canned profile. I started this whole thread because I used a profile, for which I was unable to get any good match. I thought my color management etc was at fault. While trying to adjust targets in this and that direction in order to compensate, things started to get out of hand. I did not get to use a bad canned profile till now. But the detour, the exercise that followed thanks to both of you, made me learn and understand a lot. So now I even get a lot better match for the good profiles, for which I was not complaining (that much ).

regards

nino

[tho_mas]

so case closed? fine! :-)

one last question... did you check this? (post #65) http://www.luminous-landscape.com/forum/index.php?topic=46010.msg392859#msg392859