Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: vikcious on February 10, 2019, 07:01:06 am
-
While I am still digging to find out about the weird differences two spectrophotometers are generating in their respective generated profiles (see my other thread) I am trying to reduce any discrepancies between the testing methods.
I am using two different target printing options:
A) Doug Gray's "null trick" option and B) the usual way (and recommended by Canon) by using Canon Print Studio Pro (PSP) or Canon Professional Print & Layout (PPL) plugins from Photoshop.
Yet I've discovered something that puzzles me.
Let's brake Doug Gray's "null trick" workflow into steps:
1. Open tiff target in Photoshop
2. Assign any profile (not convert) to the tiff (I chose Canon PRO-2000/520 Satin Photo Paper 240g)
3. Open print dialog
4. Select Color Handling: Photoshop Manages Colors
5. Select Printer Profile: Canon PRO-2000/520 Satin Photo Paper 240g (same as the one assigned before to the tiff)
6. Leave all the other options as they are (see attached)
7. Open "Print Settings" choose Print Quality: High
8. Choose Color Mode: Color and Color Settings: Matching Off
9. Select Media Type: Plain Paper --> and here comes trouble!!!
Basically upon wrongly choosing the "Media Type" once as "Plain" and once as "Canon Satin Photo Paper 240g" BUT printing on the same paper resulted obvious differences in the printed targets, visually noticeable.
The same is applicable for printing using Canon PSP / PPL when choosing different papers to print the targets on. Mind that in Canon PSP /PPL I don't assign any profile and just choose "Color Mode: No Color Correction".
Why is this happening? Wasn't the bypassing of color management supposed to print the target in a paper & color agnostic way?
Why is changing the Paper Type altering the colors in the printed target? If this is 'normal' how am I supposed to choose the right paper, prior to being able to profile it?
I am baffled, please enlighten me! Thank you!
-
I am sure that one of our real experts will answer this but as I remember from experimenting a few years ago you try different media types to adjust the amount of ink laid down. You would expect quite different results with different media types selected. As I remember it you have to have a grey scale in your test image and you select the correct media type to use by examining the shadow separation. Once you have determined the correct media type to use you then print out the test chart (Adobe have a special app for doing this which prints without any color correction at all).
see https://helpx.adobe.com/photoshop/kb/no-color-management-option-missing.html
Dave
-
While I am still digging to find out about the weird differences two spectrophotometers are generating in their respective generated profiles (see my other thread) I am trying to reduce any discrepancies between the testing methods.
I am using two different target printing options:
A) Doug Gray's "null trick" option and B) the usual way (and recommended by Canon) by using Canon Print Studio Pro (PSP) or Canon Professional Print & Layout (PPL) plugins from Photoshop.
Yet I've discovered something that puzzles me.
Let's brake Doug Gray's "null trick" workflow into steps:
1. Open tiff target in Photoshop
2. Assign any profile (not convert) to the tiff (I chose Canon PRO-2000/520 Satin Photo Paper 240g)
3. Open print dialog
4. Select Color Handling: Photoshop Manages Colors
5. Select Printer Profile: Canon PRO-2000/520 Satin Photo Paper 240g (same as the one assigned before to the tiff)
6. Leave all the other options as they are (see attached)
7. Open "Print Settings" choose Print Quality: High
8. Choose Color Mode: Color and Color Settings: Matching Off
9. Select Media Type: Plain Paper --> and here comes trouble!!!
Basically upon wrongly choosing the "Media Type" once as "Plain" and once as "Canon Satin Photo Paper 240g" BUT printing on the same paper resulted obvious differences in the printed targets, visually noticeable.
The same is applicable for printing using Canon PSP / PPL when choosing different papers to print the targets on. Mind that in Canon PSP /PPL I don't assign any profile and just choose "Color Mode: No Color Correction".
Why is this happening? Wasn't the bypassing of color management supposed to print the target in a paper & color agnostic way?
Why is changing the Paper Type altering the colors in the printed target? If this is 'normal' how am I supposed to choose the right paper, prior to being able to profile it?
I am baffled, please enlighten me! Thank you!
You must choose the exact same conditions in the Print Settings as what you intend the profile to be used for. By choosing the media type as plain paper you not only changed the amount of ink laid but also the speed and the dpi of the print, all of which would affect the resulting profiles.
The by-pass is only for color management (i.e. whether or not and where to use the printer profile) not about how the printer utilizes the data and interprets them for the particular media type.
:Niranjan (not a real expert... :) )
-
Let's go back to basics. The purpose of printing profiling targets is to capture a characterization of how the printer reproduces the target file colours left to its own devices without colour management active. You should not be printing with a paper profile active. But you do need to specify the printer Media Type that will be the same Media Type you will use for the paper for which you are creating the profile. So you need to specify the Media Type and the print quality you will be using. That's it. For the Canon Pro-2000 printer use Print Studio Pro for a reliable print of your profiling target. This is all you need. Of course every different Media Type will produce a different target appearance. Of course print the target on the same paper you are profiling for and you will be using to print with. No need to complicate life with anything else.
-
Oh - important reminder - when using Print Studio Pro - make sure you have "No Color Controls" selected.
-
Let's go back to basics. The purpose of printing profiling targets is to capture a characterization of how the printer reproduces the target file colours left to its own devices without colour management active. You should not be printing with a paper profile active. But you do need to specify the printer Media Type that will be the same Media Type you will use for the paper for which you are creating the profile. So you need to specify the Media Type and the print quality you will be using. That's it. For the Canon Pro-2000 printer use Print Studio Pro for a reliable print of your profiling target. This is all you need. Of course every different Media Type will produce a different target appearance. Of course print the target on the same paper you are profiling for and you will be using to print with. No need to complicate life with anything else.
Oh - important reminder - when using Print Studio Pro - make sure you have "No Color Controls" selected.
You must choose the exact same conditions in the Print Settings as what you intend the profile to be used for. By choosing the media type as plain paper you not only changed the amount of ink laid but also the speed and the dpi of the print, all of which would affect the resulting profiles.
The by-pass is only for color management (i.e. whether or not and where to use the printer profile) not about how the printer utilizes the data and interprets them for the particular media type.
:Niranjan (not a real expert... :) )
All great advice.
Victor, it's now clear what rabbit hole you have gone down. These printer/paper type settings are absolutely critical to getting the best print results. Especially the differences between matte and glossy paper. For one, they use different black inks and a mismatch will always produce suboptimal to really bad results.
Glossy type papers include descriptions such as luster, pearl, semi-gloss and sometimes even semi-matte. Matte type papers have on major distinguishing characteristics. The reflect light specularly. If you move paper around in one strong light you will see reflections from the light that vary depending on how you hold the paper. Glossy will have mirror like relfections, pearl,luster, semi-gloss will have diffuse reflections that vary as you tilt the paper.
One other thing. Stop using the null-transform trick for targets. It's of value with other printers where there isn't a direct way to print w/o color management because Photoshop some time back removed the option to do this and alternatives, such as ACPU have limited ability to position an image and worse, tend to shrink it somewhat. I only use it because I have no alternative with my Epson and it makes it easy place multiple targets on a paper roll. It's not needed for you since Canon's PSP has a direct way to print in Photoshop w/o color management.
-
Let's go back to basics. The purpose of printing profiling targets is to capture a characterization of how the printer reproduces the target file colours left to its own devices without colour management active. You should not be printing with a paper profile active. But you do need to specify the printer Media Type that will be the same Media Type you will use for the paper for which you are creating the profile. So you need to specify the Media Type and the print quality you will be using. That's it. For the Canon Pro-2000 printer use Print Studio Pro for a reliable print of your profiling target. This is all you need. Of course every different Media Type will produce a different target appearance. Of course print the target on the same paper you are profiling for and you will be using to print with. No need to complicate life with anything else.
Dear Mark,
Much to my awe the Canon PSP is rubbish! Absolutely smth wrong with it. Basically no matter what it will not print the targets correctly, meaning even if "Color Mode: No Color Correction" is engaged the output print is still color managed.
Indeed PRO-2000 is my first serious printer and I absolutely love it BUT I did not have any prior experience on printer profiling thus I have miserably failed to recognize that Canon PSP is not working right until today. Ever since I have started printing targets, last November, I had a strange feeling that the colors on the printed targets looked somehow washed-out, muted color when compared against the target TIFF. I thought it was supposed to be so but I was wrong.
I thought it was my old Dell monitor that was playing tricks on me and I've upgraded to one professional NEC Spectra View.
And yet... the printed targets looked miserable and you could see (from the other thread) that the profiling produces miserable results. But today it occurred to me I should be trying to print the target the old fashioned way:
from PS tell that printer is managing colors --> Set in printer settings "Color Mode: Color and Color Settings: Matching Off" --> And bloody click "Print" with no strings attached, no assigned profile or anything.
And for the first time ever the colors printed really looked like not-managed, raw, forceful, vivid and punchy! And guess what: first profile, rushed in, done in i1Profiler yielded one beautiful profile, in the gamut size range of the original Canon similar profile for a "satin like" paper.
So while I am really disappointed about the lost time and resources so far I am happy I realized the issue is beyond me... how can it be fixed? Hard to tell.
In short:
NULL PRINTING is not working for Canon PRO-x000 series and so is ACPU
Canon PSP plugin + PS does not turn color management OFF... yes, I stand by my statement and I am willing to future explore / evaluate the facts.
Printing from i1Profiler with printer settings "Color Mode: Color and Color Settings: Matching Off" prints targets correctly ... sorry Doug for not having listened to you earlier and trying it!
I have attached the two shots of how 100% of my printed targets in PS+ Canon PSP look like vs real NO_COLOR_MANAGEMENT. I am pretty sure it's easy to guess which one is which.
-
Firstly, I am surprised PSP is not working properly, if indeed that's the case, because it was fine when I used it - see my Canon Pro-2000 review on this website, and I am also puzzled by the blue surround of the targets and the bluishness of what may be neutral patches. None of that looks right to me.
Printing the targets from i1Profiler should also be OK.
-
Firstly, I am surprised PSP is not working properly, if indeed that's the case, because it was fine when I used it - see my Canon Pro-2000 review on this website, and I am also puzzled by the blue surround of the targets and the bluishness of what may be neutral patches. None of that looks right to me.
Hard to tell much from the scans of the prints but it's odd that one is 3% smaller than the other. Where have I seen that before? ACPU! But he didn't use ACPU. So what's up? At a minimum the print settings differ both in dimension and color. One generally having more saturated colors than the other.
Printing the targets from i1Profiler should also be OK.
But would that be the case here? Doesn't ACPU use the same printing path and settings as direct from I1Profiler and you noted ACPU was not operating properly.
It's like everything he tries is producing weird results.
-
Hard to tell much from the scans of the prints but it's odd that one is 3% smaller than the other. Where have I seen that before? ACPU! But he didn't use ACPU. So what's up? At a minimum the print settings differ both in dimension and color. One generally having more saturated colors than the other.But would that be the case here? Doesn't ACPU use the same printing path and settings as direct from I1Profiler and you noted ACPU was not operating properly.
It's like everything he tries is producing weird results.
The only way he'll know for sure whether i1Profiler floats his boat is to try it. In principle you are correct that i1Profiler and ACPU should be doing the same kind of thing, but there are "gotchas" under hood with some of this stuff - in particular with the large format Canon printers - that are very hard to unearth without first just trying the software and seeing what it does. And that's right - for the Pro-2000 we could not get ACPU to work properly, but PSP with "No Color Coontrols" on did.
-
The only way he'll know for sure whether i1Profiler floats his boat is to try it. In principle you are correct that i1Profiler and ACPU should be doing the same kind of thing, but there are "gotchas" under hood with some of this stuff - in particular with the large format Canon printers - that are very hard to unearth without first just trying the software and seeing what it does.
Which is why when I make new profiles I almost always print a separate patch set, in gamut, with an independent constellation of specified Lab values in ProPhoto RGB. Then measure those and look at the dE00 distribution. If the averages get even close to 1, I did something wrong.
-
Firstly, I am surprised PSP is not working properly, if indeed that's the case, because it was fine when I used it - see my Canon Pro-2000 review on this website, and I am also puzzled by the blue surround of the targets and the bluishness of what may be neutral patches. None of that looks right to me.
Printing the targets from i1Profiler should also be OK.
Hi Mark,
I was just as surprised as you are! :( I am veryy keen to confirm this with other PRO-x000 series, maybe with Panagiotis or someone else. Ever since your PRO-2000 review on LuLa - that actually was the decisive review in my decision to go for the PRO-2000, btw :) - I was left with the clear "lesson" in my head that by the time I would be doing paper profiling I would use that PSP workflow. Until today... :(
The test is very simple, and you can trust me I'm no fool about ticking options so we can exclude any "being distracted":
- Print a i1Profiler target from i1Profiler with the Color Mode: Color and Color Settings: Matching Off] in the printer settings
- Print the same i1Profiler target from PS using Canon PSP or even Canon PPL (tested that too!) with Color Mode: No Color Correction option ticked.
I am not sure if I had mentioned but the usage of ACPU is really out_of_discussion since most of the targets' grays will be ... kinda brown.
I apologies for the quality of the submitted samples but the WB on my iPhone XS Max might have gone bananas though the pictures were taken outside, daylight, sunny.
I hope that the two new ones do some justice in explaining the situation: in short PS + Canon PSP = pale, faded while i1Profiler + Color Settings: Matching Off is vivid and closer to the initial un-tagged TIFF. Hence my other issue with reduced gamut is now 110% solved! :)
-
Well, I hope you have been enjoying the printer - until today. As for the profiling, we struggled until we found the solution, so we were very pleased when we did and that is why I recommended it back then. While we don't understand why the PSP approach has failed you, am I correct to infer from your latest comment that you now consider the target printing problem resolved?
-
Well, I hope you have been enjoying the printer - until today.
I would say that from now on even more! :)
As for the profiling, we struggled until we found the solution, so we were very pleased when we did and that is why I recommended it back then. While we don't understand why the PSP approach has failed you, am I correct to infer from your latest comment that you now consider the target printing problem resolved?
As I've said Mark, I am still waiting for some confirmation from other Canon owners. If my saying remains fact that we should be addressing this back to Canon. As for my profiling experience, at least while I'm using the targets printing from i1Profiler (as described) or from PS but without any help from PSP / PLL or ACPU ... all is fine by now and I am ok with the quality of my actual profiles. Definitely they could be improved but they are way better than the ones based on PS+PSP / Null Print workflows.
Thanks again for your support.
-
Maybe there is a bug in PSP.
I sent an untaged tiff chart file from PS (win-latest) to Canon PSP 2.2.3 and it opens it as sRGB. I can see that in the left side Image Information column in PSP.
-
Maybe there is a bug in PSP.
It could be! Canon launched PPL to replace PSP but from an workflow perspective is a mess. Not sure if you've checked it. I've printed targets from PPL too with the same unfortunate result.
I sent an untaged tiff chart file from PS (win-latest) to Canon PSP 2.2.3 and it opens it as sRGB. I can see that in the left side Image Information column in PSP.
Weird! Anyway... best way to check is visually with two targets printed via PS+PSP and PS+Printer Manages Colors+ "Color Mode: Color and Color Settings: Matching Off" in the printer settings. The differences are striking...
-
It could be! Canon launched PPL to replace PSP but from an workflow perspective is a mess. Not sure if you've checked it. I've printed targets from PPL too with the same unfortunate result.
Weird! Anyway... best way to check is visually with two targets printed via PS+PSP and PS+Printer Manages Colors+ "Color Mode: Color and Color Settings: Matching Off" in the printer settings. The differences are striking...
Canon PPL also opens it as sRGB. It seems that there is a bug on PS or Canon side. Check the screenshot.
-
Canon PPL also opens it as sRGB. It seems that there is a bug on PS or Canon side. Check the screenshot.
Wow! Let me check on mine and will get back!
-
It's probably a Canon problem. PPL is a stand alone application also. When I open an untaged chart file from the standalone app it converts it to: Canon lJ PPL Color 1
-
It's probably a Canon problem. PPL is a stand alone application also. When I open an untaged chart file from the standalone app it converts it to: Canon lJ PPL Color 1
LOL...
Let's make it even worse... my untagged TIFF is now in ProPhoto RGB...
-
It's probably a Canon problem. PPL is a stand alone application also. When I open an untaged chart file from the standalone app it converts it to: Canon lJ PPL Color 1
No offence Panagiotis but I suspect your default workspace in PS is not sRGB, is it? Cause the only thing I could think of is that "Color Space" it mentions in the PPL is the actual workspace PS is set to use. (mine is ProPhoto RGB) Though not sure if it impacts the print or not...
Could you check pls?
-
No offence Panagiotis but I suspect your default workspace in PS is not sRGB, is it? Cause the only thing I could think of is that "Color Space" it mentions in the PPL is the actual workspace PS is set to use. (mine is ProPhoto RGB) Though not sure if it impacts the print or not...
Could you check pls?
It is sRGB but the chart is not color managed.
-
I think I found something about Canon PSP.
The file is an untaged tiff chart from iProfiler.
Photoshop version used to open the file and send it (untaged) to the plug in is CS5.
Notice the difference in "Color Space" between PSP versions 2.1.1 (blank) and 2.2.1 (sRGB) in the following two screenshots.
EDIT: I investigated what version of the plugin was most probably Mark using in his review back in Jan 2017. It was probably version 2.1.x which was introduced to support PS CC2017. PSP suspect version 2.2.x was introduced after June 2017 to support the release of MacOS hi sierra. So maybe there is a bug in PSP later versions that make it inappropriate to print profiling targets?
-
Well, After re-reading Mark's review I'm sufficiently impressed with the print capability that I've ordered the 17" baby brother.
I'm going to install it minimally then add the plug-ins. At each step I'm going to print and scan charts using ACPU, I1Profiler direct, and my null-trick. I am quite curious as to the cause of these variations and expect to learn much more by next week. Especially what works and what doesn't and detail any differences discovered.
And at the end I'll have a nice replacement for my aging 9500 II.
-
Well, After re-reading Mark's review I'm sufficiently impressed with the print capability that I've ordered the 17" baby brother.
I'm going to install it minimally then add the plug-ins. At each step I'm going to print and scan charts using ACPU, I1Profiler direct, and my null-trick. I am quite curious as to the cause of these variations and expect to learn much more by next week. Especially what works and what doesn't and detail any differences discovered.
And at the end I'll have a nice replacement for my aging 9500 II.
It is an excellent printer. I am looking forward to read your findings and observations. I can also provide older win versions of the PSP plugin if you are interested.
-
Yes, I was going to suggest Doug that you include the older PSP as part of your testing suite, as it would be interesting to see whether it does things more properly than the newer replacement. You may have also noticed that while I used ACPU successfully on the Pro-1000 (17" model), I couldn't on the Pro-2000 (24" model). So as you proceed through the test suite, useful to bear in mind that even though the inkset and the printhead are identical for 1000/2000/4000, there are differences in software and firmware that can cause discordant behaviour.
-
Yes, I was going to suggest Doug that you include the older PSP as part of your testing suite, as it would be interesting to see whether it does things more properly than the newer replacement. You may have also noticed that while I used ACPU successfully on the Pro-1000 (17" model), I couldn't on the Pro-2000 (24" model). So as you proceed through the test suite, useful to bear in mind that even though the inkset and the printhead are identical for 1000/2000/4000, there are differences in software and firmware that can cause discordant behaviour.
Mark, after reviewing the manual I suspect this difference between ACPU and going through PSP may be due to the calibration option Canon has to tune the printer to a standard. Presumably something close to their profiles. This would have to involve a secondary mapping of profile device RGB values to the printer's hardware. Depending on where and how this additional mapping is saved and used, perhaps it is responsible for the difference between ACPU and PSP when run w/o CM? It would also likely vary printer to printer.
I intend to test this specific aspect when I get the printer.
As for the 9500 II, I'm not going to bother trying to get older PSP /window combos to run. I've never been able to get them to work in the last few years and it works fine without PSP. Let sleeping dogs .....
-
Great, I'll be most interested in your observations.
-
Mark, after reviewing the manual I suspect this difference between ACPU and going through PSP may be due to the calibration option Canon has to tune the printer to a standard. Presumably something close to their profiles. This would have to involve a secondary mapping of profile device RGB values to the printer's hardware. Depending on where and how this additional mapping is saved and used, perhaps it is responsible for the difference between ACPU and PSP when run w/o CM? It would also likely vary printer to printer.
I intend to test this specific aspect when I get the printer.
As for the 9500 II, I'm not going to bother trying to get older PSP /window combos to run. I've never been able to get them to work in the last few years and it works fine without PSP. Let sleeping dogs .....
If I remember correctly, PRO-1000 during initial setup asked to perform color calibration after print head alignment. PRO-4000 which I set it up two days ago didn't ask. After the head alignment was ready to print. This was a surprise to me and this is maybe a missing setup step for new Canon PRO-2000/4000/6000 owners.
I have a general idea how Canon own color calibration works. The best literature I have found is this PRO-4000 help page:
https://ugp01.c-ij.com/ij/webmanual/Manual/All/PRO-4000/EN/LBGB/tp000883.html
The are two types of color calibration: common and unique.
Only some media types can have a unique calibration. All others use the common calibration.
The least a user must do is to perform common calibration with a Canon recommended paper type and media setting. I use Photo Paper Pro Luster. This common calibration is then applied to all media types that have not a unique color calibration.
There is an application (Device Management Control) that keeps track of the calibration status of all the PRO printers on the network.
Also some paper manufacturers (Canson for example) am1x file settings for custom media (PK papers mostly) have already a calibration target generated for performing unique calibration. When these am1x files are imported to another PRO printer the user must(?) perform color calibration, for this custom media type, to set his printer to the condition the paper manufacturer printer was the time the custom media type and icc profile created. This procedure creates an unique calibration for this specific custom media type only, and doesn't affect all other paper media types. This calibration information is stored in the printer and can be used, or not, depending on settings in the printer driver or in the operation panel.
Hope this helps a little.
-
Hi Panagiotis,
Thanks for the info. It's useful background. I've also been looking at the workflows in their manuals and have some specific ideas on what's going on. Canon has a rather novel, and potentially very useful, calibration process which facilitates consistency when a customer has multiple printers. Especially when using Canon specific papers the printer characterizes.
This makes for an interesting issue creating and using profiles for other papers and documentation for doing so is lacking from what I've seen so far. Specifically the interaction between making a custom profile and subsequent re-calibrations. However, their basic idea is quite good so I expect to create a workflow that will be as consistent but allow for custom profiles that could be shared across printers. At least I hope to do this.