Luminous Landscape Forum
Raw & Post Processing, Printing => Printing: Printers, Papers and Inks => Topic started by: Nick Walker on May 25, 2016, 10:09:48 am
-
I cannot (neither could the Canon technician) find the printer 'off (no colour adjustments)' option in Mac OSX EI Capitan - he could only find this option using the Windows platform.
Printing from Photoshop CC, (Photoshop manages colours), or printing from Lightroom - Colorsync and Canon Color matching, both options are greyed out with ColorSync showing as selected in the print options (dot shown under ColorSync option).
I was about to profile my printer with Eye One spectro but I cannot print the test patches unless I know that there will definitely be no colour adjustments - ColorSync is not where I want to be.
Has anyone else encountered this problem with the Pro 1000 using a Mac please?
-
I cannot (neither could the Canon technician) find the printer 'off (no colour adjustments)' option in Mac OSX EI Capitan - he could only find this option using the Windows platform.
Printing from Photoshop CC, (Photoshop manages colours), or printing from Lightroom - Colorsync and Canon Color matching, both options are greyed out with ColorSync showing as selected in the print options (dot shown under ColorSync option).
I was about to profile my printer with Eye One spectro but I cannot print the test patches unless I know that there will definitely be no colour adjustments - ColorSync is not where I want to be.
Has anyone else encountered this problem with the Pro 1000 using a Mac please?
If I remember correctly from when I was doing all this, with this printer on a Mac, when you select application colour management, printer colour management is automatically turned off. Download the Adobe Color Print Utility (ACPU) from the Adobe website and print your profiling targets from that application, making sure to select the correct Media Type in the Print dialog from that application.
-
Since you're on a Mac, you can also print via the ColorSync Utility which has a 'print as color target' mode which does what you want.
-
If I remember correctly from when I was doing all this, with this printer on a Mac, when you select application colour management, printer colour management is automatically turned off. Download the Adobe Color Print Utility (ACPU) from the Adobe website and print your profiling targets from that application, making sure to select the correct Media Type in the Print dialog from that application.
Hi Mark,
I have been printing test charts from Adobe Color Print Utility (ACPU) for several years.
Today, I downloaded and was experimenting with the i1Profiler application - printing charts directly from its printer profiling module. I will return to using ACPU.
I suppose my question should have been, is it becoming the norm in the printer software/driver for the 'off (no color adjustments)' to no longer be available as an option using Mac OSX?
-
Since you're on a Mac, you can also print via the ColorSync Utility which has a 'print as color target' mode which does what you want.
Thanks,
I will take a look at this ColorSync option.
-
Hi Mark,
I have printing test charts from Adobe Color Print Utility (ACPU) for several years.
Today, I downloaded and was experimenting with the i1Profiler application - printing charts directly from its printer profiling module. I will return to using ACPU.
I suppose my question should have been, is it becoming the norm in the printer software/driver for the 'off (no color adjustments)' to no longer be available as an option using Mac OSX?
You can have "no COlor Management" selected in the Printer Driver, but Colorsync will still perform some colour management - my recollection of that issue, and I don't think it has changed since the problem first occurred. That is what ACPU was designed to work around. I too have printed targets from i1Profiler and the results were not good. X-Rite, being the kind of outfit it is, i1Profiler has no documentation worth the name so I have no idea whether that was pilot error or inadequacy of the application, and no time to waste on X-Rite's nonsense trying to figure out when ACPU provides a completely workable solution.
-
Since you're on a Mac, you can also print via the ColorSync Utility which has a 'print as color target' mode which does what you want.
Howard, I am on El Capitan 10.11.5 and I didn't see any such tool in ColorSync Utility. But maybe I missed it. Can you indicate how to find it?
-
I now use the ColorSync Utility for printing targets (10.10/11) but noticed that it has a habit of defaulting to expand the image to fit the paper.
I just remember check the scale option and set to 100% ... I have several huge target sheets produced during recent testing of the P7000 and PRO-1000 where I missed this little 'feature'.
-
Mark,
Open/load the target .tif file in ColorSync Utility and then cmd-P. Select the printer you want and in the pulldown menu for the various sub-functions, under ColorSync Utility, you'll find 'Color:' pulldown. The last item on that pulldown is 'Print as Color Target'.
-
Thanks Howard, every time I try to do anything in that utility I get the spinning beachball - with 24GB RAM and 24 virtual cores. Something silly going on. But I'll try it.
-
Weird. I know sometimes ColorSync does take it's time coming up, esp. if it's querying the Devices tab but I can't think of much other reason
-
Hello all. Re Canon & turning off CM in Mac 10.11
I have the same problem with a new Canon pro-2000. I Noted Howards suggestion using the color sync utility, thought it was a great idea, however disappointingly the "print as color target" doesn't exist in the available drop downs.
I tried ACPU, however it's not playing nicely in the Mac 10.11 and only prints 1/2 the targets. At this stage, the Canon techs have no answers.
The ColorMunki and an i1 have all failed. To addd confusion the i1 appears to have corrupted the print drivers resulting in a print taking some 3 hours to run a 40 x 60cm b&w.
After uninstalling and re-installing the drivers, the printer is again printing correctly, however profiling seems near to impossible at the moment.
Any suggestions would be gratefully accepted.
-
Hello all. Re Canon & turning off CM in Mac 10.11
I have the same problem with a new Canon pro-2000. I Noted Howards suggestion using the color sync utility, thought it was a great idea, however disappointingly the "print as color target" doesn't exist in the available drop downs.
I tried ACPU, however it's not playing nicely in the Mac 10.11 and only prints 1/2 the targets. At this stage, the Canon techs have no answers.
The ColorMunki and an i1 have all failed. To addd confusion the i1 appears to have corrupted the print drivers resulting in a print taking some 3 hours to run a 40 x 60cm b&w.
After uninstalling and re-installing the drivers, the printer is again printing correctly, however profiling seems near to impossible at the moment.
Any suggestions would be gratefully accepted.
I've used ACPU to print targets in the Epson P800, Canon Pro-1000 and Canon Pro-2000. Many of them in fact. I have experienced no problem of half-printed targets. Using ACPU and El Capitan, as usual the ACPU requires that you open and print one page of your target at a time. Each page prints fully. How large is your profiling target, and is it provided in a format to allow printing of one readable sheet at a time?
You say the ColorMunki and i1 have all failed? Failed at doing what? What exactly is the problem? Is the i1 you talk about the i1 spectrophotometer, or i1 Profiler software?
I don't think there is any way an i1 can corrupt a printer driver. The two have nothing to do with each other. What makes you think this?
When you say profiling appears nearly impossible at the moment, in what ways is this the case? Only because your targets aren't printing, or are there other issues here?
Sorry to be asking all these questions, but without crystal clarity regarding the issues it's hard to understand what's going on and make useful suggestions if I could.
-
Hello Mark, thank you for your interest.
I've used ACPU to print targets in the Epson P800, Canon Pro-1000 and Canon Pro-2000. Many of them in fact. I have experienced no problem of half-printed targets. Using ACPU and El Capitan, as usual the ACPU requires that you open and print one page of your target at a time. Each page prints fully. How large is your profiling target, and is it provided in a format to allow printing of one readable sheet at a time?
I Purchased an i1Photo Pro 2 Color Management Kit as an upgrade to the Munki along with the new printer. After installing the i1, the printer took over 3 hours to produce the print I mentioned. My concern was a printer malfunction as it's brand new. Canon tech advised to remove the i1, correct the permissions and re-install print drivers. Doing so fixed the printer issue, but removed the i1. (I can see no correlation between print driver and i1 either) I tried to print an 800 patch test target from the i1 through ACPU, however only half the target printed, then the printer went into slow motion taking some 30 minutes to print the other half a page. Page size was A4.
You say the ColorMunki and i1 have all failed? Failed at doing what? What exactly is the problem?
The profiles from the ColorMunki are way off, possibly due to my ignorance, or the fact that I can't find a way to bypass colour management. Whatever the case, the profiles created are useless.
Is the i1 you talk about the i1 spectrophotometer, or i1 Profiler software?
Profiler software, however I can't get it to run without the printer failing. The i1 has since been returned, and am left with 2 ColorMunki's. A loan unit to determine if the one I own is damaged, but both produce the same results. The targets print ok, and the printer is working again, but as stated, the profiles are useless.
I don't think there is any way an i1 can corrupt a printer driver. The two have nothing to do with each other. What makes you think this?
Comments from Canon, also several hours of on site assistance from the AU Xrite distributor, who is also confused by the situation.
When you say profiling appears nearly impossible at the moment, in what ways is this the case? Only because your targets aren't printing, or are there other issues here?
The consensus of opinion (all be it possibly incorrect) from various professionals is that the issue is no one can find a way to truly bypass color management. Save target as tiff, can't get past print driver without CM. ACPU as stated failed. ColourMunki does an override at the driver and greys out options, however the profiles are as stated useless so something seems to be going on behind the scenes.
Sorry to be asking all these questions, but without crystal clarity regarding the issues it's hard to understand what's going on and make useful suggestions if I could.
Thanks for your questions, and my apologies for not being clear enough initially. Cheers.
-
Thanks for providing all those clarifications. It's unfortunate that you returned the i1Profiler kit because in principle it is far more capable than ColorMunki and I still can't conjure any reason why it should mess with the printer driver - or if somehow it does, nothing that tech support between XRite and Canon shouldn't be able to rectify, even if they need to reference their HQ support groups in Japan and the USA to resolve it. If you still had the i1Profiler kit I was going to suggest printing the target through it rather than the ACPU to see whether that solves the problem. Like you, I suspect that bypassing colour management may be part of the story, but there was a time - not too long ago - when I thought all that was licked. Anyhow, your post comes at a fortuitous time in terms of some research I'm doing that involves further profiling work with the new Canon printer line, so leave this one with me for a few days and I shall get back to you.
-
Did you try Canon Print Studio Pro plug in?
(https://dl.dropboxusercontent.com/u/19059944/Zrzut%20ekranu%202016-08-18%2014.53.00.png)
-
Marcin, that option is not provided in the Mac version - at least the one I have. Canon has advised that when printing from an image editing application on OSX, Printer Colour Management is automatically disabled. Is your screen grab from a Windows or OSX installation?
-
Marcin, that option is not provided in the Mac version - at least the one I have. Canon has advised that when printing from an image editing application on OSX, Printer Colour Management is automatically disabled. Is your screen grab from a Windows or OSX installation?
No, it's the OSX - Canon Print Studio Pro 2.1.0
-
OK, version screw-up. It wasn't an option using version 1.0 in Lightroom. It is an option using version 2.1 in Photoshop. Regardless, all that does is turn off printer colour management in the printer driver. AFAIK, it does not deal with OS colour management in a way that allows accurate printing of profiling targets. Does your experience using it differ?
-
OK, version screw-up. It wasn't an option using version 1.0 in Lightroom. It is an option using version 2.1 in Photoshop. Regardless, all that does is turn off printer colour management in the printer driver. AFAIK, it does not deal with OS colour management in a way that allows accurate printing of profiling targets. Does your experience using it differ?
AFAIK anything that turns color management in printer driver and sends unchanged RGB numbers of the target to the printer should work, unless application developers screw something or Apple screw something in OSX to the point that's beyond my imagination (which is quite common lately ;) )
-
....anything that turns ("off" I think you meant) color management in printer driver and sends unchanged RGB numbers of the target to the printer ................
Yes, that indeed is the question. With each new generation of operating systems, applications, drivers and firmware, some analytics are needed to know for sure whether that is always happening, and if it isn't, then how to determine which part of the chain is broken. But more directly on your point about PSP - it looks fine for the part of the chain dealing with the printer, but I've been advised that it does not deal with how the OS deals with the file numbers. PSP is a GUI intended mainly as a simplified window into the printer driver and to provide some print arrangement capabilities.
-
it looks fine for the part of the chain dealing with the printer, but I've been advised that it does not deal with how the OS deals with the file numbers. PSP is a GUI intended mainly as a simplified window into the printer driver and to provide some print arrangement capabilities.
That's the reason I miss the former PS plug-in a little bit - it was 100% bullet proof regarding profiling target printing
-
Did you try Canon Print Studio Pro plug in?
(https://dl.dropboxusercontent.com/u/19059944/Zrzut%20ekranu%202016-08-18%2014.53.00.png)
I did a test today trying to print a small test target without color management on the Pro-1000. Just 63 patches with a good spread around the gamut of Canon Glossy II paper.
Four cases were tested:
1) Adobe Color Print Utility (ACPU)
2) Canon Print Studio Pro (PSP), "No color correction", using default printer driver preset (as requested by PSP)
3) ColorSync ("Print as color target")
4) Canon Print Studio Pro (PSP), "No color correction", using same driver settings as case 1 and 3
Tests done using ArgyllCMS and an i1Pro.
Prints 1 and 3 look and measure similar. Peak delta E of 1.6 (only 0.66 CIEDE2000 delta E). Instrument variation on control read peaked af delta E of 0.8 for comparison.
Prints 2 and 4 look and measure the similar. Peak delta E of 2.4.
But... Comparing print 1 vs 2, that is ACPU vs PSP, things look very different (certainly doesn't require an instrument to observe the huge differences in many color patches). Peak delta E of over 50!
---
colverify target1-case-1-ACPU.ti3 target1-case-2-PSP.ti3
Verify results:
Total errors: peak = 57.614156, avg = 12.884879
Worst 10% errors: peak = 57.614156, avg = 42.173240
Best 90% errors: peak = 32.288184, avg = 9.801894
avg err X 0.058385, Y 0.052332, Z 0.016921
avg err L* 5.507523, a* 7.297215, b* 8.105047
---
When reading the chart for PSP, the tool already complained patches read very much different than expected ("There is at least one patch with an very unexpected response! (DeltaE 56.486613)"). Had preconditioned the target with Canon's OEM profile for the paper, to give the target an idea of what to expect.
At the moment I will trust ACPU and ColorSync. That has let to profiles looking fine. Haven't tried building a profile based on the PSP results, but wanted to test the difference today and was quite surprised. Even recall Mark Segal having been advised to print target in PSP by Canon, and given the differences I'm concerned. Could be PSP is right and the others not, but no indications to me that's currently the case, but could be wrong (one of the great reasons to share here).
Anyone else aware of this huge discrepency? I know macOS Sierra early on has had some issues here with Canon, but thought that was no more.
Using macOS Sierra 10.12.3. Printer driver 16.10.1.0. Canon Print Studio Pro 2.1.0.
-
I did a test today trying to print a small test target without color management on the Pro-1000. Just 63 patches with a good spread around the gamut of Canon Glossy II paper.
Four cases were tested:
1) Adobe Color Print Utility (ACPU)
2) Canon Print Studio Pro (PSP), "No color correction", using default printer driver preset (as requested by PSP)
3) ColorSync ("Print as color target")
4) Canon Print Studio Pro (PSP), "No color correction", using same driver settings as case 1 and 3
Tests done using ArgyllCMS and an i1Pro.
Prints 1 and 3 look and measure similar. Peak delta E of 1.6 (only 0.66 CIEDE2000 delta E). Instrument variation on control read peaked af delta E of 0.8 for comparison.
Prints 2 and 4 look and measure the similar. Peak delta E of 2.4.
But... Comparing print 1 vs 2, that is ACPU vs PSP, things look very different (certainly doesn't require an instrument to observe the huge differences in many color patches). Peak delta E of over 50!
---
colverify target1-case-1-ACPU.ti3 target1-case-2-PSP.ti3
Verify results:
Total errors: peak = 57.614156, avg = 12.884879
Worst 10% errors: peak = 57.614156, avg = 42.173240
Best 90% errors: peak = 32.288184, avg = 9.801894
avg err X 0.058385, Y 0.052332, Z 0.016921
avg err L* 5.507523, a* 7.297215, b* 8.105047
---
When reading the chart for PSP, the tool already complained patches read very much different than expected ("There is at least one patch with an very unexpected response! (DeltaE 56.486613)"). Had preconditioned the target with Canon's OEM profile for the paper, to give the target an idea of what to expect.
At the moment I will trust ACPU and ColorSync. That has let to profiles looking fine. Haven't tried building a profile based on the PSP results, but wanted to test the difference today and was quite surprised. Even recall Mark Segal having been advised to print target in PSP by Canon, and given the differences I'm concerned. Could be PSP is right and the others not, but no indications to me that's currently the case, but could be wrong (one of the great reasons to share here).
Anyone else aware of this huge discrepency? I know macOS Sierra early on has had some issues here with Canon, but thought that was no more.
Using macOS Sierra 10.12.3. Printer driver 16.10.1.0. Canon Print Studio Pro 2.1.0.
Something isn't right here and I can't pinpoint what it is based on this information. Anyhow, some observations: The spread in the error term between ACPU and PSP is much too large relative to my experience. Furthermore, regardless of what the Canon instructions say about using default driver settings for PSP, it is essential for correct colour management purposes to have at least the same media type selected for making the profile and printing the test chart that measures the profile 's accuracy. A mismatch of Media Type between the profile creation settings (i.e. profiling target print) and the evaluation target print is a good recipe for large, undesirable dE outcomes. As well, I do not know that Mac OSX Sierra is producing colour management issues for Canon print workflows - haven't seen any recent conclusive evidence pointing to that.
If Canon Glossy II paper has OBAs (most likely but I don't know this paper), then an incorrect or inconsistent specification of the "M" condition in either or both of profiling and profile verification could cause large-ish dE.
Don't only look at peak dE - also look at the average dE and dispersion around the average. It's a better indicator of overall profile performance.
Based on the data report at the bottom of your post these results from ACPU are not good. One thing to verify is whether the gamut of the profile is in places smaller than the gamut of the test patches you are using. If so, large dE outcomes could arise from this condition as well. This can be verified with ColorThink Pro. However, using a gloss media I rather doubt this would be the case.
-
Furthermore, regardless of what the Canon instructions say about using default driver settings for PSP, it is essential for correct colour management purposes to have at least the same media type selected for making the profile and printing the test chart that measures the profile 's accuracy. A mismatch of Media Type between the profile creation settings (i.e. profiling target print) and the evaluation target print is a good recipe for large, undesirable dE outcomes.
That's why case 4 uses the exact same settings (including media type) as used with case 1 (ACPU) and case 3 (ColorSync).
If Canon Glossy II paper has OBAs (most likely but I don't know this paper), then an incorrect or inconsistent specification of the "M" condition in either or both of profiling and profile verification could cause large-ish dE.
OBAs would never explain this amount of change observed (delta 50+). I'm making all measurements here with consistent measurement settings. I'm not evaluating profiles, only targets printed without color management.
Don't only look at peak dE - also look at the average dE and dispersion around the average. It's a better indicator of overall profile performance.
I'm not evaluating profiles here. I'm comparing target prints made from different software, all with the aim of printing without color management (as there have been issues with that between software matches before). What I'm illustrating, is that PSP target prints do not at all match target prints made with ACPU or ColorSync with the versions and settings mentioned. And I would expect they did. And I'm sticking with ACPU/ColorSync which agrees, and have previously made good base for custom profiles, evaluated with all the bells and whistles (analysis in GamutVision, visually etc).
Based on the data report at the bottom of your post these results from ACPU are not good. One thing to verify is whether the gamut of the profile is in places smaller than the gamut of the test patches you are using. If so, large dE outcomes could arise from this condition as well. This can be verified with ColorThink Pro. However, using a gloss media I rather doubt this would be the case.
No. Data report compares ACPU with PSP to illustrate the extreme difference between those two target prints (remembering ACPU is similar to ColorSync). No profiles generated or evaluated here, and not going to in this test. The target prints should all be similar, and PSP is sticking out where I assume it should not. That was what the test is set to illustrate, to make others aware in case they assume PSP is printing "No color correction" as ACPU and ColorSync ("Print as color target") - it is not, at all. At least not with the shown versions and settings.
-
No. Data report compares ACPU with PSP to illustrate the extreme difference between those two target prints (remembering ACPU is similar to ColorSync). No profiles generated or evaluated here, and not going to in this test. The target prints should all be similar, and PSP is sticking out where I assume it should not. That was what the test is set to illustrate, to make others aware in case they assume PSP is printing "No color correction" as ACPU and ColorSync ("Print as color target") - they are not, at all. At least not with the shown versions and settings.
]
I agree, OBA-induced differences would not cause dE differences of 50. That's why I said "large-ish", and didn't say huge.
I appear to have misunderstood the primary purpose of your exercise. Sorry about that. It's probably because my mind is focused on the practical question of optimizing profile performance so I get consistently correct prints out of the printer.
I am not surprised you get different results from the different applications. I observed quite some time ago as well that they do not produce nearly identical outcomes even though they all advertise as "no colour management" algorithms. Why they differ to the extent they do goes beyond my competence to analyse - one would need in-depth, detailed understanding of the algorithms, but they do, and you have once again confirmed that phenomenon. Beyond that I'm not sure what this buys me. I shall continue to use profiles that demonstrate the greatest tested average accuracy and least dispersion around the mean. But if your approach satisfies you on what you should be using, of course it's fine. Whatever works best for each of us is what we will be most comfortable with and should go on using. Cheers.
-
I appear to have misunderstood the primary purpose of your exercise. Sorry about that. It's probably because my mind is focused on the practical question of optimizing profile performance so I get consistently correct prints out of the printer.
No problem. And me too about end result :). And to me that would include verifying that the base of my custom profiles, the printed targets, can be relied upon. Currently they vary where they shouldn't. I can make profiles I'm happy with - that's not the issue. It's inconsistency between software that promises "no color management" applied (to a degree in a practical sense, not for measurement's sake alone - they are wildly different). When they promise I would like to know I can trust it, e.g. if assigning printer profiles in Photoshop (for various reasons, one could be use of BPC with PSP which doesn't support it) and printing elsewhere (like in PSP if I preferred to use Contrast Reproduction or other features only available somewhere special. Or I just want to know my target is trustworthy before building profiles upon it. Basing them on wildly different target prints seem inconsistent to me).
Again, trying to point out the difference in case others think they obtain the roughly same target print using the different techniques shown. They don't, not even close.
-
Even recall Mark Segal having been advised to print target in PSP by Canon, and given the differences I'm concerned.
Did I remember wrongly here? I remember a reviewer getting into issues with target printing resulting in surprising profile problems, and was advised by Canon to print target through PSP. Perhaps it was Keith Cooper instead, or someone third of many good reviewers :). Sorry can't remember, but appreciate all your great efforts and write-ups.
Anyhow, I wanted to test, and share, if target print inconsistencies had been solved, and apparently it's still not consistent (or close to). And since profiling is about consistency, it's an issue in my opinion, and worth sharing (including quantification to show how very different they really are, not just nitty gritty insignificances).
-
You didn't remember wrongly. It was me.
-
I did a test today trying to print a small test target without color management on the Pro-1000. Just 63 patches with a good spread around the gamut of Canon Glossy II paper.
...
Purpose was to see if different techniques of printing target without color management (all seen recommended here or there, as you would assume they were identical or pretty close) are consistent in behaviour, and if different, by how much (insignificant or major discrepancy). That has been done already above, showing major discrepancies using the software and setting combination as described.
Such a target is used as base to create a profile (how does device input RGB actually render in perceptual/tristimulis vision we agree on, e.g. CIE Lab, and reverse), with multiple features, one being 1) consistency between soft proof on monitor and what you get out of the printer, another 2) consistency with other printers, so you can trust that colors within common gamut abilities actually can render the same. So if multiple people individually and correctly profile their printer and print the same image on same paper with colorimetric rendering intent, they should look the same - within common gamut. If printers were even the same, they could share the very same profile, given printers were calibrated (as for instance the Pro-1000 can be).
So I currently wonder what will happen if you profile with two significantly different looking targets, like is currently possible, unknowingly, because PSP differs vastly from ACPU/ColorSync. Would 1) be achieved, but 2) not? That's a guess currently, but not thought complete through. Comments please. But I find it hard to believe the different target prints won't result in different profiles, making at least 2) impossible to achieve. And I would aim for having both 1) and 2) right, so we each agree on colors, getting the same consistent print output, and not just are satisfied that our own soft proof looks like our own print.
So if the target print does matter and causes inconsistency if not done right, how do we determine "right"?
Well, I thought that if Canon has gotten their own profiles roughly right, then I can predict using that OEM profile if my target prints look "right" (as their profile includes the device RGB input mapping to Lab). Having made my small test target preconditioned with Canon's OEM profile, I have, using ArgyllCMS, included the expected raw print output for each patch RGB device input - as Canon have measured it when creating their own profile for a similar combination of printer/paper.
So, does the PSP print match the expected values from Canon's own profile measurements?
colverify target1-case-2-PSP.ti2 target1-case-2-PSP.cie.ti3
Verify results:
Total errors: peak = 58.401719, avg = 14.152884
Worst 10% errors: peak = 58.401719, avg = 43.449070
Best 90% errors: peak = 32.835848, avg = 11.069075
avg err X 0.063412, Y 0.058000, Z 0.021320
avg err L* 6.300949, a* 7.518214, b* 8.974196
No, not even close. With peak dE of 50+ and worst 10% error averaged of 40+, I will surely conclude a no. The print does not in any way represent the expected raw device response.
Does the ACPU and ColorSync target prints (which are similar to each other as shown earlier) match the expected values from Canon's own profile measurements?
colverify target1-case-1-ACPU.ti2 target1-case-1-ACPU.cie.ti3
Verify results:
Total errors: peak = 6.774103, avg = 2.858761
Worst 10% errors: peak = 6.774103, avg = 5.876139
Best 90% errors: peak = 4.843732, avg = 2.541143
avg err X 0.010225, Y 0.009825, Z 0.008567
avg err L* 1.287127, a* 1.323744, b* 1.461922
Yes, to a degree that could reasonably be explained by printer, paper, environmental and measuring variations, and an indicator for why you might want to profile on your own as differences can show up. Peak dE of 7, worst 10% average of 6 and best 90% avg of <3 (under 1.6 in CIEDE2000) doesn't sound unrealistic to what I would call an expected match - certainly compared with the PSP results.
So I conclude that ACPU/ColorSync are the ones to trust for target prints, both agreeing with each other and Canon's own measured device responses contained in their OEM profile.
Please let me know if I got this line of reasoning all wrong, or someone agrees.
If you agree, I will hope PSP/macOS/driver gets the printing with "No color correction" correct, as it is useful in several scenarios, if done right.
-
Purpose was to see if different techniques of printing target without color management (all seen recommended here or there, as you would assume they were identical or pretty close) are consistent in behaviour, and if different, by how much (insignificant or major discrepancy). That has been done already above, showing major discrepancies using the software and setting combination as described.
Such a target is used as base to create a profile (how does device input RGB actually render in perceptual/tristimulis vision we agree on, e.g. CIE Lab, and reverse), with multiple features, one being 1) consistency between soft proof on monitor and what you get out of the printer, another 2) consistency with other printers, so you can trust that colors within common gamut abilities actually can render the same. So if multiple people individually and correctly profile their printer and print the same image on same paper with colorimetric rendering intent, they should look the same - within common gamut. If printers were even the same, they could share the very same profile, given printers were calibrated (as for instance the Pro-1000 can be).
So I currently wonder what will happen if you profile with two significantly different looking targets, like is currently possible, unknowingly, because PSP differs vastly from ACPU/ColorSync. Would 1) be achieved, but 2) not? That's a guess currently, but not thought complete through. Comments please. But I find it hard to believe the different target prints won't result in different profiles, making at least 2) impossible to achieve. And I would aim for having both 1) and 2) right, so we each agree on colors, getting the same consistent print output, and not just are satisfied that our own soft proof looks like our own print.
So if the target print does matter and causes inconsistency if not done right, how do we determine "right"?
Well, I thought that if Canon has gotten their own profiles roughly right, then I can predict using that OEM profile if my target prints look "right" (as their profile includes the device RGB input mapping to Lab). Having made my small test target preconditioned with Canon's OEM profile, I have, using ArgyllCMS, included the expected raw print output for each patch RGB device input - as Canon have measured it when creating their own profile for a similar combination of printer/paper.
So, does the PSP print match the expected values from Canon's own profile measurements?
colverify target1-case-2-PSP.ti2 target1-case-2-PSP.cie.ti3
Verify results:
Total errors: peak = 58.401719, avg = 14.152884
Worst 10% errors: peak = 58.401719, avg = 43.449070
Best 90% errors: peak = 32.835848, avg = 11.069075
avg err X 0.063412, Y 0.058000, Z 0.021320
avg err L* 6.300949, a* 7.518214, b* 8.974196
No, not even close. With peak dE of 50+ and worst 10% error averaged of 40+, I will surely conclude a no. The print does not in any way represent the expected raw device response.
Does the ACPU and ColorSync target prints (which are similar to each other as shown earlier) match the expected values from Canon's own profile measurements?
colverify target1-case-1-ACPU.ti2 target1-case-1-ACPU.cie.ti3
Verify results:
Total errors: peak = 6.774103, avg = 2.858761
Worst 10% errors: peak = 6.774103, avg = 5.876139
Best 90% errors: peak = 4.843732, avg = 2.541143
avg err X 0.010225, Y 0.009825, Z 0.008567
avg err L* 1.287127, a* 1.323744, b* 1.461922
Yes, to a degree that could reasonably be explained by printer, paper, environmental and measuring variations, and an indicator for why you might want to profile on your own as differences can show up. Peak dE of 7, worst 10% average of 6 and best 90% avg of <3 (under 1.6 in CIEDE2000) doesn't sound unrealistic to what I would call an expected match - certainly compared with the PSP results.
So I conclude that ACPU/ColorSync are the ones to trust for target prints, both agreeing with each other and Canon's own measured device responses contained in their OEM profile.
Please let me know if I got this line of reasoning all wrong, or someone agrees.
If you agree, I will hope PSP/macOS/driver gets the printing with "No color correction" correct, as it is useful in several scenarios, if done right.
This degree of error is about what I've seen doing similar tests between OEM profiles and custom ones I've made. At least for glossy, high DMax, targets.
Unfortunately, I can't explain the differences you are seeing with the various target printing techniques. This sort of variation indicates some serious differences in data flow with presumably no color management.
I've never seen a difference printing targets but I've only used 3 techniques and they were all done on Windows with Epson and Canon printers. I do not use PSP They are:
1. Printing directly from I1Profiler
2. Printing the target file image using ACPU
3. Printing using Photoshop manages color using an arbitrary profile that had also previously been assigned to the target.
They all produce targets that measure well under .5 dE average.
-
You didn't remember wrongly. It was me.
I did find your article now at https://luminous-landscape.com/canon-pro-2000-review-and-related-items-of-interest/. It's very insightful with good discussions of both technical and practical details of profiling and paper observations. Adding many uncovered aspects to the Pro-1000 review. Thank you.
I see how you fought the ACPU/PSP target printing monster during profiling with you i1Profiler setup, and found a practical match in the end with PSP with impressive results after a long research.
I also notice how your OEM profile dE results in general compares very much to mine from yesterday, both on peaks and averages. Different methods and tools used, but agreeing on results - that's always reassuring to see, and what's it's about, consistency.
I look forward to my own profiling result comparisons with ACPU and other profiling tools (like Argyll, where I prefer the better insight to what is going on, especially when something is goofy like we see here with the target print mismatches). I will still prefer ACPU/ColorSync as target printing tool, as it seems objectively to be the correct one with current state of software matches. It didn't work out with your i1Profiler suite, but I hope I'm not affected by that using another tool. I will know soon enough.
Is it possible that PSP and i1Profiler on your tested setup somehow works out best together because they share the same behaviour, which actually isn't intended / correct, but if the same, happens to work out in the end? If so, we might of course watch out if PSP suddenly changes behaviour (fixed) to match the rest (ACPU/ColorSync), and/or i1Profiler changes something. It's always suspicious when existing solid workflows suddenly change behaviour, as you noted so clearly. I would love to see PSP print correctly, as-is, without color management as it can be useful if trusted and consistent with other tools printing without color management. If Canon/macOS/driver doesn't change the behaviour it's inconsistent and problematic to work around I think, leading to potential surprises, that shouldn't be needed. But also having worked with software development I recognise how many facets can get mixed up between application, OS and drivers, that I'm not surprised it screws up from time to time.
I'm very intrigued by your interesting personal and panel observations of the new Canon FA-SM (FineArt Smooth or FAS) and the perceived improved blackness, and the discussions about Lab measures vs visual impression. I will try out FAS if available here.
The two other new papers also look interesting, but were rebranded existing stocks as I recall, so I'd like to know which they compare to. Do you know, or have a feeling for what it most closely resembles? I would prefer buying it from a source where I can also get cut sheets. If no one currently knows the stock behind it, someone could share a spectral reading file of paper white (not UV-cut) with proper details, preferably with and without black backing and front and back side readings, it could be compared to known databases of existing papers, whereby you can often find out which ones are rebrands of same stock (not taking surface structure much into consideration though, but reflectance). That's a valuable shortcut to finding similar paper substitutions.
-
.................
Is it possible that PSP and i1Profiler on your tested setup somehow works out best together because they share the same behaviour, which actually isn't intended / correct, but if the same, happens to work out in the end? If so, we might of course watch out if PSP suddenly changes behaviour (fixed) to match the rest (ACPU/ColorSync), and/or i1Profiler changes something. It's always suspicious when existing solid workflows suddenly change behaviour, as you noted so clearly. I would love to see PSP print correctly, as-is, without color management as it can be useful if trusted and consistent with other tools printing without color management.
Thanks - glad you found the article useful.
Turning to your question, I'm in the OSX context now - don't know how this works with Windows. A most important function we expect from a profiling target-printing application is that it either completely turns off or completely neutralizes colour management in the printer driver and in the operating system, so that the resulting profile will do what it is intended to do: i.e., describe the native behaviour of the printer in respect to how it lays down colours according to the pixel colour values in the image file. Here I am surmising based on quite a bit of reading, as I am not a programmer or application developer. Turning it off in the printer driver is probably straightforward; the operating system however may be another matter and I suspect here the goal is neutralizing whatever colour management is "baked-in" and inaccessible to the developers. I am surmising that each of these target printing algorithms is doing this somewhat differently, which may explain the variability in performance that we have both observed. I'm offering this suggestion hoping that someone who really knows the innards of these applications and processes could jump in and either confirm or correct.
As you will have read from my empirical work, I did find that on OSX El Capitan or Mavericks (haven't tested Sierra) and working with i1Profiler 1.6.7/i1Pro2, for the new line of Canon "Pro xxxx" printers, Canon's advice to use PSP with "No Color Correction" selected for printing profiling targets is the preferred way to assure the objective of simulating unmanaged target printing. I still find that ACPU is very reliable for the Epson SCP800 and SP4900 printers. There are several sets of actors involved with this: Canon, XRite and Adobe (and perhaps Apple); hence it's far from clear whether or when managerial and technical resources would be devoted to assuring cross-application coherence between all these printing vehicles; I agree with you it would be desirable, but how essential is less clear given that we know what works with what.