Severe gamut reduction in profiles Colormunki vs i1Pro2 in ArgyllCMS

[vikcious]

The CM profile I used for that comparison was quite close to the I1Pro2 profile was:"test_target_cm.icm", which was made with 200 patches and has the embedded description: "ArtMaker Cristal Luster 270g CM". The I1Pro profile it was close to has the file name:"argyll_650_on_A4_null_print_M2.icm" and description:"ArtMaker CristalLuster NULL ARG_M2_2_M2"


Don't change too many things at once. It's now clear from the other thread you started that the basic problem is variation in printer settings using PSP.

Your I1Pro 2 is probably behaving perfectly well.

Dear Doug,

I have just posted in the other thread about my latest discovery related to the failure of PS + Canon PSP in printing with COLORMANAGEMNT_OFF.
I am now working on producing a couple of test profiles with the right gamut and almost all flawless.
There are still a couple of thoughts I need to share with you but I have pretty much identified the culprit.

I will get back to you as soon as I'm done with the new profiles to discuss them together. I am sorry I haven't listened to your suggestion of trying to print form i1Profiler with the color management OFF in the printer settings... My bad!

[Doug Gray]

Dear Doug,

I have just posted in the other thread about my latest discovery related to the failure of PS + Canon PSP in printing with COLORMANAGEMNT_OFF.
I am now working on producing a couple of test profiles with the right gamut and almost all flawless.
There are still a couple of thoughts I need to share with you but I have pretty much identified the culprit.

I will get back to you as soon as I'm done with the new profiles to discuss them together. I am sorry I haven't listened to your suggestion of trying to print form i1Profiler with the color management OFF in the printer settings... My bad!

BTW, one of the ONLY ways I know of to cross check a profile AND printer settings is to print an image of a ColorChecker (from bablecolor.com amonst others) using Absolute Colorimetric then spot read the 24 colors with a the same spectro used to create the profile.

My Epson 9800 has tons of settings (color management, head height (paper thickness), vacuum levels, quality, paper type, roll sie, cutter, etc) and I've found some improvement tailoring these for off brand papers. Then I save the setup using a name specific to the configuration and make profiles with targets from the same one. Don't know about the Canon nor how it interacts with PSP. I can't get PSP running on my Windows setup for the 9500 II probably because it's old and Canon hasn't updated it for current Photoshops when using the long-in-the-tooth 9500.

[vikcious]

BTW, one of the ONLY ways I know of to cross check a profile AND printer settings is to print an image of a ColorChecker (from bablecolor.com amonst others) using Absolute Colorimetric then spot read the 24 colors with a the same spectro used to create the profile.

Will do that for sure to get the "ultimate" peace of mind!

My Epson 9800 has tons of settings (color management, head height (paper thickness), vacuum levels, quality, paper type, roll sie, cutter, etc) and I've found some improvement tailoring these for off brand papers. Then I save the setup using a name specific to the configuration and make profiles with targets from the same one. Don't know about the Canon nor how it interacts with PSP. I can't get PSP running on my Windows setup for the 9500 II probably because it's old and Canon hasn't updated it for current Photoshops when using the long-in-the-tooth 9500.

Yep! It looks like there's no more love for your oldie Canon...
https://www.canon-europe.com/printers/inkjet/pixma/professional_photo_printers/print_studio_pro/

[vikcious]

I have uploaded the following new profiles, based on the target printing from within i1Profiler with "Color Mode: Color and Color Settings: Matching Off]" in the printer settings.
I1P_AMK_650_SatinSemiGloss270g v2
ARG_AMK_650_SatinSemiGloss270g M2 MQ

https://drive.google.com/open?id=1JAZd45TaIKzA5ZWAG_q_Vw2u9fF-LLV7

Second is Argyll version based on the measurements export from i1Profiler.
As of now I am a bit confused since if I print from PS with Printer is managing colors --> Set in printer settings "Color Mode: Color and Color Settings: Matching Off", the target prints nice but upon scanning in Argyll I get quite a lot of warnings related to Delta >=30+ in some patches. Not sure what exactly could be the problem though! The profile is build but it doesn't look good... You can see it as "ArtMaker_SatinSemiGloss270g_i1p2"

Btw... the paper now is a RC, Satin SemiGloss since I have run out of the previous Luster.

[Doug Gray]

Btw... the paper now is a RC, Satin SemiGloss since I have run out of the previous Luster.

Your printer should be printing blacks with L*=3 or so. The fact most all your profiles and datasets are showing L* in the teens indicates you are selecting the wrong paper type which causes the printer to select the wrong ink. Matte ink will produce really bad results on paper designed for PK (glossy, semi-gloss, luster, etc).

You need to focus on getting the correct paper setting.

Print some black patches measure them with the I1Pro2 in spot read mode. I1Profiler and Argyll both have tools to do so. In I1Profiler select "Measure Chart", enter the number of rows/cols and select spot reading mode. Double click on the patch after it's read and you will see a breakdown on the left.

[vikcious]

Your printer should be printing blacks with L*=3 or so. The fact most all your profiles and datasets are showing L* in the teens indicates you are selecting the wrong paper type which causes the printer to select the wrong ink. Matte ink will produce really bad results on paper designed for PK (glossy, semi-gloss, luster, etc).
You need to focus on getting the correct paper setting.

Let me see if I got this right? When I print from i1Profiler the "Media Type" I should choose for this type of paper I am trying to profile (Satin Semi-Gloss) would be "Canon Satin Photo Paper 240g"... being the closest I could identify to mine, correct?

As for this 2 specific profiles I've created I am not sure what paper I've chosen and it could have been not the right choice.
Btw...  there is also a thick-box in i1Profiler, Measurement area, called Paper Information... could that too, if wrongly chosen, impact the quality of the profile, specifically the L* for blacks? 

Print some black patches measure them with the I1Pro2 in spot read mode. I1Profiler and Argyll both have tools to do so. In I1Profiler select "Measure Chart", enter the number of rows/cols and select spot reading mode. Double click on the patch after it's read and you will see a breakdown on the left.

Will do! Now I'm more in control so I should be able to progress faster in profiles quality...

[Doug Gray]

Let me see if I got this right? When I print from i1Profiler the "Media Type" I should choose for this type of paper I am trying to profile (Satin Semi-Gloss) would be "Canon Satin Photo Paper 240g"... being the closest I could identify to mine, correct?

Yes

As for this 2 specific profiles I've created I am not sure what paper I've chosen and it could have been not the right choice.
Btw...  there is also a thick-box in i1Profiler, Measurement area, called Paper Information... could that too, if wrongly chosen, impact the quality of the profile, specifically the L* for blacks? 
Will do! Now I'm more in control so I should be able to progress faster in profiles quality...

The "Paper Information" in I1Profiler's measurement area has no effect on anything. It's just informational when you save the workflow.

[vikcious]

Yes
The "Paper Information" in I1Profiler's measurement area has no effect on anything. It's just informational when you save the workflow.

Good to know, thanks!

Btw... looking forward for having you on board the Canon ship too! I just hope you'll not be disappointed with the PRO-1000 weird ink wastage behavior (cleaning cycles related) that has already made it "famous"!

[Doug Gray]

Vic,

Your last 3 profiles are much better:

ArtMaker_SatinSemiGloss270g_i1p2.icm
P_AMK_650_SatinSemiGloss270g v2.icm
ARG_AMK_650_SatinSemiGloss270g M2 MQ.icm

All look pretty decent with low L* consistent with PK (glossy) black ink.

One has quite a bit of uV, Another is targeted for the Spectrolino? Two are Argyll profiles and the other an I1Profiler.

All the gamuts look decent. However, best to use only M2 profiles. This provides the most consistency and lets you check color using both the I1Pro2 and CM.

[vikcious]

Vic,

Your last 3 profiles are much better:

ArtMaker_SatinSemiGloss270g_i1p2.icm
P_AMK_650_SatinSemiGloss270g v2.icm
ARG_AMK_650_SatinSemiGloss270g M2 MQ.icm

All look pretty decent with low L* consistent with PK (glossy) black ink.

Indeed I have reached the point where I am happy with the overall quality of the profiles. Right now the reported gamut is always on par or superior to the original Canon profile, for similar paper, and this is the first time this happens.

One has quite a bit of uV, Another is targeted for the Spectrolino? Two are Argyll profiles and the other an I1Profiler.

That "Spectrolino" is what Argyll is wrongly adding to the "TARGET_INSTRUMENT" when converting the CGATS from i1Profiler.  Nothing to worry.

All the gamuts look decent. However, best to use only M2 profiles. This provides the most consistency and lets you check color using both the I1Pro2 and CM.

Dear Doug,

I think that we can conclude that, once the target printing mystery has been somehow solved /me happy with the results I'm currently getting, the profile comparison between CM and i1Pro2 is no longer relevant.
Again, your help and advice have been priceless and really guided me a lot through this Canon PSP mess and I thank you.
Anyway I am really curious and eager to see your input once you get your new Canon baby!

[Doug Gray]

That "Spectrolino" is what Argyll is wrongly adding to the "TARGET_INSTRUMENT" when converting the CGATS from i1Profiler.  Nothing to worry.

The issue is rather subtle. Argyll will interpret spectra based on the instrument that recorded the data and over the years there has been some variation between models. Tagging CGATs from I1Profiler with something other than the I1P can cause the spectrum to be interpreted slightly differently. These are very minor effects but was one of the reasons for XRite standardizing on XRGA. In any case it's close to a negligible effect.

Congrats on getting a handle on your printing color management. Normally, I would not jump in on such a panoply of printer issues. Especially a printer I don't have or am not familiar with. In this case it was a printer type I have been considering purchasing for some time after reading Mark's detailed review. That and your facility with Argyll and command line processes kept me interested.

Best wishes printing.

[GWGill]

The issue is rather subtle. Argyll will interpret spectra based on the instrument that recorded the data and over the years there has been some variation between models. Tagging CGATs from I1Profiler with something other than the I1P can cause the spectrum to be interpreted slightly differently. These are very minor effects but was one of the reasons for XRite standardizing on XRGA. In any case it's close to a negligible effect.

Traditionally Argyll uses the instrument values just as they are. For consistency, I did not changes that when I added XRGA support (all the measurement tools default to "Native").
Importing from .txt CGATs formats perhaps could be smarter - I'm not attempting to parse X-Rites instrument and XRGA information. You can work around this manually for imported files using spec2cie, which will allow you to do arbitrary XRGA conversions. (Building this into txt2ti3 would make it simpler).

[Doug Gray]

Traditionally Argyll uses the instrument values just as they are. For consistency, I did not changes that when I added XRGA support (all the measurement tools default to "Native").
Importing from .txt CGATs formats perhaps could be smarter - I'm not attempting to parse X-Rites instrument and XRGA information. You can work around this manually for imported files using spec2cie, which will allow you to do arbitrary XRGA conversions. (Building this into txt2ti3 would make it simpler).

Thanks Graeme. Good to know. So the tag is, in Vic's workflow, informational only.

I did notice, back in 2017, an average difference of about 1.0 dE00 in patches read from the I1Pro 2 in M2 and the iSis 2 in M2. I chalked it up to the age of the I1Pro2. OTOH, it's not inconsistent with X-Rite's published data on spectro differences.

These are all minor effects for most purposes.