Refilling Pro-200 with inks for Pro-100? A Good idea?

[stratman]

If you can get better, there is a$1000 on the line,.

This isn't about your wager. It is about your test leading to a result that is of indeterminate utility.

This is about the comparison of your custom ICC profile vs a mismatched ICC profile and passing this off as not only of value but also as relative to OEM Canon quality - "Closer" and "Catching Up".

This is about your methodology, not about the quality of your ink.

This is about respecting the forum. A mismatched ICC profile and mystery ink used in the reference print does not promote confidence.

You have presented yourself as an advanced practitioner in the printing field with your own designed inks and ICC profiles. You sell your custom inks. The forum has been a source of referral and revenue for you. Therefore, the standard by which you are held in your data and claims is higher than that for the novice.

It is unproductive to deflect with a straw man argument about a wager.

Please explain clearly to the members your reason for selecting these variables, why you did not identify all your variables, and why you believe your results are useful despite the use of an unmatched ICC profile and unknown ink for the reference print.

Or, delete and replace it with quality data so a useful conclusion can be made.

 

[palombian]

I have no experience with dye inks and the PRO-100/200, but also with pigment inks I observed the profiles published by PC are in general shifted to the left and the bottom (so to a- and b-) but this can be related to the color temperature (D-level). I am no expert, the only thing I could do was test the inks myself.

According to my own measurements (with the Colormunki) for the PRO-10 the gamut of the PC inks is very wide and the extreme points of CMY have even a larger value than the OEM ink for this printer (but not as the wide format OEM inks, plausible since those are the inks where 3th party labs can make a business).

I agree with previous posts the used "strawberry" test sheet is not targeted to extreme gamuts and that with correct calibration exactly the same image can be reproduced with most inks on most printers (as a test last year between the Epson P400 and the Canon PRO-10 showed).
I made a lot of tests with this strawberry sheet with 5 different brands of ink on all paper types, and except for the black level and some red nuances, all look almost the same.
For the average photo a correct profile is more important than the gamut of the ink.
But since the average refiller can't make his own profiles he will be as dependent on the 3th party ink provider as to OEM. A comparision with OEM using the strawberry sheet is of coarse a proof the ink is up to the task but does not exclude other inks are also.

Again, I can't judge @mikling 's theories about the gamut tricks Canon uses with dye inks, but for pigment inks until now custom profiling can compensate the (sometimes very visible) color values of the different inks.

 

[Ink stained Fingers]

@palombian , you are addressing various issues with profiles and some of the complexity behind it which the average user does not see. Let me just illustrate this with the strawberries as an example with a crop from the complete test image

This image comes with a range of quite saturated red tones, I'm displaying the image colors - by image pixel - in the Lab color space together with a profile mentioned above already - the PC42SE Red River PecosRiver Glossy paper as an overlay

The light red gamut shows the mentioned profile, and all the yellow dots are the image pixels, almost half of them are outside the gamut which means that such out of gamut colors are mapped to colors with a lower saturation level. But that's not the prime function of a profile, and using images with out of gamut colors are not very helpful to judge and compare color profiles on prints. That could be a seperate action to see how out of gamut colors are printed with different rendering intents.
I'm using the old MonacoGamutWorks utility for this type of display, the Gamutvision software let you display the mapping of colors by a profile - the out of gamut colors - with vectors , but it is not very illustrative I think so I don't use it here.

 

[Ink stained Fingers]

@mikling has published a large collection of icc-profiles for the Pro-200 and Pro-300 for the new corresponding inksets

http://precisioncolors.com/PC65.html

http://precisioncolors.com/PC300.html

precisioncolorts is the only website I'm aware of which publishes profiles not just for the standard D50 illumination condition but for a few more of those.

I compared a few of those with the icc-profiles which are published by RedRiver for the same papers for use with the genuine Canon inks. Profiles have apparently been generated with different software - not specified by RedRiver/Chromix,a and probably different hardware as well.
Those profiles I checked are very close together - slightly beyond measuring tolerances - the gamuts match very well and the variations should not be perceiveable in actual printouts. But the main part of the color corrections with a profile happens within the gamut anway and cannot be demonstrated that easily.
I just assume that all profiles are made with the M0 measuring condition which includes some of the influence and color shift caused by optical brighteners.

 

[palombian]

@mikling has published a large collection of icc-profiles for the Pro-200 and Pro-300 for the new corresponding inksets

http://precisioncolors.com/PC65.html

http://precisioncolors.com/PC300.html

precisioncolorts is the only website I'm aware of which publishes profiles not just for the standard D50 illumination condition but for a few more of those.

I compared a few of those with the icc-profiles which are published by RedRiver for the same papers for use with the genuine Canon inks. Profiles have apparently been generated with different software - not specified by RedRiver/Chromix,a and probably different hardware as well.
Those profiles I checked are very close together - slightly beyond measuring tolerances - the gamuts match very well and the variations should not be perceiveable in actual printouts. But the main part of the color corrections with a profile happens within the gamut anway and cannot be demonstrated that easily.
I just assume that all profiles are made with the M0 measuring condition which includes some of the influence and color shift caused by optical brighteners.

Impressive profiles, large volume and deep blacks.
In practice this is a good indication of real performance (if the deepest black of the profile is L=2, I can measure the same value in a test print.
Never expected the PRO-300 inks would be different from PRO-1000, maybe why all 10 colors could be reproduced (instead of 8 of the 12 for the PRO-1000).