ArgyllCMS 875-patch printer profile with i1Pro 2

[Ink stained Fingers]

Is this problematic reflected light coming from the devise or is it ambient light?

It's how the paper surface is reflecting the measuring light coming from the spectrophotometer, and the vast majority of papers can be tested with the standardized 0/45 deg light path in the photometer before the light hits the sensor. But as mentioned a few - a very few papers come with a coating which let the light being dispersed wider than on most other papers.

Here are some links explaining the working of a photometer and the light path, it is mentioned somewhere that the 'normal' spectrometer is to be used on matte - smooth surfaces which excludes highgloss structured surfaces

https://www.xrite.com/blog/what-is-a-spectrophotometer

It was the above gamut plot at the lower luminance, the ondulated line of the gamut which made me remember that I had seen a very similar plot long time ago. Again - yes - the explained effect can cause some weird looking gamut plots but as a rare instance . Try the above mentioned actions - rescanning and as well using a smaller patch count

 

[pkk]

I've attached another profile on the semimatte paper. We can at leasat see if the paper is the problem.
Thanks again.

 

[The Hat]

Yes, changing the patchcount can get you around the problem , but the first problem is that you need to identify it

I can read along and can understand what your trying to achieve, but this type of profiling is beyond me, so my question:- why do you need so many colour patches to achieve your goal, after all we’re only dealing with basic Cymk and their uncle and aunts

Excuse the ignorance.. !

 

[Ink stained Fingers]

This is the gamut plot at L*=13 for the 3880 PremSemimatt profile of June 29/24, it looks pretty normal

after all we’re only dealing with basic Cymk and their uncle and aunts

every image pixel has a particular color which is a mix of the CMY base colors, and the mix ratio of these need to be controlled and adjusted because the CMY base colors are not theoretically pure , and the image carrier - the paper adds a tint as well which needs to be compensated. But you are right - there is no specific requirement that exactly 924 or 875 patches need to be used in the profiling process , interpolation does a huge job to calculate these correction factors for all of the in between 2 exp 24 = 16 million - possible RGB (8 bit) colors. The profiling process actually controls the color appearance of these 16 million possible colors based just on these few 100 or 875 or 924 color patches.

 

[pharmacist]

I suggest to scan quite slowly: start with pressing the button and wait for about 2 seconds before moving the i1Pro2 to the other side of the row in about 4 seconds time and when you reach the other side, hold the button another 2 seconds before releasing it. It take some exercise to move the i1Pro2 with a steady slow speed over the row to be scanned. This way the scanning will be more repeatable and steady.

But maybe you can have test with this simplified 480-patch target with elongated colour patches: this ensures that the i1Pro takes more samples of the same colour when swiping the i1Pro2 over each row. @Ink stained Fingers : maybe you can give this target set a try with ArgyllCMS to profile your problematic reflecting glossy paper and report back.

@pkk : I did a softproofing in PS and it seems to be perfectly ok, suggesting the previous profile had some scanning errors. Also in iccview.de the volume and body of the profile seems to be perfectly ok. This confirms that the second scan has been properly done. I always do a softproofing to see if there are any strange colour transition. If everything is ok, then you know you have properly scanned the target.

 

[pkk]

Here is a completely new profile from the 9880 on Epson semimatte. I used all of the suggested techniques for scanning the targets.

 

[pharmacist]

Both newer profiles are ok to my eyes and don't have any visible errors in both iccview.de and softproofing in PS (relative colorimetric and NO black point compensaton). You are doing more than well.

 

[pharmacist]

Today I made a profile with the lower 480-patch target and see if there is any visible difference compared to the higher cound 875-target. The result might surprise you: to my eyes the difference is not visible.

Above the 875 patch target and below the specially made 480 low patch target

Here below a print with a test image: upper image with the 875-patch target profile and lower image with the low 480-patch target profile:

 

[Ink stained Fingers]

Yes, tlhis last 9880 Epson Semimatte profile looks pretty much o.k, gloss related problems may show up earlier on glossy - semiglossy papers.

maybe you can give this target set a try with ArgyllCMS to profile your problematic reflecting glossy paper and report back.

I regret but I threw away those offending papers long time ago already

 

[Ink stained Fingers]

The result might surprise you: to my eyes the difference is not visible.

It's the visual impression at the end which counts , and if the profiling passed without errors differences in prints should be limited to very small color ranges , and you would need some monochrome area with such color as well in that image at which you could see that difference.