z3200. How to determine the optimal ink limit for particular paper?

[SergeyT]

This will be a long one, sorry...

Preface.
I have measured the PK patches density of the Color Calibration target on HM Photo Rag Baryta with the Spyder3Print and to my surprise got the following numbers(from dark to light):
2.31 2.43 2.33 2.24 2.07 1.89 1.71 1.58 1.41 1.25 1.07 0.93 0.74 0.56 0.39 0.21. Seeing the 2.31 followed by 2.43 is something unexpected for me.

Out of curiosity I went ahead and measured all the other Color Calibration targets ever printed on my z (I keep them all). This is what I've got( a few samples)

HP Pro Satin
From the first for this type of paper calibration: 2.26 2.36 2.31 2.18 2.06 1.88 1.73 1.57 1.43 1.29 1.11 0.96 0.78 0.59 0.42 0.23
The next one few minutes after(same roll):       2.49 2.41 2.34 2.14 1.96 1.78 1.63 1.49 1.38 1.22 1.06 0.90 0.74 0.58 0.41 0.24

HP ID Satin
From the first for this type of paper calibration:                2.16 2.27 2.27 2.22 2.09 1.93 1.77 1.61 1.45 1.29 1.12 0.96 0.77 0.59 0.41 0.24
The next one a week later (same roll YM head replaced) : 2.37 2.38 2.27 2.14 2.00 1.76 1.58 1.49 1.27 1.13 1.01 0.88 0.73 0.56 0.40 0.23

HP Everyday Satin
From the first for this type of paper calibration: 2.29 2.29 2.25 2.19 2.07 1.97 1.83 1.69 1.57 1.39 1.18 1.03 0.85 0.64 0.42 0.24
The next one (same roll)                               : 2.27 2.27 2.29 2.20 2.09 1.97 1.84 1.70 1.58 1.41 1.23 1.05 0.86 0.64 0.44 0.25

On Litho-Realistic the sequence is always decreasing for PK: 1.50 1.46 1.37 1.31 1.24 1.17 1.10 1.02 0.95 0.87 0.77 0.67 0.57 0.45 0.32 0.20
but similar to Satin story for                                         MK: 1.75 1.80 1.59 1.50 1.39 1.31 1.23 1.14 1.04 0.93 0.82 0.73 0.61 0.48 0.34 0.19
Another one (same roll)
                                                                                 PK: 1.53 1.44 1.35 1.29 1.23 1.15 1.08 1.01 0.93 0.86 0.76 0.66 0.56 0.44 0.32 0.19
                                                                                 MK: 1.73 1.70 1.61 1.51 1.42 1.33 1.26 1.18 1.07 0.94 0.63 0.74 0.62 0.50 0.35 0.19

I do not think there is an error on my side or on the tool (at least in relative values).
I would expect the first patch to have the highest density or at least have it equal to the next dense patch. But this is mostly not the case. In most cases the sequence is improved on the next Color Calibration run (see Pro Satin).

Questions:
1. What do you think is going on here?
2. How to make sure the printer came up with the "best" and consistent ink density numbers after the C. Calibration?

Now what I'm trying to achieve ...
Given that there are a few built-in presets to start with working with non-HP papers and each has its ink limit set at a fixed number such as 46, 37.5, 32.5, there is no guarantee that any of these numbers will be the precise match for a non-HP paper. We can pick the "best" matching preset out of 4 or 5 built-ins but  in most cases it will be just a relatively good match (a starting point), since the ink limit is fixed.
Yet, there is the Ink Limits setting in each preset , which allows to change the actual ink limit from 0 to 120% of the selected default limit. It is set at 100%.
Let's say I would like to make a custom preset for HM Photo Rag Baryta and I start with the HP Pro Satin preset. Now my ink limit is fixed at 46 which (can assume for now) is the best for that paper out of 46, 37.5 32.5 ... available in z3200 for PK papers but it does not mean it is optimal.

Initially I was thinking about the following approach:
* Running the C. Calibration. Measuring the PK patches . Writing down the numbers.
* If the first patch has the highest density already I would increase the Custom preset Ink Limit to let say 105% and do another C. Calibration and measuring. If the first patch density is increased and still the highest I would increase the ink limit some more up until there is no change or the second patch started approaching the first in density. Then I would decrease the % to the prev step. And finally I would run a test print with plenty of blacks(dark tones) to see if the paper can handle the ink load.
* If the first patch does not have the highest density(equal to the second) it would mean there is too much ink already and I would decrease the Custom preset Ink limits by 5% and run another C. Calibration and measuring. etc.
   Should I be paying attention to other ink patches as well?
** Finally profiling the paper and printing a test image.

But now seeing the inconsistency with the numbers between each C. Calibration on the same paper I'm not sure how to tune-up the ink limits%. Any suggestions?

[John Caldwell]

The progression in densities that you expect is what I, out of relative inexperience, would also have expected. The method you describe for adjusting ink limits, to just avoid "clipping" the darkest ranges sounds sensible to me, but again, I am inexperienced. Did you ever solve this question?

What success have you had in incorporating ICC profiles made "outside" the Z3200's internal profiler into Paper Presets, if I may ask?

Thank you,

John Caldwell

[Ernst Dinkla]

Finally profiling the paper and printing a test image.

But now seeing the inconsistency with the numbers between each C. Calibration on the same paper I'm not sure how to tune-up the ink limits%. Any suggestions?

Very familiar observations. My logic was that the 100% patches had too much ink (went beyond Dmax) and more ink limitation was needed so I created custom media presets with less ink. The measurements still showed Dmax at the second or even third patch. After more trials + and - my conclusion was that the ink limitation isn't a true RIP like ink limitation but a gamma change and the Dmax will change but not its position on the patches. The Z3100 showed very linear greyscale wedges after calibration, the Z3200 doesn't. It is called calibration so there doesn't have to be the linearity you and I expect.

I think that the media presets got heavier gamma curves compared to the Z3100 ones. That and the heavier Chroma Red ink gave the Z3200 prints a more Epson style gamut, something people didn't see in the Z3100 prints and complained about. At the same time the nice B&W print linearity of Z3100 was sacrificed. See my measurements for QTR profiles:

http://www.pigment-print.com/review/Z3200FirstPage_2.htm
bottom page

Keith Cooper observed the same non linearity in QTR greyscale wedges:
http://www.northlight-images.co.uk/reviews/printer/hp_z3200ps.html


met vriendelijke groeten, Ernst Dinkla

New: Spectral plots of +180 inkjet papers:
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

[SergeyT]

Very familiar observations...After more trials + and - my conclusion was that the ink limitation isn't a true RIP like ink limitation but a gamma change and the Dmax will change but not its position on the patches.

Thanks for the replies!

I kind of suspected that same thing and just did not have time to prove / deny it. I still want to make an attempt at that once I have some spare time...will keep this thread updated with my findings.

[SergeyT]

What success have you had in incorporating ICC profiles made "outside" the Z3200's internal profiler into Paper Presets, if I may ask?

John,
 
 I find both the canned profiles (which I can download for all the papers I use) and created by the built-in profiler accurate enough for my requirements. On the other hand, according to the info I have gathered, Spyder which I have, is a no-competition to the z3200 built-in profiler, so I just did not bother trying to build any profiles with the Spyder.

SergeyT.