Sandy,
Thanks for posting your response.
>...Your platinum print made using the inkseparation page printed with an Epson K3 inkset gives similar results to what I get with carbon printing using 50% calibration.
I see from the linearization data in the profile you posted that you have the same low-slope curve as you approach the paper white. In traditional silver printing the toe and shoulder lower slopes were, in my view, major issues. Bruce Barnbaum bleached many if not most of his prints to fight this at the paper white end. I did that step occasionally, but with the advent of digital, using digital tools to fight, particularly, the gray highlight problem, seemed like a better (and safer) way to go.
The linearization data in your profile have such a low slope that you're close to the point where QTR will reject the data.
I'm curious why the linearization data is from dark to light. I had to reverse the order to get QTR to accept it. I'm on Windows, but I thought QTR was the same with respect to this issue for both platforms.
I've used the straight line curves a lot for testing, but typically we need a sharp upturn at the dark end to have a more linear response on the inkjet paper. The QTR built-in "black only" curves have this. On the other hand, those built in BO curves do not have the rest of the slope characteristics your data show.
I would be curious to see scans of the highlights of prints made with this sample curve. Much of what we do with partitioning is to "hide the dots" in the highlights. When the image in inverted, I'm not sure what the differences are going to be. The platinum print is so rough, I doubt anything will show.
> ... And just for the record, carbon is a process that is as sharp or sharper than silver, and carbon prints are often made on very smooth papers similar to those used for silver.
(We do share our fondness for carbon as the imaging substance, even if in different realms.)
What papers are you using that are as smooth as the base for the silver print? I work with a lot of watercolor painters and have found that Arches Hot Press is about the best, but then I'm also rejecting many/most papers based on poor dmax results when printed directly with the carbon. That is an issue your carbon process may not be concerned with.
> ... In my personal work with carbon I use a QTR profile of my own creation based on a all gray ink set using various mixtures of Epson K3 PK, LK and LLK.
Do you stay with the straight line CURVE_K= "0;0 100;100" and copy curves for the inks?
I did a lot of experimentation with Black Only curves for the old Epson 1800, where MIS Associates commercialized what I called a "3 MK" inkset that used 3 MK positions. I found that the optimum number of MK curves seemed to be about 4, but 3 hid the microbanding sufficiently for most printers. I also found that having a single MK in the highlights made them look smoother. The "ordered" QTR dither pattern actually looks smoother than a random pattern. The single ink did this. When another one was added the pattern became more random and looked rougher. But, again, what happens when an image is inverted is something I've never explored (and probably will not, given my expected short term involvement in the alt process/internegative end of the field).
> ... In testing profiles for carbon printing I make extensive use of a UV transmission densitometer as it saves a lot of time in evaluating step wedges. I also linearize with an iOne spectrophotometer.
I've been using my Nikon 8000 scanner to look at and measure the steps of the negatives. I'm uncertain how accurately this maps to the UV spectrum. The little info I found on the web seemed to suggest that for carbon the curves would be rather congruent but that the carbon blocked more UV light than green light. Have you ever seen or generated data on this issue and the applicability of film scanners to the process?
One thing that I have found with film scanners is that it's hard if not impossible to stop them -- or at least their OEM drivers -- from applying a built in profile. This can significantly compress the dark end of the curve. I try to use manual exposure and pull the sliders out to the max to try an avoid the effects of this. It's hard to say how successful this approach is.
Paul