2007-03-15

Ok, I'm even further behind than I first thought. I have been working on a standardized format for posting printer color and tonal accuracy information on tested papers. I know it isn't going to be perfect in the first round, but I was hoping to zero in one set of optimized results obtained by using a custom ICC profile. I realized that the GM Profilemaker software I use can produce a number of "flavors" of perceptual rendering, and it wasn't clear to me at this point in time whether I should just standardize on one perceptual rendering flavor or opt for relative colorimetric rendering and black point compensation. Would any rendering method I chose always produce the best result? Probably not. Then too, there is the issue of UV cut-off filter versus no UV cut filter on the Spectroscan used in conjunction with Profilemaker's software compensation for optical brighteners (OBA's). I had pretty much made up my mind to standardize on the use of a UV cut-off filter so that I wouldn't be dependent on a software detection of OBA's and subsequent proprietary compensation algorithm. However, one paper I've been testing, Innova Fiba F Gloss Bright White (with OBA's) reads warmer with the UV cut-off filter on the Spectroscan than it looks visually under a Solux 4700K light source compared to other papers that contain no OBAs. This is a little puzzling because the Solux lamp is claimed to have negligible UV output. If this low UV output lamp still triggers florescence that leads to a more bluish paper white appearance than is measured with UV cut filter on the spectroscan, it gives pause to rethink my choice.

Given the myriad variables, a little more testing and refinement is in order before I start posting routine results. Maybe all the different rendering methods are interesting and I should put all of their results into each report! I fear that would be TMI (too much information). Yet I do want to sort out what gives the best results for each printer/ink/paper combination. The goal is always to show best results. Anybody can make mediocre results at any time.

Can the I* metric even distinguish the difference between different profiles? An interesting question and one that needs more study. So I've turned my attention to Case Study #2 which will explore printing without profiles, printing with generic profiles, and printing with custom profiles on my Epson R1800 printer using Epson Premium Luster Photo Paper. Numerous test results have now been collected, and yes, the I* metric does identify differences in the custom profile "flavors" generated by Profilemaker 5.0.8 software. It also easily distinguishes between the Epson standard driver setting, the colorsync setting with a generic profile, and the custom profiles printed in conjunction with the Epson no color adjust setting (NCA). I'm busy writing up the results, but it is taking longer than expected and I probably won't be finished for another few days. Check back soon if you are curious about the test results.