Digital BW, The Print

While tweaking some curves, and checking out some of the grayscale gradients
I've collected over the past few months I finally gave a serious look at
Tyler's Zees (which can be downloaded in the files section of this group on
Yahoogroups. Files: General image processing info: Tyler's Zees.

Below, I've included Tyler's explanation to me of how he made the chart, and
what to look for.

Anyway, the reason I bring it up is that it is a very challenging image to
print for my MIS VM workflow, and I would suspect any monochromatic
workflow. As I don't own the Piezo BW driver, I can't test how successful it
is at printing it, and I am most curious to know. I've been able to pass
Cone's "Piezo Challenge" test image in the past with Roark's curves, but I
believe this "test" is a greater challenge than that.

I encourage anyone who's interested in really seeing what their workflow is
capable of handling to give this test image a whirl. I'd also be interested
in hearing how well you all fare. I'm particularly curious to hear how well
Piezo handles it, at least for papers with their own profiles, as well as
any of the other workflows that have been discussed of late (black ink only,
Wolf's curves, Color ink Duo tones, etc.).

In my own case, using Roark's VMQNC curve on my 1160, with my own custom
gamma curve (which gives me my best results so far), I'm able to get
discreet separation between all but three squares, and reasonably discreet Z
separation in all but five squares.

Now I can make good looking prints in spite of this, so failure to excel at
this test is by no means cause to chuck your system, but it is terribly
revealing of where flaws live. Unfortunately it only helps us recognize
where to look for trouble, but not how to fix it....

Todd Flashner

This from Tyler Boley:

Within each square is a gradation of 5% right to left. In other words,
the bottom right square gradates from 1% to 5%, the next square from 6%
to 10%, with enough total squares to get to 100%. The last (upper left)
square is 96% to 100%.
Within each square is a Z made up of solid value of the middle
percentage of that square. So the Z in the bottom right square is 3%,
the one in the next square 8%, etc.. It was created in the 2.2 gamma
gray space.
I made this file to evaluate my sep curves on paper. The workflow is
applied to the file, and printed.
On the print, first look for a delineation from one square to the next
for overall gradation problems. Secondly, each Z should appear against a
background square that is faintly lighter behind it on the right, and
faintly darker behind it on the left. I've found areas that need work
are fairly well revealed with this.
Dan decided it was useful for actual curve development by eye on the
monitor if you are working with RGB quad profiles, that's why he put it
with his procedure.
But I use it mostly for verification, it seems to show me specifically
where things need attention that are creating problems in images. The
other interesting thing is that often curves seem fine, then suddenly a
different sort of image looks bad, but this will show almost any problem.