Digital BW, The Print

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, 
"Charles Bandes <byronbulb@y...>" <byronbulb@y...> wrote:
> All I can say is "wow this is hard"
> 
> I have been trying to modify the VM curves for my 1280 without 
too
> much success so far. The standard curve Roy made works ok 
but the
> darker midtones get all bunched up without much separation 
from 75-55%
> I tried playing with the magenta curve to see if I could increase 
the
> separation b adding a little of the toner to those darker values, 
but
> I just made things worse. Basically I just don't have enough
> understanding to fiddle with the curves yet with anything 
beyond
> "error and error" results :)
> 
> Anyway, yes, Roy, Paul, Tyler, whomever, please educate us! :) 
I have
> a feeling that with suitable curve development the gimp-based
> quad/hextone systems might become _the_ systems, it's so 
exciting!!
> 
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, Kevin 
Gulstene
> <kevin@d...> wrote:
> > Do you have a systematic method you used when developing 
the curves 
> > used to generate the  "quadtone curve description files" or 
did you use 
> > your own experience and experiment.
> > 
> > I am interested in modifying some of the curves you provided 
and it is 
> > unclear to how best to proceed.  Trial and error seems like it 
would 
> > take a lot of experimentation.
> > 
> > --
> > Kevin Gulstene

Glad to see some of you guys trying it out.  First of I have to
say that I'm far from being an expert at designing curves, but
I do have a general method in principal.  The major advantage
to this system versus the RGB workflow is that you can 
control each ink individually.  So here's my approach:

My ink set is the MIS VM on an 1160.  So I have 4 inks but
only 3 are the grays to produce the smooth step wedge.  The
4th is a toner which can be added or not.   In case anybody
is not familiar with this ink set,  Cyan is dark gray, Magenta
is light gray, Yellow is the toner and Black is of course black.
For those with 6 ink systems, the magenta and yellow are
reversed -- i.e. Y = light gray, M = toner.  There are 2 other
inks (to make 6) but the inks and the driver have all that
builtin and invisible to this whole issue.  So if you have a 6
ink system just reverse M and Y in this description -- every
thing else should apply.

Part of my principal is to use about 200% to max 300%
total ink coverage.
I made 4 test curves that just use one ink at a time so you
can easily see what densities they each have.

The first thing that is desirable is to have target densities for
the whole step wedge (21 steps should be fine).  I would
build the wedge then in three parts.  The low density end
will basically be constructed using just M (light gray).  So
based on the target ramp you ought to be able to look at
the M (light gray) only test curve and see how steep is
necessary.  I figure it makes sense to have the light gray
go up straight to at least about 80% then have it smooth off
and come back down to 0 some where about 2/3 of the way
across. It's not going to have a lot of effect in the denser
areas so it just ought to be smooth.   Next, I would work on
the black end.  This is a little different.  If you run the TestBlack
curve i.e. just use the black ink its really not dense enough.
So the idea is to have 100% C dark gray as an under coat
for the black.  This also has the effect of hiding the black
dots.  So similar to the light end, we need to use the appropriate
slope of the black curve with 100% cyan to make the dark
end of the step wedge.  Finally the C dark gray curve has
to join the light end with the dark end.  Start the C where the
M is topping off (the 80% line I had above) and end the C
curve at the top where Black is starting to come in.  This ought
to be very close to a smooth wedge.  I think you'll need to
measure a few intermediate steps to get the C curve in
exactly the right place and slope.  I haven't done much work
with the toner curve yet but I think starting with just a straight
line from 0 to 25%,  50%, and 75% will be a pretty good start.

It looks like I should make up a TestBlackwith100Cyan
curve for the dark end.  One major question I have is how
smooth the curves should be  -- I wonder if simple straight
lines would actually work just fine.  Also if the paper can take
more ink, some more light gray all the way to the black
end might give higher dMax.

For those with 4 gray inks (like the MIS FS or Piezo Inks) I'd
do the two light grays together -- up and down -- same
shape but shifted slightly.

As I get more practice myself and any feedback from others
I'll include that in the download.

Enjoy! 

Thanks,
Roy