Digital BW, The Print

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Robert 
Morrison <rmorrison@p...> wrote:
> 
> On Monday, June 16, 2003, at 08:37 PM, Roy Harrington wrote:
> >>
> >> But, correct me if I'm wrong, without single channel control.  
You are
> >> limited to a CYMK approach?...meaning if I wanted six (or 
seven) 
> >> different
> >> inks or toners in the printer I wouldn't be able to control 
them
> >> individually, because, for example, cyan and photo cyan 
channels 
> >> would be
> >> coupled?
> >>
> >> Robert
> >
> > Yes, the 1.1 version that I think you tried was limited to
> > CMYK, but the new 2.0 version controls 4,6, or 7 inks and
> > uses full 16bit curves.  I think the best thing is that I've
> > got automatic partitioning that works very well -- no need
> > to work on going from light grays to dark grays.  You do
> > get to shape the final overall curve if you like though.
> >
> > I did post about it a couple of weeks ago, but you may have
> > missed it.  I'm still doing little tweaks but the guts works
> > just fine.
> >
> Thanks for the clarification, I must have missed the update.  It 
seems 

Here's my post about QuadToneRIP:

I'm not much for pre-announcing things -- but I guess I've got to 
say
some about QuadToneRIP. In between a bunch of other things 
I've
managed to upgrade to a new version.

First QuadToneRIP 2.0 now has all the obvious stuff:
1) Separate control of all 4, 6, and 7 ink epson printers.
2) All curves are internally 16 bit to match the gimp-print 
internals.
3) blending of two curves for neutral--to--toned ink sets.

After working for a while with these "raw" 16-bit curves, I found 
that
partitioning at this level is very simple. So the new and exciting
thing is another program I'm calling QuadProfile. With a couple
of very straightforward calibration steps to measure the relative
densities of the various inks, all the partitioning curves are
generated automatically. This works for any set and any number
of inks. The generated curves give a monotonically increasing
density but the exact shape can be customized. Adding toner to
the gray is fairly easy but manual so far.

I'm sure this will raise lots of questions so here's a few answers:

System?
This is a Mac OS X system -- Linux if you do the port.

Printers?
Any Epson printer supported by Gimp-print is possible. I'm
figuring on all the 4-digit model numbers, and several of the
3-digit models (860,870,890,960,980). Others?

When? 
The basic QuadToneRIP code has been running for quite a 
while.
The QuadProfile system is working on my computer, but it's
currently several programs and shell scripts to connect it all
together. I've been working on putting this all in one program
and one descriptor file. I have a lot of printing that I need to
do this month, so work on this is lower priority -- sorry. If
anyone is familar with running shell scripts and the Terminal
program, a beta version with shell scripts can be used. Email 
me at
quadtone <at> harrington <dot> com

Densitometer?
It's certainly a very useful tool, but I've been able to get good
results trying a flatbed scanner for the partitioning. Contrast 
decisions can for the most part be done visually.


> then that only remaining difference between the Quadtone RIP 
and 
> IJC/OPM are then the following:
> 
> 1. IJC provides expected density values (AIMS) for the profile 
and 
> allows you to enter density values and plot these to see where 
you are 
> at.

Good idea.   I should at least provide density goals.
I don't have a fancy graphical interface, but there's much less
trial and error.

> 
> 2. IJC allows you to linearize final profiles for an algorithmic 
> solution to straightening the final profile

I've only read the documentation of IJC.  But with QuadToneRIP
you DON"T linearize individual inks -- the automatic 
partitioning works with the raw ink channels.  The final
profile can be tweaked using a Photoshop Curve if you like.
I've actually gotten great curves without resorting to this.
Builtin is a final curve gamma correction and a fine tuning
highlight versus shadow correction.

> 
> 3. IJC provides an easy way (within the program) to generate a 
whole 
> variety of test patterns including single channel test wedges

I also provide a special .psd file that allows you to print
test patterns that have each ink in a separate 21step wedge.
You can use this for any curve set .  The difference is that you 
print it from Photoshop.

> 
> 4. OPM provides a blender so that you can simply make two 
profiles...a 
> cool one and a warm one and then use a slider to get 
anywhere in 
> between, instead of using separate profiles for anything.

Got that too.  On my 7500 I've got a cold curve and a sepia
curve and print with a 85/15 blend lately.

> 
> I'm not sure whether you have actually used IJC...so some of 
this may 
> not make much sense.
> 
> Also, can you remind us all what printers the Quadtone RIP 
supports? 
> and whether there are any paper size restrictions, does it 
support roll 
> paper in the wide format printers?

I use the ordinary gimp-print interface and it does show
Roll Paper as an option but I've never used it.  The basic
print dialogues allow custom paper sizes.  I created a
17x22 and a 22x17 (horizontal) size and they are working
just fine.

As far as what printers: all the 4 digit epson printers,
plus several of the 3 digit ones.

> 
> Robert

I'm currently printing a whole show of prints for the end
of the month so I've been pretty busy  -- but everything is
being printed on my 7500 with my own custom VM-sepia mix.

One last note,  although I mentioned about shell scripts
above,  you don't need to type a single command line.
Its all editing a text file, double-clicking on scripts, and dragging
the text file onto the script window.

I've had some feedback from the last beta, so I'll be getting
a new version with more documentation asap.

Roy
www.harrington.com