<I guess the response to this would
be, "How does one know that the max has been reached?" - as it was
argued that the limits are too low and cannot be increased, thereby decreasing
the gamut relative to other RIPs>
Rich, because each printer model is its
own project so to speak. When a new printer comes in the door, they create a
new set of recipes for that model. These recipes are created by their custom
software, and control ink limiting, screening, etc. This is why I had mentioned
that running a third party inkset might create problems, if their attributes
were very different than the properties of the Epson inkset (or Canon, HP). As
far as ink limiting, the “standard” behavior of the driver ( in
your case it would be the 9600 Ultrachrome driver) puts down as much ink as
possible without flooding the media. I know that a variety of papers are used
to determine this sweet spot. Some media may require ink limiting down to say
90% per channel, this would typically be backlight film, some vinyl materials, reverse-print
polycarbonates, and other non–paper materials. The bottom line in terms
of photographic media is that if you find a paper that can’t cope with
the standard ink limits in ImagePrint, the media sucks. I suppose that its
possible you could find yourself with a slightly larger gamut by really
slamming certain papers, but it’s a fine line and ImagePrint is
attempting to reduce bronzing and create the smoothest output possible. Its
black box in that sense but certainly tuned for a specific printer model.
<<There is no linearization
feature, its is already done at a level much higher and more complicated
than the number of steps you get with applications like Studioprint.>
Can you be more specific? Is this a part of building a profile, or can it
be done separately?>
If I remember correctly each channel in ImagePrint
is linearized in at least 256 steps, but of course the whole process is more
complicated than that. All of these properties are rolled into the recipe which
is part of the individual driver. In Studioprint 10 you had the option to linearize
each channel with 10, 20, or 40 patches. This isn’t enough sample points,
and using the same profiler (which takes that out of the equation) I was consistently
seeing smoother tonal transitions with ImagePrint, especially in the
highlights. If I sampled 40 patches for each channel with a spectrolino and did
no such thing with ImagePrint yet my output was still smoother with ImagePrint,
I would consider Colorbyte’s technique pretty sound. You cannot alter the
recipes as an end user, they are simply built into the driver. I admit this isn’t
all that scientific, but I’m not interested in the math, only what is in
my hand. That’s what customers buy.
There are two reasons one might not
choose to run ImagePrint. First the B+W crowd with dedicated inks like Cone. I
don’t see the point in doing this any more but Studioprint would give you
more flexibility. The other reason would be if you had the 9600 and something
like a Mimaki JV-160, Mutoh, or Colorspan. ImagePrint doesn’t support
many printers that are supported by say Onyx Production House, Wasatch, and
many of the other companies. It probably would not make a lot of sense to deal
with training staff on 2 different RIP’s.
Lastly whatever RIP you decide on Colorburst,
Ergosoft, Colorbyte, Wasatch, please realize that there are yearly agreements
for support and upgrades. In the case of Colorbyte it is $500 a year for
upgrades and unlimited phone support. I figured I’d warn you now.
Cris Daniels