Re: [Gimp-print-devel] Strategy for profiling Epson R280 with Ultrachrome K3 inks

[Robert K. <rl...@al...>]

Marko Milisavljevic wrote:
> Hello list,
> 
> I have an Epson R280 printer loaded with MIS K4 inks (3rd party
> substitute for Ultrachrome K3 inks), without LK and LLK, of course. I
> am trying to figure out the best way to profile it. I will probably
> use ArgyllCMS for profile generation.
> 
> If these inks were standard on this printer, I would treat it is RGB
> printer and print patches using "Uncorrected" mode, but that is not
> the case. I am thus looking for advice on general approach, so I don't
> go barking up too many wrong trees. I imagine I need to do this:
> - Print a gradient for each of the 6 inks using "Raw" option to figure
> out ink limits for individual channels and where is the crossover
> between photo and regular inks. Also print combined inks to figure out
> total ink limit.
> - Tell Gutenprint about these limits, ideally without recompiling
> (simply change settings in driver? modify PPDs or
> /usr/share/gutenprint/5.2/xml/escp2/media/something.xml maybe)
> - Linearize - not exactly sure how. I see Gutenprint has some HSL
> curves in media/something.xml - I assume this is the famed
> "linearization"? I don't see how to linearize individual inks or
> combinations of inks (M+LM) other then some gamma settings.
> Alternatively, print unlinearized with "Density" option and let the
> profiling software sort it out (I'm working in 16-bit).

So this printer is essentially a CcMmYK printer, correct?

One option you might start with -- you'll need the latest CVS to do this 
-- would be to copy the appropriate data file from escp2/inks and change 
escp2/model/model_81.xml to point to the new file, and change the 
appropirate escp2/media file (probably start by copying the file for one 
of the K3 printers).  Assuming that the K4 ink really does match the K3 
ink, you'll get (basically) the same results as a real K3 printer, other 
than the light black inks.  The good news is that you won't need to 
recompile anything to make this work!

After you do this, you'll need to regenerate the PPD file.  You should 
be able to do this with cups-genppdupdate.

The HSL curves are not linearization curves; they're color correction 
curves for people not doing their own color management.  You'll be using 
Uncorrected, Density, or Raw color correction, so they won't come into 
play.  The parameters you're interested in are:

	+ The Value parameters (such as Light Cyan Value) specify the
	  relative darkness of the light vs. dark inks.

	+ The Cutoff parameters now specify the point at which the
	  darker ink is first used, relative to the value of the light
	  ink.

	+ The Scale parameters now specify the amount of light ink
	  used (essentially the density of the light ink).  This is
	  the same as the parameters previously called Cutoff.

Does this help you get started?

-- 
Robert Krawitz                                     <rl...@al...>

Tall Clubs International  --  http://www.tall.org/ or 1-888-IM-TALL-2
Member of the League for Programming Freedom -- mail lp...@uu...
Project lead for Gutenprint   --    http://gimp-print.sourceforge.net

"Linux doesn't dictate how I work, I dictate how Linux works."
--Eric Crampton