[argyllcms] Re: some questions for using targen

  • From: Graeme Gill <graeme@xxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Tue, 28 Sep 2010 10:48:54 +1000

Michael Schulz wrote:
it's quite a while ago I had the discussion below about using targen for
proofing purposes. At that time I didn't understand Nikolay's advice and I'm
not sure that I understand it now. But I would like to ask people more
familiar with Argyll if I'm right or wrong when I think this was meant:

This is certainly one approach to maximizing final quality while printing
and measuring as few patches as possible. The disadvantage is that you
have less flexibility to alter parameters such as ink limit and black
generation, since you would need to re-do the second print again.
(A more subtle disadvantage is that targen doesn't really know that
 the second pass is 3 dimensional, and still uses a lot of the points
 attempting to explore the K space. If you're lucky these points
 will fill out the space roughly evenly, but in practice there will
 still be more points near the black end than not. My plan was to use
 a proper two pass approach to do multi-ink profiling at some stage.)

Overall it is possible that you might end up with a better quality result
for a given patch budget. The only way to know for sure is to try it out.
To "do the experiment" you might compare:

A) "one pass", where you print 5000 points using the default OFPS, and
   create a profile from the result. You can decide your ink limit and
   black generation in this trial.

B) "two pass", where you print 1000 points using default OFPS, create
   a profile with your chosen ink limit and black generation, and then
   use that profile with targen and body centred cubic to create a second
   test chart, merging the resulting readings before generating the final

Compare the two approaches and see if one is better than the other.

Graeme Gill.

Other related posts: