I spent some time on the archives, but my search terms may not have been
very good. What I'm looking for is if anyone has ever determined the
smallest number of patches for a pre-conditioning profile that works
properly on modern inkjet printers.
I understand that you can use a very small set and get some benefits, and
use a huge set and maybe get a benefit, but looking for something in the
middle ground.
As I understand the process (and may be wrong), you make a profile from a
set of patches and feed this back into the target generation tool. Targen
then looks at the out of gamut areas and only creates patches that fit
within gamut and therefor allow a better profile because you aren't wasting
patches on out of gamut areas. It probably also linearizes the output
characteristics so that patch distribution is well placed. Probably more
functions but this is how I'm understanding it.
So back to the original question, is there a patch count that will
accurately map (or decently estimate) the out of gamut areas that will fit
on a small sheet of media? Trying not to waste ink and paper when possible.
And after that, assuming a well behaved printer, what's a good number of
patches that should be created for the second run?
Back when I was using Profiler Platinum, decent profiles were made from
around 900 patches, with better profiles made from around 1500 patches (all
RGB). If I spend 500 patches on the pre-condition step, will 1000 be good
for the second round? Or is that too few or many patches?
If it matters, right now my concern is a Canon Pixma Pro-100s (dye ink with
gloss/luster papers). Using either a Pulse UV (if I can get it working
again) or an i1Pro rev.A (both UV and non-UV versions available). I will
probably want to print some B/W or selective color images once in a while,
so I'll add a bunch of gray (grey) patches in the mix.
Thanks, Greg
