[argyllcms] V1.1.0 RC1 RGB Profiling

  • From: Nikolay Pokhilchenko <nikolay_po@xxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sun, 08 Nov 2009 11:44:51 +0300

1. Will this be useful or not, but I've not found any clear way to include 
first linearizion chart *.ti3 data to the next linearized 

chart *.ti3s - the first have direct device data, the last have linearized 
device data. So it's impossible to add just printed channel 

wedges to further main ti3 for colprof.
Another problem: I forgot to input -K *.cal parameter to printtarg and have 
print and read non-linearized chart. How can I join this 

chart with another one but linearized?

2. Quick RGB profiling
Printing and measuring PrintRGB 130-patch wedges (my A5 page have 391 patch 
capacity):

targen -v -d2 -G -e3 -s130 -f0 RGB
printtarg -v -ii1 -T360 RGB
CHARTREAD -v RGB
printcal -v -i RGB
targen -v -d2 -f391 -G -w RGB_OFPS
printtarg -v -ii1 -T360 -K RGB.cal RGB_OFPS
CHARTREAD -v RGB_OFPS
colprof -v RGB_OFPS
targen -v -c"RGB_OFPS.icm" -A0.95 -G -f391 -d2 -w RGB_OFPS_2
printtarg -v -ii1 -T360 -K RGB.cal RGB_OFPS_2
CHARTREAD -v RGB_OFPS_2

profcheck -k -x -w -e RGB_OFPS_2.ti3 RGB_OFPS.icm
Profile check complete, errors(CIEDE2000): max. = 4.197503, avg. = 0.970148, 
RMS = 1.177961

Joining of RGB_OFPS and RGB_OFPS_2 data into RGB_OFPS_782.ti3

colprof -v -qu -iD50 -oshaw -S"sRGB.icc" -P -dpp -A"Hewlett-Packrd" -M"HP 
Photosmart 8253" -D"HP Photosmart 8253 @ Lomond
Matt 20091107_Lin not included" -C"Nikolay Pokhilchenko 2009" RGB_OFPS_782

profcheck -k -x -w -e -iD50 -oshaw RGB_OFPS_782.ti3 RGB_OFPS_782.icm
Profile check complete, errors(CIEDE2000): max. = 1.741373, avg. = 0.321470, 
RMS = 0.389358
See attached archive.

The whole impression: With ArgyllCMS 1.1.0 RC1 I can to profile the difficult 
RGB printer by few of patches. With previous versions I 

had to print about 3000 patches for reasonable profile quality. The one of a 
great improvement I noticed is A-to-B table precision and 

smoothness for both RGB and CMYK modes. The curves are smooth, but the 
tolerance is higher! Now ArgyllCMS have no drawbacks. It's a 

complete and stable. I'll check the black generation difference (or not) 
between xicclu and colprof/collink soon.

I Hope the information about RGB profiling will be useful.


Other related posts:

  • » [argyllcms] V1.1.0 RC1 RGB Profiling - Nikolay Pokhilchenko