[argyllcms] Re: Is it possibile to "improve" a RGB printer profile with more patches from chartread using patches that I already have printed with printtarg?

  • From: Vittorio Villani <vittoriovillani@xxxxxxxxxxx>
  • To: ArgyllCMS List <argyllcms@xxxxxxxxxxxxx>
  • Date: Tue, 16 Nov 2010 01:21:28 +0100

I would like to say thank your for the userfull informations that you gave me 
I am using the profile for a Epson 4880 printer, so I think it should be enough 
consistent. I can tell you that recently I read some patches made in 2009 and 
they had the exact same Lab values of the frist read (did 24+ hrs after the 
print job).
I will try to add new patches using targen.
It would be really interesting to add specific patches to the .ti1 file.

What I was also thinking about is to make a "test print" to see how good the 
profile is.
I have this question (hoping, again, to don't write too much rubbish): is it 
possibile to print a chart set using the profile that I made and read them and 
then use profcheck to see how far the patches are from the right Lab values?

> Date: Mon, 15 Nov 2010 19:25:26 +0000
> From: profiling@xxxxxxxxxxxxxxxxxxxxxxx
> To: argyllcms@xxxxxxxxxxxxx
> Subject: [argyllcms] Re: Is it possibile to "improve" a RGB printer profile 
> with more patches from chartread using patches that I already have printed 
> with printtarg?
> Hi :)
> On 15/11/10 10:17, Vittorio Villani wrote:
> > So what I am asking should sound like "can I add more patches to an
> > already existing .ti1 file, can I add them to the .ti2 file and finally
> > to the .ti3 file?".
> No, but you can merge multiple .ti3 files using a spreadsheet or text 
> editor to merge them into one.  In other words, create a new .ti1 file 
> with your extra patches, follow the usual procedure to read them, then 
> add the readings from the new .ti3 file to the existing one.
> If memory serves, colprof doesn't care about the SAMPLE_ID field, so you 
> can just paste the big block between BEGIN_DATA and END_DATA from one 
> file into the other, and fix up the NUMBER_OF_SETS field to reflect the 
> number of patches in the new, combined .ti3 file.
> Note that you'll probably get a slightly higher fitting error if you do 
> this, especially if your printer's not amazingly consistent between 
> prints (i.e. a laser printer, or to a lesser degree an offset press.) - 
> but that's not necessarily a bad thing - the profile will be doing a 
> better job of averaging out page-to-page consistency errors.
> > I know that I could use targen with the -c parameter to use the profile
> > that I already made to "partially justify" the targets that I am
> > serching for "something more".
> Yup, and doing that when creating the new set of patches is probably a 
> good idea.
> > I hope I didn't write too many stupid things!
> Not stupid at all - I've done this myself when an inkjet profile has 
> performed well overall but had a specific trouble-spot with, say, dark 
> greys.  With care you can edit .ti1 files and add hand-selected patches 
> to cover weak points.
> All the best
> --
> Alastair M. Robinson

Other related posts: