Pages: [1]   Go Down

Author Topic: Using i1Profile -- Unexpected results (I suspect my ignorance is showing)  (Read 2864 times)

photographist

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 74
    • Dancing with Light Photography

Good day, all ...

   I've been lurking around the discussions re: i1Profiler, noting the success and frustrations of others and comparing them with my own.  I have an i1 (rev D) and an i1iO table with which I've been successful in the past using iMatch 3.6.n over the last couple of years.    I too had run into the general error of "... unable to create profile..." initially.   That issue seems to have gone away or at least has not reasserted itself recently.   

   I continue to run into questions and issues that are not answered in the XRite docs I've seen thus far and are frankly butting up against the limits of my knowledge.  I'm hoping that those of you out there that are far smarter than I in this arena can lend some insight and suggestions.   


My current issues and questions are as follows:
   1) I don't truly understand the differences in application of V2 v. V4 of the ICC.  It appears that I get 'darker' resulting images using profiles generated with i`Profiler.   In terms of apparent light interpretation to darker, it is:  1) iMatch ICC, 2) i`Profiler ICC v.2 and 3) i`Profiler ICC v.4.   
 
   2) When selecting the Profile Settings (Perceptual): - Should I select Custom, Colorful, or Saturation?  I'm used to creating various "Logo" varieties in iMatch, but am not sure what or if there really is a default here.

My workflow so far:
  • Create Test Chart,  9x9mm targets, 1780 and 4096 charts, non-scrambled
  • Print via the Canon Photoshop plug-in with no color management
  • Let set for min 24 hrs
  • Using i1 and i1iO table, measure the results using i1Profiler
    o Have used standard strip mode
    o Have used spot x2 sample mode
    o Have NOT yet tried averaging (using same read method for multiple reads)
  • Lighting setting left at 50K (same as when using iMatch).
  • Profile Settings:
    o Perceptual: used Custom, Colorful and Saturation( at different points in time to
        see what the results would be -- jury is still out on these)
    o Tables: Optimize Quality
    o Advanced:  Default values, except ICC Profile version (Have tried bothV2 and V4)

Thank you in advance for your comments and time!
Jeffrey 

Additional technical details:
   - i1 (rev D)
   - i1iO Table (a year old)
  - Windows 7, 64 bit
  - Intel i7 260, 18 GB RAM, 5+ TB of disk space
  - Cannon ipf-6100 Printer (using Adobe CS5 Canon Print plug-in)
  - Epson 9880 Printer (no RIP)
  - calibrated and color managed hardware and workflow
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20650
  • Andrew Rodney
    • http://www.digitaldog.net/

Stick with V2 profiles, V4 bring nothing to the party but issues with some 3rd party applications. V4 profiles in i1P are basically V2 profiles in disguise anyway (they don’t fully support the V4 spec and usage).

In terms of Custom, colorful, saturation, that’s really up to you and based on the output device and the perceptual mapping. You could build a profile each way and view the soft proof of each (with only the perceptual table) and see if you have a preference. Its a bit like asking, should I use Velvia, Ektachrome or Agfachrome E6 films? They all have differing renderings. Its subjective.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

photographist

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 74
    • Dancing with Light Photography

Hi Andy ...

  Thank you for your quick responce and your comments.  I've noticed something else in examining the output of i1Profiler during the chart creation stage.  Frankly, I'm not sure if this makes a difference or not but here's what I've noticed:

1. i1Proflier output (when saving to file)
  a. resolution 101.6 pix/inch
  b. Mode:  8 bit

2. When using Bill A.'s test charts
  a. resolution: 203.3
  b. Mode:  16 bit

In your or anyones opinion, is testing from an 8 bit image make any difference? 

Thank you!
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20650
  • Andrew Rodney
    • http://www.digitaldog.net/

The bit depth of the target isn’t really an issue. Be nice if the product provided the option for 16-bit but its not anything that’s going to stop you from getting good results with the current implementation.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

photographist

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 74
    • Dancing with Light Photography

Thank you again very much for your answer.

Do you (or does anyone else) know what effect the MeasurementTolerance= line in the XRi1G2WorkflowSetting.ini file manages? 
 - If I lower the value does it decrease the tolerence of each measurement or
   increase it? 
  - Is there any practicle effect on final profile's quality?

Thanks!

Jeffrey
Logged
Pages: [1]   Go Up