Pages: [1]   Go Down

Author Topic: Experiment with print profile  (Read 2108 times)

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Experiment with print profile
« on: April 05, 2010, 06:03:05 am »

Hi,

I 'm looking into the question of consistency, a way of checking printer profiles regarding the ongoing discussion about Snow Leopard color handling.

The idea I had was to print a synthetic color checker and compare visually with the original color checker, also to measure each field and compare to older data.

Recently I observed that colors (a and b values) are pretty close but lumonisity (L) is pretty much different between print and reference. Am I doing something wrong?

The enclosed screendump shows correct values for each field in red and measurement in blue.

[attachment=21283:Screen_s...56.03_AM.png]

The way I made this was that I downloaded a synthetic CC from Bruce Lindblom's site. This one is in Lab-space and correctly tagged, I think. The synthetic CC was imported into Lightroom 2.6 and printed using a profile created with my Color Munki using relative colorimetric intent.

So it seems that my print is darker than intended.

Any explanation would be much appreciated.

Best regards
Erik
« Last Edit: April 05, 2010, 06:17:51 am by ErikKaffehr »
Logged
Erik Kaffehr
 

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Experiment with print profile
« Reply #1 on: April 05, 2010, 09:27:50 am »

A Lab file in Lightroom? I wouldn’t be using Lightroom for this for one, I’d use Photoshop, apply the profile with Convert to Profile and use the No Color Management settings (which isn’t available in LR), and measure. I’d send out a target (something like one used to build a profile) or roll your own, doable in the free ColorPort app from X-Rite which you can also use to measure the target. Then I’d feed the CGATs file into something like ColorThink to provide deltaE reports of the various targets.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Experiment with print profile
« Reply #2 on: April 05, 2010, 10:11:59 am »

Andrew,

Thanks for your info. So I don't know what I'm doing? Fine opportunity to learn!

What I don't really understand is that if we have a correctly tagged LAB image and it is imported to Lightroom it would be converted to Prophoto RGB (or rather the Melissa RGB). Very little should be affected by that conversion. If I export the image from Lightroom using Prphoto RGB and reimport to Photoshop and convert to LAB again I still have the same L values on my gray patches (at least in my test). So I think LR handles Lab correctly on import.

My question is really should not a Lab 20,0,0 patch print as a LAB 20,0,0 patch with correct profiles?

Best regards
Erik



Quote from: digitaldog
A Lab file in Lightroom? I wouldn’t be using Lightroom for this for one, I’d use Photoshop, apply the profile with Convert to Profile and use the No Color Management settings (which isn’t available in LR), and measure. I’d send out a target (something like one used to build a profile) or roll your own, doable in the free ColorPort app from X-Rite which you can also use to measure the target. Then I’d feed the CGATs file into something like ColorThink to provide deltaE reports of the various targets.
Logged
Erik Kaffehr
 

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Experiment with print profile
« Reply #3 on: April 05, 2010, 10:23:42 am »

Quote from: ErikKaffehr
My question is really should not a Lab 20,0,0 patch print as a LAB 20,0,0 patch with correct profiles?

We don’t know what correct is really until we compare the expected value to the measured color through the profile. Now in ColorThink, using the profile, we can get Lab to RGB values but we also need to print and measure that to see what the values result from the printer. That’s why taking a target which has the Lab reference values, printing it, measuring it and popping it into ColorThink would allow one to get the deltaE to the profile report. You have the reference Lab values (expected) and the output Lab values (measured) and you can then plot the two. You can see the deltaE values for each patch in color space, even sorted by best or worst in the ColorThink Color List window.

You’d want to do this with each application or OS or driver in question.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Experiment with print profile
« Reply #4 on: April 05, 2010, 10:46:38 am »

Hi,

Would I need ColorThink Pro or could I do with Color Think 2?


Best regards
Erik



Quote from: digitaldog
We don’t know what correct is really until we compare the expected value to the measured color through the profile. Now in ColorThink, using the profile, we can get Lab to RGB values but we also need to print and measure that to see what the values result from the printer. That’s why taking a target which has the Lab reference values, printing it, measuring it and popping it into ColorThink would allow one to get the deltaE to the profile report. You have the reference Lab values (expected) and the output Lab values (measured) and you can then plot the two. You can see the deltaE values for each patch in color space, even sorted by best or worst in the ColorThink Color List window.

You’d want to do this with each application or OS or driver in question.
« Last Edit: April 05, 2010, 11:08:25 am by ErikKaffehr »
Logged
Erik Kaffehr
 

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Experiment with print profile
« Reply #5 on: April 05, 2010, 12:11:34 pm »

You don’t need pro to do this.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Experiment with print profile
« Reply #6 on: April 05, 2010, 03:15:24 pm »

Thanks for feedback! I have ordered Color Think 2 today. It was on my to do (maybe) list anyway.

Best regards
Erik


Quote from: digitaldog
You don’t need pro to do this.
Logged
Erik Kaffehr
 
Pages: [1]   Go Up