Calculating contrast index

Discussion in 'B&W: Film, Paper, Chemistry' started by lamda, Jan 5, 2010.

  1. lamda

    lamda Member

    Messages:
    37
    Joined:
    Apr 1, 2006
    Location:
    Vermont
    Shooter:
    Medium Format
    Now that I finally bought a densitometer, I am interested in calculating the contrast index, so I can see how my negatives compare with Kodak suggestions. (Don't worry, I will use other ways of evaluating them too, such as Adam's The Negative.) I found two formulas in the Kodak publication Monitoring and Troubleshooting KODAK Black-and-White Film Processes.

    More accurate: CI = 0.128 + (0.267 x D-min) - (0.969 x TD) + (0.454 x LD) + (0.183 x HD) + (0.039 x D-max)

    Less accurate: CI = (HD - TD) / 2.26 + 0.10

    These, of course, are meant for use with Kodak process control strips, which I don't have. If I use an ordinary step pattern (21 steps) I figure D-min is the film base and D-max is any of the patches with such a strong exposure they are all the same. So which patches should I use for Toe Density, Low Density, and High Density?

    Or, maybe there is a different formula to use with 21 step patterns.
     
  2. ic-racer

    ic-racer Member

    Messages:
    7,473
    Joined:
    Feb 25, 2007
    Location:
    Midwest USA
    Shooter:
    Multi Format
    Last edited by a moderator: Jan 5, 2010
  3. ic-racer

    ic-racer Member

    Messages:
    7,473
    Joined:
    Feb 25, 2007
    Location:
    Midwest USA
    Shooter:
    Multi Format
    Personally, I do mine on the computer and I use a linear regression to find the slope of the first 11 steps higher than 0.1 on the "Y" axis and use that as my "Personal Contrast Index."

    All the myriad of methods for measuring contrast of the film have problems because some fiilms have a straight portion and some are totally "S" shaped or "J" shaped. Find a method that you like and stick with that, realizing that none are perfect.
     
  4. BradS

    BradS Subscriber

    Messages:
    4,219
    Joined:
    Sep 28, 2004
    Location:
    S.F. Bay Area
    Shooter:
    4x5 Format
    I would think that a plot of the empirical H-D curve would be much more interesting than an estimate of the CI.
     
  5. RalphLambrecht

    RalphLambrecht Subscriber

    Messages:
    8,213
    Joined:
    Sep 19, 2003
    Location:
    Florida
    Shooter:
    Multi Format
    They are both interesting.
     
  6. RalphLambrecht

    RalphLambrecht Subscriber

    Messages:
    8,213
    Joined:
    Sep 19, 2003
    Location:
    Florida
    Shooter:
    Multi Format
    That's a good book. I prefer the average gradient method, but CI is fine. Note that Ilford is using yet another method. To measure the CI yourself, I have attached two files. Print the first to plot your data. Print the second on clear overhead foil and use to measure the CI. make sure to print both at the same scale.
     

    Attached Files:

  7. lamda

    lamda Member

    Messages:
    37
    Joined:
    Apr 1, 2006
    Location:
    Vermont
    Shooter:
    Medium Format
    Thanks, between the two of these I should be able to manage.