Re: graphs

Greg Schmitz (gws1@columbia.edu)
Tue, 25 Feb 1997 00:15:45 -0500 (EST)

On Sun, 23 Feb 1997, Judy Seigel wrote:

> I wonder if Greg, or Red, or whoever, would say a few words about why
> computer for this. I've plotted densities to make a graph (or "curve") on
> a piece of paper xeroxed from Kodak's chart. It seemed *to me* to be
> fairly quick, easy and clear. So are there things simple pencil & paper
> doesn't do? Surely it can't be quicker -- can it? (Greg, I think we had
> this discussion before & it's just my memory failing. But I'm probably not
> the only one wondering.)

Yes Judy we have talked about this, but it was a long time ago.

Plotting my tests by computer has saved me countless hours of making
tiny little dots and then connecting them. Briefly, many moons ago
before I owned a desktop computer I developed a system for testing
film which yields a number of graphs including characteristic curves
and curves which plot density/exposure against development time; the
later being handy for determining development times which yield a
specific density range for a given range of exposures. In my
pre-computer days the actual testing took me between 1-2 hours. But
after conducting the test in the darkroom it then took me many more
hours (like 4 or 5) to plot the curves. Generally I had between 6 to
10 twenty-one step plots (thats 126-210 tiny little points) just for
the characteristic curve sets. Now with the aid of a spreadsheet
program to record the data and do the math along with a plotting
program I can now have hard curves and a data print-out within 1/2
hour of the time I've read the densities (it takes that long because I
read the densities into my cassette recorder and then play the tape
back to enter the numbers - sort of a poor man's RS-232).
Additionally, the computer has allowed me to create 3D plots from which
I can take "slices." This has allowed me to use data which varies by
exposure, development time and dilution for any given developer/film
combination that I have tested in depth; very handy!

-greg

>===for PGP Key finger Greg Schmitz <gws1@cunix.cc.columbia.edu>===
:: ::
Key fingerprint = 73 D3 91 15 96 69 74 E5 14 51 71 44 C8 ED 01 11
Key ID: A7026B1D Key Size: 1024 bits Created: 1994/11/10
==================================================================<