Path: utzoo!utgpu!jarvis.csri.toronto.edu!cs.utexas.edu!uwm.edu!ux1.cso.uiuc.edu!ux1.cso.uiuc.edu!m.cs.uiuc.edu!shirley
From: shirley@m.cs.uiuc.edu
Newsgroups: comp.graphics
Subject: Re: photographing screens
Message-ID: <4400053@m.cs.uiuc.edu>
Date: 28 Nov 89 01:38:39 GMT
References: <3777@hydra.gatech.EDU>
Lines: 33
Nf-ID: #R:hydra.gatech.EDU:3777:m.cs.uiuc.edu:4400053:000:1536
Nf-From: m.cs.uiuc.edu!shirley    Nov 27 10:39:00 1989




We use 100 speed film for both prints and slides.  We want a slow
high quality film, and this is the slowest that we can get processed
quickly.  We want a slow shudder speed to avoid electron gun visiblity,
but not too slow to avoid nonlinearity problems in the film.  For our
monitor we usually use an f8 apature for about half a second.

We get as far from the screen as we can to avoid distortions near the
corners of the image.  If you have a zoom lens it helps.

We usually use slides because it seems to give the lab less of a 
chance to "correct" (screw up) the colors.  Taking some shots
of color bars at the beginning of the roll seems to help.

Like monitors, films have their own gamma.  We've found that with 
Ektachrome 100 the whites are white, the blacks are black, and the
greys are too dark.  Our monitor's gamma is around 2.3.  We've found that
using a gamma of about 3.0-3.2 we get better results with Ektachrome.
The image looks crummy on the screen (washed out) but the photos look
much better.

You may find that even with a 24-bit monitor you see contouring.  This
can be especially true in photos.  To avoid this we often dither in
the 24 bit image (I think Roy Hall's book suggests this).  When you've
found an intensity between 0 and 1 and are about  to output this as a 
byte, use "byte = floor (254.99999 * I + random_from_0_to_1)".  If you
don't have the ability to gamma correct your monitor, then you might 
want to first raise I to an exponent: I = I**(1.0/gamma).

peter shirley
shirley@m.cs.uiuc.edu


Brought to you by Super Global Mega Corp .com