[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Green Light



At the risk of belaboring a point ...

Jim Mendrala writes:

<< Negative film on a telecine has already captured the broad spectrum of
light on the negative by recording the image in its yellow, cyan, and
magenta images. there are only three color dyes used for this image Yellow,
Cyan and Magenta (YCM). [...] >>

True, though if one looks at the response charts for any color negative film,
the spectral responses of Y,C, and M are rather broad, relatively speaking.

<< Film is subtractive color and TV is additive color. Therefore you only need
a vary narrow bandpass for the red light, the green light and the blue light
to pass through the film and be detected >>

I don't see how it follows that in going from subtractive to additive color,
only narrow bandpass is necessary.  The bandpass of filters in front of
telecine sensors should bear some relationship to the color space of the
television system, which does have some bandwidth although it is smaller than
that of most color films.

<< By using narrow band red, green and blue light the orange mask, which is
primarily for contact printing on film with white light, is not even a factor
because the telecine is adjusted for equal RGB output from the three sensors.
In other words the orange mask is transparent on an RGB light source. >>

What I was trying to point out is that with a predominantly green or blue
light source (as in a telecine), the orange mask presents a considerable
amount of attenuation which must be made up by the intensity of the light
source, sensors, and electronic amplification.  Negative is not fully
transparent to any light source.  In telecine work, we invert the mask to its
video "positive" color which is a murky mix of green and blue, then calibrate
the electronics at the telecine and color corrector to get rid of any residual
color. 

<< A telecine with an adjustable RGB light source can give outstanding results
because you can optimize the telecine on the worst channel for best signal to
noise and reduce the lights on the other two channels for a balance. This
means you don't have to turn up the gain (and noise) to acheive a balance.  >>

This is all very true, but it begs the issue.  If you have sufficient light to
begin with and sensitive enough image sensors, it's hard to rationalize an
adjustable RGB lamphouse on a telecine.  Why go through all that complication
if you could just obtain a proper balance on any film electronically, and
noise did not become a problem?  I'm asking this hypothetically, I'm not
trying to say any previous telecine designs have reached this ideal yet!

<< The Peterson/Bell & Howell RGB lamp house used on most optical printers is
a
classical example of this. This technique I have used many of times when doing
matte shots from TV cameras and film matte transfers. >>

I am under the impression that the B&H model "C" printer which appeared in
1955 was the first to have an additive RGB lamphouse, and even the newest BHP
Modulars use the same basic system, albeit with electronic light valves.  I've
spent many happy (?) hours maintaining both B&H/P and Peterson printers of
various types and am familiar with their optical systems.  I'd just point out
that these electromechanical RGB systems are a necessity in color film
printers built to change light values automatically during a print run; they
have no other means of controlling the light used to expose the negative.  In
telecine work, adjustable RGB light sources are an intriguing concept and one
that might have a tremendous amount of merit in conjunction with certain types
of image sensors, but not essential for producing equally good--or perhaps
better--results in other telecine designs.

Best regards,
Christopher Bacon

(Not affiliated with any equipment manufacturers or vendors.)

---
Thanks to Peter Stansfield for supporting the TIG in 1998..
No product marketing allowed on the main TIG.  Contact rob at alegria.com
956 subscribers in 36 countries on Mon Mar 23 08:37:55 PST 1998 
complete information on the TIG website http://www.alegria.com/tig3/