[Tig] D.I and 6500k

Kevin Shaw kevs
Sun Mar 6 09:16:33 GMT 2005

>But I do think we ought to at least consider Geoff Boyle's point of view.

Sorry but IMHO there are some non sequiturs here.

> Product and systems designers creating DI suites should take note of what
> cinematographers are shooting before setting their own standards.

First, the display medium is the one thing that should be constant, as it is
ultimately the one thing we as a group have least control of.
Second, the white balance used by the DOP is based on the film stock
characteristic, not the final viewing conditions. Correct me if I'm wrong,
here is how I see it...

If the standard for TV is 6500k based on old CRT technology, that?s what we
should  aim for when working on tv programs. If the standard for projection
is 5400 based on old film projector technologies that is what we should aim
at for cinema. Post production is much easier/better if what you see is what
you get. It will be interesting to see if these standards ever change
because of new technologies, but as Dick points out we generally try to fit
new tech into existing conditions.

BTW - do LCD and plasma screens perform better at a different white balance?
Do newer projectors, and indeed digital projectors need to stick to the old
projection standard?

Adapting a project to work on different mediums (even EBU to SMPTE
phosphors) requires color correction to be perfect. When the viewing
conditions cannot be replicated, as in grading for film finish, complex LUTs
should emulate the final conditions. (Although the lab graders have never
worked that way!)

In a similar, but surely unrelated way, the white balance used by the DOP
has to relate to the characteristics of the capture medium. Simply put is it
a daylight or tungsten film stock. In a theoretical "pure" scenario daylight
lighting on daylight stock produces neutral white, which should be
reproduced as a neutral white on the final viewing mediumS. Obviously if the
viewing mediums are different the results will be different unless different
masters are generated.

A classic example of how this manifests itself occurs when grading a neutral
gray for cinema projection. If we simply remove all color, the final on
screen image appears warm rather than neutral because of the processes after
color enhancement. To get a true neutral we have to color balance allowing
for film out emulation LUTs - to my knowledge none of the color system
manufacturers have yet built a control system that calibrates "neutral/ zero
chroma" AFTER monitoring LUTs and many do not apply monitoring LUTs to the
scopes. Of course user defaults can be stored but my point is we are still
in ground breaking mode even though DI getting pretty commonplace

I have the impression Geoff was teasing a bit with his comment ... Reminding
us that the true purpose of color enhancement is to make the DoP look even
better :-). 

Happy Coloring

Kevin Shaw                 consultant colorist
kevs at finalcolor.com          www.finalcolor.com

More information about the Tig mailing list