[Tig] Do you Do It in the dark?

Martin Euredjian ecinema
Tue Mar 1 21:22:02 GMT 2005

> Grading in the dark should not be an issue. If the calibration has
been done the colorist will perform......

Despite reported success, it is a scientific fact that the human visual
system is subject to many factors that conspire against you when viewing a
monitor in the absence of a reference field.  

I've seen severe examples of this.  One such test consisted of a computer
displaying an image with a program that slowly varied color parameters.  The
rate of change was adjustable.  At slow rates the human observer could not
tell the difference in the absence of a reference field.  Expert observers
(what a color scientist would call a colorist) are not immune to this.  Only
when color-shift extremes were reached did observers start to notice

Of course, this effect is easy to understand.  Put on a pair of green
sunglasses and go outside.  Within a very short period of time your brain
has adjusted for the shift in color and white is still white, red is still
red, etc.

And so, in the context of monitoring, without a reference field, it is quite
possible to either have human-induced color drift or color drift caused by
such things as the monitor drifting.  A reference field, which should be
significantly larger than the surface-area of the monitor --as well as
designed and calibrated in context-- provides a constant reference not only
for color but also for contrast and black level.  Don't forget that, more
often than not, more than one person is looking at images in a grading room.
How are you calibrating your clients' visual system?

SMPTE RP162 is one document that deals with this.  I feel that it is
somewhat outdated.  Despite this, I think this is better than darkness.

I think this is particularly important in another context:  Distributed

What happens when content must be viewed and judged at geographically
distant locations?  LA to London?  Budapest to NYC?  San Rafael to New
Zealand?  etc.  A properly established surround field is of significant
importance here in order to be able to even begin an attempt to discuss
color.  The same could be said about color decisions made on set brought
into post.

Now, from that broader context, I hope that it can be seen that unilaterally
working in a dark environment might not guarantee best results.  It could
actually be dangerous.  While you (plural) may have total trust in your
colorist as the "reference entity" (a myth that I can dispel in about 15
seconds flat --no offense meant), the very ability to communicate color
reliably throughout the production and post-production process must be
standards bases and these standards have to take into account the viewing
environment because we humans do not see as a colorimeter does.  We see in
relation to other factors surrounding the subject.  The print, paint and
fabric industries have known this for a long time.  They have standards and
are very aware of such issues as viewer and instrument metamerism.  It might
be time to take a look at RP162 and bring it up to date so that everyone,
from production to post, can understand what to do before attempting to
either adjust or discuss color.

Interestingly enough, we are going to have a calibrated viewing environment
at the eCinema booth this year.  A room with proper surround lighting.
Being that I've landed right-smack in the middle of trying to make a new
display technology play with the big boys I have to deal with this issue of
viewing environment all the time.  A great many lessons were learned so far.
Some of which I hope to share during NAB.

Martin Euredjian
eCinema Systems, Inc.
voice: 661-305-9320
fax: 661-775-4876
martin at ecinemasys.com
ecinema at ieee.org

More information about the Tig mailing list