P3D Re: color temperatures

[roberts@xxxxxxxxxxxxxxxxx (John W Roberts)]

>Date: Fri, 30 Apr 1999 13:40:07 -0600
>From: "Greg Wageman" <gjw@xxxxxxxxxx>
>Subject: P3D Re: color temperatures

>>From: Rod Sage <rsage@xxxxxxxxxxxxxxxxxxxx>

>>All this talk of color temps brings up a question. Don't our brains
>>automatically adjust white balance so that we aren't so aware of the
>>sickening green of fluorescent lighting or the orange cast of tungsten
>>lamps etc.

>Yes.  However, there are limits to what it can do.  Go look at various
>familiar objects under a high-pressure sodium light, for example.  Some
>colors will look black, because the sodium light simply doesn't emit at
>some frequencies, and there's nothing your brain can do to compensate
>for that.

It will be even more pronounced under a low-pressure sodium light (which
we have here in the parking lots).

>>Why do camcorders etc.
>>need automatic white balance. If they record the images in the same
>>light we see them, wouldn't we be seeing them correctly? Do films record
>>light more accurately than we are able to perceive? Or are films limited
>>in their accuracy?

>As I understand it, the problem is magnified with film because the
>different emulsion layers have different sensitivities.  When you use a
>daylight-balanced film under fluorescent light, which is lacking in
>reds, the already less-sensitive red layer gets even less exposed.  The
>resulting slide is highly deficient in reds.  You can't view what isn't
>captured.

Another way of looking at it is that films don't "see" the same as humans -
the color balance they produce is based on certain assumptions about
subject and ambient conditions. If these assumptions are violated, the
reproduction of colors may be less faithful than desired.

>Also, I think for the brain to do the color correction, something close
>to your entire field of view would need to be taken up by the image in
>question, which is rarely, if ever, the case with a television or color
>slide image.  

I also think that may be a significant factor.

John R


------------------------------