[Top] [All Lists]

Re: [OM] Colour temperature?

Subject: Re: [OM] Colour temperature?
From: Christopher Biggs <chris@xxxxxxxxxxxxxx>
Date: 31 Jul 1998 08:59:34 +1000
Siri Bjoner <sbjoner@xxxxxxxxx> moved upon the face of the 'Net and spake 

> So; can anyone recommend any literature on this subject? All I have is two
> half-pages from "The Bare Bones camera course for Film and Video" ... not
> much use for still pics....

May I suggest "The Ilford Manual of Photography".   

Later retitled "The Manual of Photography". 

It's often used as a textbook in photography degrees.

First published in the lat 19th Century, and revised and republished
about a billion times since.

I have two copies ('72 and '77 edition).  If anyone wants the '72, they
can have it (next time you're in Brisbane :-).

A basic bit of reading on Max Planck's discoveries in physics and the
idea of a "black-body radiator" would give you the background to
understanding what "colour temperature" really means.

I'll have a go.

Any object warmer than it's surroundings radiates energy (heat).

Consider a "perfect radiator", an object that absorbs all energy that
hits it, and can radiates it away.  This object will be pure black,
hence the term "Black body" (a coloured object obviously reflects some 
incoming light!).

A blackbody radiates energy (light) in (almost) all colours

At one time the distribution of this energy was not well understood.
Nobody could see why ordinary objects didn't give off at least some
very high-frequency radiation (X rays, gamma rays etc).

Planck theorized that energy was "quantized", that it could only be
given of in discrete "packets" of a certain amount of energy.  These
packets correspond to what we call the photon.   So a rock, for
example, doesn't radiate x-rays because it doesn't have enough energy
to "fill up" an x-ray photon.   

Light can be though of as a wave or a particle.  A photon with more
energy corresponds to a wave with a higher frequency.

Now, the distribution of energy of photons versus number of photons
always looks like this

              ..,             ---> increasing frequency (energy)
            /~   \
          _/     |
       __/       \       /\
     _/           |      |
___,-/             \      | increasing amount of photons (intensity)

That is, it builds up slowly from radio, microwave, infrared, visible
light etc, then drops off sharply after a "peak".

The light from the sun peaks in the green part of the visual spectrum.

An object at a particular temperature will have a "blackbody spectrum"
that always looks like the curve above, the hotter the object, the
more to the "right" the curve would peak.

Another feature of blackbody spectra is that below the peak, the
hotter the object the more energy it gives off at ALL FREQUENCIES (i.e
the curve moves "up" the graph, too).

So an object at 5000K always gives off more infrared, red, yellow and
orange light than an object at 3000K, but the 5000K object gives off
more green and blue (because the 3000k object has peaked below green).

Now given that our *physiological* response to light consists of
receptors in the eye sensitive to only red green and blue, any colour
we can see can be synthesized from red green and blue light.  Any
given "colour temperature" can be represented as proportion of

Thus we arrive at the typical manifestation of colour temperature:
lower temperatures are more reddish, middle temperatures are orange or
yellow, and very high temperatures (>10000k) are "bluish".

That we are able to percieve all these colours as "white" at various
times is a miracle of visual perception, and it's all done in the

This got much longer than I expected, hope it helps,


p.s. I'm not a physicist, but I know some of the same long words as
        physicists do. :-)  (Oh yeah, I married one, too).

 chris@xxxxxxxxxxxxxx, when he visits the Real World, is Christopher J. Biggs
 Stallion Technologies, Australia.   Ph. +61-7-3270-4266 Fax. +61-7-3270-4245
 I dig PGP, MIME and Rush. Send mail with "Subject: sendpgpkey" for my pubkey
 ------------------ Power grows out of the barrel of a GNU ------------------

< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz