White Balance and frequency contradiction??

[Killakoala]

The point -

A star emits light at a specific frequency dependent upon it's temperature.
A cool, dark star emits light that is very much in the red part of the spectrum.
A hot, bright star emits light that is very much in the blue part of the spectrum.

SO therefore, blue is hot and red is cold.

So why is it that we consider blue to be cool and red to be hot? It's the exact opposite of the way it acts in nature.

Is it because blue is the same colour of the sea and red is like fire?

Surely if White Balance is based on 'true light' then red is cool and blue is hot.

Has anyone else thought about this or is it just me and my warped mind?

[stubbsy]

Steve

This confuses me too. If I increase the colour temperature in NC or PSCS the image gets more red (decreasing makes it more blue), but if I change my monitor colour temp from 6500K to 9300K it gets more blue and less red (and vice versa) so in one case increasing colour temp moves from blue to red and the other from red to blue

[phillipb]

Maybe because it's hard to think of temperature as anything other then heat of cold. Color temperature is a different thing to thermometer temperature. Maybe there should be another term used instead of temperature.

[digitor]

The point -

A star emits light at a specific frequency dependent upon it's temperature.
A cool, dark star emits light that is very much in the red part of the spectrum.
A hot, bright star emits light that is very much in the blue part of the spectrum.

SO therefore, blue is hot and red is cold.

So why is it that we consider blue to be cool and red to be hot? It's the exact opposite of the way it acts in nature.

Is it because blue is the same colour of the sea and red is like fire?

Surely if White Balance is based on 'true light' then red is cool and blue is hot.

Has anyone else thought about this or is it just me and my warped mind?

Yes, I'm afraid it is your warped mind...

But seriously folks, I think it is because in our genetic memory, (or something like that) blue light is associated with cold temps, and yellow light with warm ones.

Cheers

[stubbsy]

This is lifted from Peter iNova's excellent D70 ebook:

As things heat up, they glow. First red, then orange, then white hot. Then what? Blue. Like the flame on your gas stove.Glowing objects throw off photons that are evenly distributed in a continuous spectrum—a bell curve of colors—and the peak of that curve is described by the heat it emits. Heat and color are intimately connected, and Irish mathematician Lord Kelvin figured this all out in the 1800s.On the Kelvin temperature scale, 5200K (sometimes written as 5200°K) is the white balance setting for full daylight conditions, 5000K is the white balance temperature of the Sun's direct rays at noon (notice that daylight conditions are higher—meaning bluer—due to the presence of blue sky's contribution to the lighting). Electronic flash tubes usually create 6000K light, cloudy skies typically produce from 6500K to 9000K depending on cloud depth. Your computer monitor usually exhibits a color temperature of about 6500K, but can be as high as 9000K. Deep blue sky itself might reach a reading of 20,000K with its overwhelming preponderance of shorter, hotter, blue photons. Incandescent quartz bulbs put out 3000K while 100 watt household bulbs throw off warm 2900K glows while a 40 watt bulb shines at only 2600K. A candle flame is about 1800K.

From this, my monitor getting bluer makes sense, but increasing the temp making an image less blue & more red doesn't <sigh>

[digitor]

snip snip...
From this, my monitor getting bluer makes sense, but increasing the temp making an image less blue & more red doesn't <sigh>

If you think that the temperature is of the light that's illuminating your scene, it may start to make sense. When you increase the temp, you're saying to the imaging program "this photo was taken in blue light!" so it makes the pic redder to compensate.

Cheers

[stubbsy]

snip snip...
From this, my monitor getting bluer makes sense, but increasing the temp making an image less blue & more red doesn't <sigh>

If you think that the temperature is of the light that's illuminating your scene, it may start to make sense. When you increase the temp, you're saying to the imaging program "this photo was taken in blue light!" so it makes the pic redder to compensate.

That's it digitor. No longer confused. Hope Steve is fine now too

[Matt. K]

Steve
Don't do this to us!

[gstark]

Peter,

If I increase the colour temperature in NC or PSCS the image gets more red (decreasing makes it more blue),

Yes, that would be correct.

You're not actually changing the colour temperature of the image that you shot, but the colour temperature that you apparently recorded it at.

Thus, increasing the CT in NC will reduce the relative CT of the image that was recorded.

but if I change my monitor colour temp from 6500K to 9300K it gets more blue and less red (and vice versa) so in one case increasing colour temp moves from blue to red and the other from red to blue

Changing the monitor's CT is merely changing the way that you view the image. Best to callibrate the monitor and not worry about its CT.

[stubbsy]

Best to callibrate the monitor and not worry about its CT.

Thanks Gary. I think I might have to spend the $$$ and get a Spyder for my monitor and buy a printer profile for my Canon printer & paper

[gstark]

Peter,

Once your monitor is calibrated, you'll be surprised how little more work you'll need to do to get great looking prints, even from your Canon printer. We get excellent output from Leigh's all the time.