Here's a gross oversimplification (because I don't know the details) the more or less explains what is going in a way that also makes it clear why Pentax chooses not to automatically correct for very red lights:
What happens is the camera takes the picture and looks at the data. If it sees that everything looks pretty reddish, it assumes the light is red in color, and sets the WB according to try to *reduce* the red cast. However, if it tried to remove the red cast, it would be in very big trouble. As far as the camera is concerned, there is no discernible difference between a white rose shot under a red light and a red rose shot under white light. Both would appear mostly red. So if the camera completely eliminated the red every time it saw a mostly-red object, the picture of the red rose and white rose would *both* come out white. not what you want at all.
How do P&S cameras do it? Good question, and I can only speculate. First, though, I would observe note most P&S cameras do not tend to be used in tungsten light without flash. And if the flash fires, that eliminates the red cast right there. So you need to be sure you're comparing apples to apples and use the P&S without flash. My guess is most of the pictures you are thinking of won't look so great any more.
But I'm sure some P&S cameras do try. If you find a P&S camera that does not retain much red cast when shooting under tungsten light even when shooting without flash, try the rose experiment. Or a simplified version, anyhow - a plain red sheet of paper. My guess is the shot will come out not as saturated as it should, particularly if you shoot RAW. If you shoot JPEG, most P&S cameras do some very heavy-handed processing - bumping up saturation and so forth - in a way that might be able to produce passable results. I think most DSLR's are reluctant to anything like that in JPEG processing because they don't want the JPEG results to be that different from the RAW results.
I'd be curious to see some posted examples of P&S cameras that manage to do well in the rose test - producing a very white rose without flash even under red light, but still retaining plenty of color in the red rose shot under white light. BTW, the rose has to fill the frame - no fair having other objects in there that the camera could use to tell the difference.
|