The thread seems to came out a little bad, but i believe the topic is interesting enough, to be worth continuing in a polite and qualitative manner.
Correct me if i'm wrong (and since tr13 being well educated in physics may know this a lot better than me), but as far as i know the main factor determining the colour response characteristics of a sensor itself is the photo active material (if you could say so), which in both - CMOS and CCD sensors currently is Si (and maybe have some additional materials untold by manufacturers).
The spectral response of the Si is following:
You can clearly see that it has the best response from red to near infra-red, but is very poor at blues. This could hint that the cause of bad red's could be dynamic range, which cannot cope with the big differences in each colour component well.
Other sensor related factor of colour reproduction could be its geometry - either there are any nano-scale shapes or other materials interacting with specific wavelengths - currently CMOS sensors have a lot more of "junk" in front of the photo diode, but lately Sony has brought backside illuminated sensors to the consumer market, where all this junk is moved behind the photo diodes, leaving it with even "cleaner view" than CCD sensors.
Improving blue colour response aswell:
Now most of the other factors affecting colour performance are sensor independent, like colour filters used, IR filter, ADC and its colour configuration, lens colour characteristics and
RAW processing (sorry for bad jpeg's).
Images for some additional thinking:
CMOS:
CCD:
Last edited by ytterbium; 11-21-2009 at 04:38 AM.