I have several portraits from my own CCD Nikons and CMOS Nikons and I see the difference SOOC.
Hi,
As has been explained several times, silicon based sensor arrays are monochrome, and have a roughly panchromatic spectral sensitivity. They are also relatively sensitive to Near Infra-Red light, but that's filtered out.
The color response of silicon based sensor arrays, regardless whether built as CCD or as CMOS device, is determined by the Bayer CFA filter characteristics. Some manufacturers choose to have clearly separated color channels with little overlap, others sacrifice color separation to gain a little sensitivity with les dense and more overlapping color channels.
Then comes the profiling. Now all bets are off in how colors will be rendered, even SOOC. Same file, different Raw converters, different colors. That's not because the sensor suddenly switched form CCD to CMOS, that's ludicrous. Color is created as it is demosaiced and profile converted.
What many do not seem to grasp is that the CCD versus CMOS technologies have other differences that explain why they are predominantly used in certain devices. Traditional CMOS technology is e.g. less suited for large surface sensor arrays where the angle of incident light can vary a lot, e.g. due to Tilt/Shift movements, or traditional lens designs where the exit pupil is relatively close to the sensor array.
That also explains why AA-filters and micro-lenses are rarely used in such camera backs,
not because CCD does not need it, they both do, but it may cause other bigger issues than they solve for a particular use. Of course, technology has not stood still, and we'll see new improved designs as time goes by, but the color rendering it has virtually nothing to do with silicon sensor fabrication technology.
Calling the very technology that allows to make technically better images mumbo jumbo is silly, to say the least.
Cheers,
Bart
