Each pixel in a CMOS sensor has its own amplifier, while on a CCD, pixels are read in serial fashion, one at a time through the amplifier (though some have multiple amplifiers, the general idea still holds). Given that there is only one amplifier you can "bin" the signals from adjacent pixels so that the signal is larger relative to the "read noise" in the amplifier (which is fixed). This will result in more dynamic range provided the signal is not too large for any of the "buckets" where the electrons are accumulated.
Of course you get into all sorts of trouble if you are binning from pixels with different colour filters applied to them (as the Bayer mask does above a "one-shot colour" sensor), so people with colour CCDs are less likely to using binning I expect.
Hope that helps answer your question,
Cam
Quote:
Originally Posted by cometcatcher
If only Sony made a low light APS-C version of the A7s...
Why can't DSLR's "bin" or combine pixels to get better low light performance? Too much junk electronics around the pixels? I see no improvement in noise or sensitivity when I use a lower megapixel count.
|