ScottGee1 wrote:
>
> Following that line of logic, packing more pixel sites onto the same
> size sensor *should* result in lower quality and more noise even
> though the resolution is increased. But apparently that ain't true.
>
> I gotta lot to learn . . .
-----------------------------------
That is true in some cases and would be true in all cases except that
technology marches on. A physically smaller pixel has less area with
which to capture photons and the fewer the photons the lesser the ratio
between signal (photons) and noise. Noise is generated within the
sensor's electronic circuits and is somewhat analagous to the sound
distortion of your 1960's audio amp when you cranked the power way up.
Cranking up the ISO is done by amplifying the signal and, after awhile,
the noise and signal get to be indistinguisable and red, green and blue
pixels start turning on randomly.
I'm not sure if any sensor manufacturer has actually figured out how to
increase the quantum efficiency of a pixel but they have certainly made
advances in producing quieter circuitry and employing various tricks to
differentiate signal and noise and subtract the noise. And sometimes
they're just guessing what's noise and what's signal and end up
subtracting the wrong thing which leads to loss of subtle detail in the
image. Check, for example, this Canon page on sensor technology.
<http://web.canon.jp/Imaging/cmos/technology-e/index.html>
Chuck Norcutt
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|