[filmscanners] RE: scanner dmax discussion

["Chuck Cole" <cncole@earthlink.net>]

> -----Original Message-----
> From: filmscanners_owner@halftone.co.uk
> [mailto:filmscanners_owner@halftone.co.uk]On Behalf Of Austin Franklin
> Sent: Wednesday, July 09, 2003 1:19 PM
> To: cncole@earthlink.net
> Subject: [filmscanners] RE: scanner dmax discussion
>
>
>
> No, but since the CCD is basically logarithmic, as are our eyes, it works
> out very nicely.

This is not correct: CCDs are intrinsicly highly linear and not logarithmic
to any subtle extent until ancillary circuitry starts to saturate or become
gain starved.



> > Your point about the minimum step increment being the noise function is
> > correct.  The hard part is that I don't think the noise
> function is linear
> > over the full dynamic range of the sensor so theoretically the step
> > increments aren't linear either.
>
> Actually, they really are quite linear...check out the linear
> sensor specs.


??  Noise is accurately represented as additive light-independent functions
on top of the basic linear transfer of the silicon (et al) detectors.  Only
certain types of detector (eg, avalanche photodiodes) have gain dependent or
signal level dependent noise functions, and none of these are used in
"civilian photography".

I think you're mostly on target, but the bits above on noise are ambiguous
at best.  The "steps" mentioned above have nothing to do with intrinsic
properties of sensor chips and I doubt that anyone tries to size an A/D's
lsb to some rms noise level - except to exceed Shannon sampling criteria
when doing enhancement of low light images. and trying to average out the
frame-to-frame noise.

Is/was there a point to this technical thread I stumbled into?

Chuck


----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe 
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or 
body