[filmscanners] RE: Dynamic range

["Austin Franklin" <darkroom@ix.netcom.com>]

> Yes, I am, because that is where the DyR will be captured. The
> bits are just
> there to represent the DyR digitally, no?

Correct.

What the "dissent" was on others part, is I claimed you NEEDED N bits to
represent a particular dynamic range, (which the fact is, you do), but
"others" claimed I was wrong.

> > But the bit-depth is material to the scanner user, since if it is
> > inadequate, then the quantization error will be larger than the
> noise, and
> > value of the minimum discernable signal rises. If the bit depth is
> > excessive, then the noise becomes the limiting factor, since
> the device will
> > return a distribution of random values for a given density on
> the film. If
> > you start throwing away excessive bits, the range of the distribution of
> > those random values will get smaller and smaller...
>
> I think I agree with this - not sure about the last line, but
> I'll take your
> word for it. HOWEVER, this is where this conversation often bogs
> down and it
> is simply immaterial to what DyR is, and what DyR is is what this
> conversation is really about. I believe the only way to really get at what
> DyR is is to discuss it in the analog domain, (though I'm not
> sure how to do
> it since most references seemed geared to digital applications). After we
> understand DyR then we can look at how it applies to scanners, if we still
> care to.

David made a post about dynamic range a while back that said something to
the effect that every number you "measure", is actually measuring a "range"
of numbers, simply because you measure with the resolution of the minimum
discernable signal...  Example, you are measuring voltage, and your meter
measures to 1V.  When it measures 1V, your actual value could be 1V +- .5V,
or .5V to 1.5V.  That is, why, in my book, dynamic range represents a
resolution.

Now, because I say that it gives me resolution information, does not mean
that it doesn't also give other information, like "size".

Regards,

Austin

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