Hi Austin,
>It does, you are right...but only to the point of reaching the noise level.
>Once you have resolved down to the noise, resolving any further simply
>doesn't give you any more information.
Here is where I tend to get myself in trouble; but here it goes. As a
practical empirical matter of pragamtic significance, I agree and in my
comments have sort of taken that as a given eventhough logically,
analytically, and theoretically you could extend the descrete segments
beyond the noiseless portions of the density range into the noise. If one
did so, it would be meaningless from an empirical and practical standpoint.
That is my thinking and why I could not visualize noise in the equation (I
suppose one could just define as a matter of definition the density range as
being only the noiseless portion of the density spectrum ignoring that
portion which lies below; but I guess I am reluctant to do this for
analytical reasons of defining a phenomenon like the density range which
often extends beyond the clean, noiseless, visable - taken metaphorically -
elements of that phenomenon). But with respect to my conceptual
understanding as it is, am I incorrect about this?
>Simply put, dynamic range is the number of steps within the density range.
>I believe we have no problem with the definition of density range...right?
That is my generic understanding of what you have been saying, and one with
which I understand and agree. To carry it further, I also understand you to
be saying in general terms that the bit depth (number of bits) has not
bearing on the size of the density range, but it does determine the number
of steps or segments that that density range is divided into (not
necessarily the size of any of those steps or segments). Or to put it
another way, the bit depth describes the structure of the dynamic range and
not the size of the density range. Am I off base or close to what your
understanding is.
-----Original Message-----
From: filmscanners_owner@halftone.co.uk
[mailto:filmscanners_owner@halftone.co.uk]On Behalf Of Austin Franklin
Sent: Monday, June 10, 2002 7:02 AM
To: laurie@advancenet.net
Subject: [filmscanners] RE: Density vs Dynamic range
Hi Laurie,
> >Dynamic range is, in our case, (dMax - dMin) / noise.
> I guess I tend to want to stay away from that definition in part because I
> am not really able to visualize it very well
> Alas, you are beginning to lose me because of my limited technical
> engineering knowledge - I suspect - with your first paragraph.
> With respect
> to the first part of the second paragraph, I understand what you
> are saying
> and I agree, for what it is worth. However, my understanding and
> agreement
> with the second part of this paragraph may turn on the
> terminological use of
> the notion of "accurrately." If I understand what you have been saying, I
> can see that the number of bits has no bearing on "accuracy of
> representation" in the literal corrspondence notion of truth
> sense; but in a
> more figurative sense where the terms "accuracy of
> representation" stand for
> ability to define the density range in terms of finer gradiations or more
> descrete segments, I would suggest that it does increase the ability to
> discern and designate finer differences within the the density range.
It does, you are right...but only to the point of reaching the noise level.
Once you have resolved down to the noise, resolving any further simply
doesn't give you any more information.
> Unfortunately, I think the confusion stems from your desire to use the key
> terms in their very precise technical sense as defined in engineering
> formulas and concepts whereas I am only able to really grasp the general
> theoretical sense of the concepts in more metaphorical meanings.
Well, you DID have it, so I thought...then someone confused the situation by
adding some inaccurate statements...which we are still sorting out.
Simply put, dynamic range is the number of steps within the density range.
I believe we have no problem with the definition of density range...right?
Damn, I have to write this whole issue up and post it on the web some
time...it just keeps coming up...and up...and up... ;-)
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
----------------------------------------------------------------------------------------
Unsubscribe by mail to listserver@halftone.co.uk, with 'unsubscribe
filmscanners'
or 'unsubscribe filmscanners_digest' (as appropriate) in the message title or
body
