Digital BW, The Print

----- Original Message -----
From: "Austin Franklin" <darkroom@...>
To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
Sent: Sunday, March 24, 2002 9:48 PM
Subject: RE: [Digital BW] Dynamic Range


> Hi Martin,
>
> > > What definition says what?
> >
> > Dynamic range: the ratio of the strongest to the weakest sound intensity
> > that can be transmitted or reproduced by an audio or broadcasting
system.
>
> Well, it's not really a good definition, as it leaves the definition of
> "weakest" and "strongest" ambiguous.  Which is at the heart of the
> difference we have!
>
> I am only trying to tell you where your misunderstanding here is, so
please
> bear with me.  The problem you are having here is you are confused by the
> ambiguous definitions of "weakest" and "strongest".  Here is a better
worded
> definition of dynamic range:
>
> Dynamic Range(dB) = log10(largest amplitude / smallest DISCERNABLE
signal).

Okay but taking the log does not change the fundamental ratio.
>
> Note the term "discernable".  Not lowest, not weakest...but discernable.
> Discernable means perceptible.  One can only perceive a signal that is
> greater than noise.

Granted, but the difference between "discernable" and "weakest" doesn't seem
to be discernable in this context. <G>
>
> "Discernable" is the exact same as noise...it is the +/- with which the
> signal can be discerned...lowest, weakest and smallest are ambiguous
terms.

That was what I was trying to say earlier but did not state it well.
>
> Also, you are taking strongest to be the top number of the range.  That is
> not what is meant by strongest.  A better term is largest amplitude.

That's fine but I don't see the distinction between "strongest" and "largest
amplitude" except that amplitude places us in the realm of wave forms which
may not be appropriate to discussing print density.

>  Here
> is an example:
>
> You have a signal that can range from 1 to 10.  The largest amplitude of
the
> signal is 9, NOT 10.

Yes the largest possible step change and the range, 10 minus 1.

>  Now, let's say your noise is .1 units.  That makes
> your smallest discernable signal .1.  The dynamic range of this is
> log10(9/.1) or ~1.9.
>
> Now, say you have the same range from 1 to 10, but your noise is .5...
Note
> you have the same range, 1 to 10, but only have the ability to measure to
> within .5 units.  That gives you a smaller dynamic range, or log10(9/.5)
or
> ~1.6.

I understand this in the sense of electronics, sound, digital/analog systems
where there is input and output that change with time but I do not see how
to apply it to a static print. In a print the largest step change is from
paper white to maximum black which also happens to be the range. What are
you going to use as a value for the noise?

>
> Did that make any headway?  My 15 minutes of lucidity is over, it's been
two
> hours ;-)

Yes except as noted above. I do want to point out that we started with
differing opinions on whether inkjet prints had more tones than a silver or
traditional print. Which then lead to a difference over the definition of
analog and the existence of infinite tones between any two tones on a
continuous tone curve.

I am not sure that the dynamic range is a part of this issue although we
seem close to clarifying our differences on that subject.

Martin

(I hope Bill does see this, he'll feel bad we didn't post it on his new
list.)