Apache-Talk @lexa.ru 

Inet-Admins @info.east.ru 

Filmscanners @halftone.co.uk 

Security-alerts @yandex-team.ru 

nginx-ru @sysoev.ru 




      :: Filmscanners
Filmscanners mailing list archive (filmscanners@halftone.co.uk)

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[filmscanners] RE: Dynamic range


> > Their equation for dynamic range is precisely what I've used...
> I think I should just requote myself until I understand what you don't get
> from me. Okay, I may embellish it a bit.... I'm saying:
> Note, there are at least two critical differences between The Higgins
> diagram with Austin's terms, and the Analog Devices white paper figure 5.
> 1). What Higgins calls "Largest", Austin defines as, "the maximum signal
> level minus the minimum signal level, and is the largest range or absolute
> range that the signal can go from <=> to"...
> This RANGE is what Austin uses as the numerator in his DyR equation...
> However, In figure 5 this RANGE is called: Dynamic RANGE - The difference
> between the loudest and quietest representable signal level, or
> if noise is
> present, the difference between the loudest (maximum level) signal to the
> noise floor.

One diagram is in log (the paper), and one is in non-log (Higgins).  You can
not compare the two.

> Let me repeat, this paper says DyR is: if noise is present, the difference
> between the loudest (maximum level) signal to the noise floor.
> This is in contrast to Austin who says DyR is: (maximum signal level -
> minimum signal level) / noise)

They are exactly the same equations, as I've shown countless times...one is
in log and one is non-log.  Subtraction using log numbers is EXACTLY the
same as division using non-log numbers.

> The key is not that one is expressed as a difference, the key is that what
> Austin uses for his numerator (or Dmin) is the range between
> maximum signal
> level and minimum signal level, whereas the Analog Devices paper
> just uses:
> loudest (maximum level) signal.

Again, they are both the same thing.

> Austin takes the range between max signal and noise, what they consider to
> be the DyR, and makes it the numerator of his ratio, with noise as the
> denominator.

Again, you are reading two diagrams that represent two completely different
things, one is in dB and one in voltage.  It just so happens, that they end
up being the exact same thing, if you understand what they represent.

> This gets to the issue of whether DyR defines a range or a resolution. I'm
> contending that the Analog devices paper shows it to the a RANGE, the
> useable range of the device, the range between noise and clipping. Not the
> range between noise and clipping relative to noise.

Well, it simply shows the area over which the dynamic range is calculated,
that does not make it a range...and it shows it in dB.

> 2). Further, Austin has claimed that if Higgins had meant for the
> numerator
> to simply be what Austin calls "maximum signal level" he would not have
> illustrated the concept of "Largest" with a double sided arrow which spans
> the range between noise and clipping (what fig 5 calls DyR).
> However, in figure 5 we clearly see them point specifically to the highest
> signal (Peak Level) as a single point where anything greater
> would be in the
> "distortion region". In fact they define it as the clipping point. This is
> the value they use as their numerator.

Which is fine, because one is in log and the other in non-log...  Not a
single thing in the Analog Devices paper contradicts what I've said, it's
just that you are confusing log diagrams/arithmetic with non-log

> Remember, their definition of DyR is:
> Dynamic Range = (Peak Level) - (Noise Floor)
> Austin-s definition is:
> ((maximum signal level - minimum signal level) - noise)
> The difference again is that Austin claims DyR to be (essentially) the
> useable range of the device (between clipping and noise) relative
> to noise,
> while Analog Devices claims it to be the the useable range of the device:
> the range between clipping and noise.

Here, simple example:

Max signal in non-log notation 10000, in log notation 40dB, minimum
discernable signal in non-log 100 in log notation, 20dB.

DR (dB) = 10log10(10000/100) = 100
DR = 40dB - 20dB = 20dB

Well, 100 is the same as 20dB.  EXACT SAME THING.

> So, in summary, I believe this paper shows that dynamic range is a range,
> the range between the noise floor and clipping - which is in opposition to
> Austin's premise that DyR is a resolution.

Come on, Todd, the paper CLEARLY says dynamic range is a resolution.  Why on
earth do they say so many times that you need so many bits to represent a
particular dynamic range?  Forget the diagrams, you are confused by them, as
they clearly represent two different things.


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


Copyright © Lexa Software, 1996-2009.