From: Fernando Gonz=E1lez-Garc=EDa <>
>
> Hi Graham, i have been reading the last post and i have the following
> question.
> I see you are using MD for recording bat sounds. Is there any problem wit=
h
> the compression? Are you losing some frequencies with the compression?.
> Saludos
I can see where there may be some confusion about bat detectors. I don't
know which type of bat detector Graham is using, but they don't record
the actual bat call. That's way to high a frequency to hear.
A bat detector can be thought of as a interface. Every time a bat call
turns up on the input side it puts out some indication on the output
side that's well down in our easy hearing frequency range. The better
bat detectors the signal put out will have a pattern that resembles the
original call, but is not the call. Only the most expensive high speed
recording systems that then slow the recording to lower the frequency
for audibility would give you the actual original call in full detail.
I'm sure that's not what Graham is using as those really eat disk space
and would require a much larger sampling rate.
So, what's the job of the recorder? To record fairly simple sounds that
are easily audible. Generally a bit above voice frequency, but not all
that much. A very easy job, and why Graham can use the extra long
recording methods of LP MD, which don't give really good audio.
As a matter of record, the compression does not lose frequencies,
except, maybe in the top couple khz of the audible range, which adults
don't really hear anyway. It selectively increases quantization noise by
selectively reducing bit depth. And does so in ways that the
quantization noise won't be noticeable on listening. It does not toss
frequencies. (And, no, I don't care to engage in that religious war.) It
will have no trouble at all perfectly reproducing the sounds produced by
the bat detector.
Walt
________________________________________________________________________
________________________________________________________________________
|