naturerecordists
[Top] [All Lists]

Re: Re: Cornell and rec formats

Subject: Re: Re: Cornell and rec formats
From: Walter Knapp <>
Date: Fri, 05 Jul 2002 13:25:38 -0400
Doug Von Gausig wrote:
>
> At 06:51 PM 7/4/2002, Walt wrote:
> > wrote:
> >I'm not so sure they are recording to .wav files, I could not find that
> >confirmed. How variable a sample rate will .wav files handle?
>
> You can resample .wav files in CoolEdit down to 100 Hz and still play the=
m
> back, but they do not start playing back (at least on my system) until at
> least 7800 or so samples have been read. A 60-sec. 8-bit file sampled at
> 100 Hz takes up only 5K!

Thanks for the note, I thought .wav should be able to store a arbitrary
sample rate. I checked in my mac programs. Spark XL has presets for it's
sample rate choices ranging from 11k to 192k with no provision for
typing in any other. Peak 2.6 has the ability to type in sample rates,
and I got it to go down to 1 Hz, which could be saved in .aif just fine.
A tiny, but useless file.

Of course this is software conversion, actual sampling hardware is a
different story. I wonder how many computer digital inputs can be set
down to low sampling rates. My mac with the Roland UA-30 could play all
these sample rates. At least down to where my ears gave up. I don't have
any way of testing the inputs, however. They are settable in dialogs at
the standard steps down to 11kHz.

Of course there was no meaningful data left in the sound sample I was
mangling. Even at the 100 Hz sample rate you are just keeping sounds
below 50 Hz. The 2000 Hz that Cornell bioacoustics chooses for their
whale arrays would probably be the low end of useful for natural sounds.

Their logger sounds like a nice design. Most loggers just use a timer
for control, and cannot adjust sample rates and such like. This one has
a microprocessor setup that's programmable for such things. Too bad they
don't sell it, from the look of all their denials on their pages they
must be getting a bunch of requests. I'd be happy if they would just
provide the circuits and programming, though it would be nicer if the
boards were available. Even bare boards would speed building a lot. I
expect the completed units are probably a bit costly for them as they
are probably still hand building them, though the boards look like they
may be contracting them now.

There are some other companies making some pretty nice loggers that you
can buy. I don't remember any that had sample rate control at this
level, however.

Walt



________________________________________________________________________
________________________________________________________________________

<Prev in Thread] Current Thread [Next in Thread>
Admin

The University of NSW School of Computer and Engineering takes no responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the naturerecordists mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU