naturerecordists
[Top] [All Lists]

Re: Minidisk data transfer and bat recordings

Subject: Re: Minidisk data transfer and bat recordings
From: "?" <>
Date: Sun, 25 Jan 2004 20:51:41 +0100 (MET)
Wakt wrote :

> > One web site that did concern me was
> > http://www.avisoft-saslab.com/compression/compression.htm
> >
> > but I assume you are ware of this one.
>
> That's Raimond's attempt at analyzing on paper something he's no
> experience with.  He heavily overstates his case. And he uses a
> contrived test with sounds no animal or group of animals would ever
> produce. Testing mp3 compression as if it was ATRAC. Kind of like saying=

> Fords and Chevys are exactly the same, something that might get you shot=

> around these parts.

Sure, such extreme artificial sounds will never occur in animals. But I
think that it is valid to demonstrate the worst case. This will enable you =
to
better understand what's really going on in lossy compression systems. Once=
 you
know the potential artifacts, you would be able to identify them in your ow=
n
recordings.

I made these tests certainly in response to Walt's statements on that
subject. Please note that there is also a test on ATRAC, not only MP3! Actu=
ally, it
turned out that MP3 adds less distortion than ATRAC (at least in these
artificial test signals).

Finally I have also added a comparison on real bird sounds (the test
conducted by Jeremy Minns).

Regards,
Raimund

--
+++ GMX - die erste Adresse f=FCr Mail, Message, More +++
Bis 31.1.: TopMail + Digicam f=FCr nur 29 EUR http://www.gmx.net/topmail



________________________________________________________________________
________________________________________________________________________

<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