[Top] [All Lists]

Re: Making MP3 files

Subject: Re: Making MP3 files
From: "evertveldhuis" <>
Date: Sat, 14 Jun 2003 16:07:25 -0000

It all depends on the source and the codec used.
sample rate: If your source contains hardly any signal above 10 kHz
then a sample rate of 22 kHz is fine, otherwise use 44,1 KHz or
higher (try not to resample)

bitrate; this is where most difference is heard.
bitrate goes together with the codes though; MPegSuite is great but
only if you use 256 kbps or higher. Use other coders for lower
bitrates. I can't give advise on which codec ; the opinions about
them vary a lot, and info about them is often too old/not valid
anymore. (there are even people saying MP3 isn't the best anymore,
but MP3's can be played at most computers, it is a double edged sword)

most important and most overlooked setting: true stereo versus joint
Joint stereo is a pseudo stereo; the low freqs are mono, mids and
high is true stereo. I have heard a recording of the flushing of a
toilet that was moving from left to right; but at the end of the
movement some low parts were audible at the left (where they should
not be) while at the same moment other parts were only audible at the
opposite speaker ; this is very unnatural!
(It was my own recording, that is why I am so convinced.)

I would strongly advice using true stereo all the time. Unfortunately
true stereo can only be used at higher bitrates.

Maybe you should make multiple MP3's with different settings to hear
for yourself which setting is good at 'a' recording ; each recording
might need other settings.
Or you can just 'make' your own standard and always use for example:
true stereo, 44,1 kHz sample rate and 256 kbps ; this is VERY close
to CD quality.

Regards, Evert


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

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