[Re: [vorbis] RC2 vs. beta4 (test)]

[Henry Rearden]

/> Additionally, it has been stated many times that you /
/> _should_ _never_ _transcode_ _any_ _lossy_ _codec_. /

This isn't transcoding. The point about transcoding, say, MP3->OGG
is that the two codecs use different assumptions (psychoacoustic models,
etc.) in performing the compression -- so transcoding often results in
one codec trying to chase down and reproduce artifacts generated by the
other.

The "lather-rinse-repeat" encoding test described by Atilla is a bit
different: roughly speaking, it simply magnifies the artifacts introduced
by oggenc and makes them easier to identify. Obviously this is a non-
linear process so it isn't perfect, but it's far from useless.

I believe this type of testing has historically been done by other codec
designers as well (including tests of various tunings of the MPEG codecs).

-HJR

--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'vorbis-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.