[vorbis] channel coupling in rc2

John Morton jwm at plain.co.nz
Wed Sep 5 07:09:34 PDT 2001

On Wednesday 05 September 2001 23:59, you wrote:

>  Yes it's probably an ATH/masking issue - but I think that's the way it
>  should work.

I recall their was some discussion on r3mix.net about a study that 
demonstrated that most people can't distinguish samples with and without an
18kHz lowpass filter applied. They use a 19.5kHz lowpass with lame, but 
obviously an observation like that belongs in the ATH profile.

> So using the highest bitrates (256+) the encoder should
>  encode up to 19.5kHz-20kHz if there is really audible high frequency
>  content there (MPC works in the same way, right?). IMO the encoder
>  shouldn't encode past 20kHz under any circumstance since CD players
>  also cut-off at 20kHz.

What about encoding DAT? DVD-Audio? Your own 48kHz sample rate creations? If 
there isn't any frequencies above some point, the encoder won't encode them. 
If you can't hear them, the ATH profile should ensure that they're thrown 
out. Otherwise it doesn't seem wise to make blanket assumptions about what 
the encoder shouldn't encode just because some presently common format doesn't
use those frequences. 

But maybe you know that and where making some other point :-)


--- >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.

More information about the Vorbis mailing list