AW: [vorbis-dev] PlusV algorithm -> CBR

[Segher Boessenkool]

Michael Smith wrote:
> On Thursday 30 January 2003 03:16, Dan Miller wrote:
> 
>>fwiw, there are definitely applications that would use Vorbis but can't
>>until there is a _strict_ CBR mode.  Meaning, architecturally the datarate
>>simply cannot exceed X no matter what happens to the quality.  Send silence
>>if you have to, but limit the datarate.
> 
> 
> Dan,
> 
> In situations like this, how is bitrate measured? When you say "cannot exceed 
> X no matter what", what time-period is this over? Since vorbis is a 
> block-based codec (as are many others, including mp3), it's essential to 
> specify some time window - even if that's just a single block (but note that 
> since there's more than one block size, even that can be problematic).

There's also the page header overhead, which has to be counted
to the first packet fragment in the page; this overhead is
non-trivial at low bitrates.

> Nobody seems to want to say what they actually MEAN by "CBR".

*All* codecs are block based -- with "block" being "bit".
There Is No Such Thing as "instantaneous bitrate", with no
codec whatsoever.

Hey, we agree -- something is wrong here :)

<p>Segher

<p>--- >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-dev-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.