Re: Information about errors in estimated times for ogg files.

Robert Hart wrote:
> Hi guys, just to add a bit of information about the "length of ogg
> files" debate.
> 
> I have scanned through my vorbis collection using ogginfo, and computed
> the ratio of nominal bitrate to actual average bitrate. Based on 788
> files, encoded at various nominal bitrates (mostly 128k, but some 112
> and some 160)
> 
> max	1.171
> min	0.624
> stdev	0.0696
> mean	0.941
> 
> So it looks like the majority of tracks are actually shorter than the
> nominal bitrate. For a typical 4min track, this would be less than 30
> seconds. 
> 
> Rob

yea, the problem is in how the ogg decoder rounds it's average bitrate 
to some whole number that you'd typically find a cbr mp3 at.  I dont 
know why oggenc does something stupid like that but whatever, we'll be 
using it ...at least for first-time/non-scanned files.




-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en