Some more bitrate graphs ...

2002-01-15 16:38:33

(Part 2 of my quest to use Ogg Vorbis to help me learn Python scripting  )

I've written a little tool which takes the output from ogg123, hits it with a rock, and generates nice graphs of bitrate-vs-time for Vorbis audio files. It's nothing complicated, but it's possible that some of you may like to play with it.

I've generated two example graphs - they both involve Vorbis encodes of 'Radiohead - Everything In Its Right Place' from 'Kid A'. Both were generated with a datapoint smoothing value of 10.

http://www.maths.warwick.ac.uk/~jingram/og...tyvsmanaged.png
This demonstrates quite well the differences between quality mode and managed bitrate mode. The track was encoded three times:
1) with -q 5, which averaged 181kpbs
2) with -b 160, which averaged 160.1kpbs
3) with -b 160 -M 160, which averaged 152kpbs.
As you can see, -M enforces the bitrate limit very strictly, -b less so.

http://www.maths.warwick.ac.uk/~jingram/og...yingquality.png
This shows how bitrate increases throughout the track as you increase the quality setting.

As with my previous graphs, there are postscript versions, which will be of higher quality for those of you on OSes that understand them . These, along with the Python script I used to generate the graphs, can be found in the obvious place:
http://www.maths.warwick.ac.uk/~jingram/ogg/bitrate-time/

(Edit: added some explanation for the first graph).

Some more bitrate graphs ...

Reply #1 – 2002-01-15 16:44:49

Looks like managed bitrate could still use a bit of work... but I guess it's hard without 2-pass encoding