Ogg overhead comparison numbers [LWN.net]

Ogg overhead comparison numbers

Posted Apr 8, 2010 1:49 UTC (Thu) by DonDiego (subscriber, #24141)
Parent article: Ogg and the multimedia container format struggle

If you are missing numbers for Ogg container overhead in comparison to other containers, here are some for the famous YouTube comparison conducted by Gregory Maxwell. After extracting the video and audio streams from the containers via 'mplayer -dumpvideo' and 'mplayer -dumpaudio' you arrive at the following numbers:

  17307153 bbb_theora_486kbit.ogv    (the complete file)
- 15009926 bbb_theora_486kbit.theora (the video track)
- 02107404 bbb_theora_486kbit.vorbis (the audio track)
  ========
  00189823 (container overhead)

  17753616 bbb_youtube_h264_499kbit.mp4  (the complete file)
- 13898515 bbb_youtube_h264_499kbit.h264 (the video track)
- 03796188 bbb_youtube_h264_499kbit.aac  (the audio track)
  ========
  00058913 (container overhead)

So in this application, Ogg has more than 300% the overhead of MP4. It's a typical figure.

(Log in to post comments)

Ogg overhead comparison numbers

Posted Apr 8, 2010 2:27 UTC (Thu) by xiphmont (guest, #58693) [Link]

Diego,

We've pointed out previously that your mp4 number is wrong and your conclusion about Ogg overhead is misleading at best. Directing you again to http://lwn.net/Articles/377928/

For illustrative purposes, at that comment you'll also find an Ogg file that we repacked with the same overhead as the mp4. In addition, the most recent release of libogg (drop-in upgrade) reduces default overhead of Ogg-framed video by about 35%. In the grand scheme of things, that means it reduces the overall rate of a bitstream by about .38%. Personally, I think it's almost a misplaced optimization. I'd rather be looking at much larger wins in Vorbis or Theora or CELT. For example, the new libvorbis reduces 5.1 surround bitrates by 10-40%, which is between 25x and 100x more savings.

Ogg overhead comparison numbers

Posted Apr 9, 2010 13:10 UTC (Fri) by marcH (subscriber, #57642) [Link]

> We've pointed out previously that your mp4 number is wrong and your conclusion about Ogg overhead is misleading at best. Directing you again to http://lwn.net/Articles/377928/

There is not even the need to go there.

The numbers posted (and contested) above are:
- 1.0% overhead for Ogg
- 0.3% for MP4

=> who cares?

Ogg overhead comparison numbers

Posted Apr 8, 2010 21:25 UTC (Thu) by kalahann (subscriber, #33885) [Link]

why do we have to suffer the same exact comment from you each time an article about Ogg appears on any website? are you a rantbot in an endless loop?