[vorbis-dev] lossless ogg encoding with good compression rate

Graham Mitchell graham at grahammitchell.com
Thu Jun 26 14:48:08 PDT 2003



> The way I see it, the combination of ogg and diff file can never be encoded
> as easily (or as compact) as the original wave file itself.

I concur.  I've been playing around all afternoon, and I can't find a method 
diff'ing a decompressed OGG with the original WAV in such a way that the 1) 
original WAV is completely recoverable, and 2) the OGG + diff is any smaller 
than a FLAC-compressed version of the original file.

Using a byte-for-byte XOR I get okay results, and slightly better results with 
a byte-for-byte difference, but neither approach even equals FLAC, much less 
beating it.

The Ogg + diff approach does beat storing the uncompressed WAV, and is 
lossless, but only on my machine.  And since oggdec isn't guaranteed to 
produce identical WAVs on different architectures, I'm guessing it's not even 
a workable solution in general.

There is a binary diff utility that proposes to work like the "diff/patch pair 
does with text files", but I can't get it to finish running, even after 
letting it churn on smallish files (16MB or so WAVs):
        g_pochini at virgilio.it/">http://space.virgilio.it/g_pochini@virgilio.it/</a>

Maybe if someone can get this to complete, it might provide better 
compression.


-- 
Graham Mitchell - computer science teacher, Leander High School
Cats are Spiderman,
Cats are Spiderman,                                                   ^ ^
Cats are Spiderman,                                                   o o
NO-body KNOWS who they ARE!                                          ==Y==

--- >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.




More information about the Vorbis-dev mailing list