[vorbis] Ogg too good?

[Ed Sweetman]

Why cant people wait until it's really released before questioning why a
prerelease isn't as fast as lame or some other encoder.  So, oh my the
encoder is slower than flac and lame can run circles around it.  When
it's released and people still aren't happy with it being so slow, the
whole big kicker with this open source thing is that you can read it,
then go and change it and put out your own copy.  See, you can turn that
energy of wanting change to actually making change.  oggenc is an
example encoder.  It uses libraries implementing the ogg codec in a way
that allows it to be used on the most amount of systems.  That's xiph's
purpose in creating an open source codec like this. I'm glad they're not
wasting their time pushing cycles out of an algorithm written in x86 asm
with special instructions from the latest cpu spec sheets because it
shouldn't be their place to and it doesn't serve them any better to do
so.  

After the codec hits the big 1.0 all the x86 asm bufs can go optimize
the hell out of the codec and encoder, fork off their tree and maintain
it just like mp3s are done now. Fragmentation didn't kill off mp3, but
fragmentation makes codec evolution extremely difficult if not
impossible. So at least wait until the codec is complete before calling
on the hand tuning cpu specific optimization of it.

Now that we're sufficiently off of the topic of the original post.  I'd
like to point out that you can specify an abr for the files you want
sufficiently crappy sounding. Nobody is forcing you to use -q  but the
lowest level of -q shouldn't have to sound bad if it doesn't need to
be.  You're sacrificing an easy to understand argument system for a full
range scale and most people wont go for it.  But once again, open source
allows the magical property of being able to be reprogrammed by anyone
so you can rewrite oggenc to do what you want.  Nobody is forcing you to
use oggenc.  Think of the whole xiph package for vorbis and ogg as the
ISO, it doesn't have to be the only implementation of vorbis and ogg.
The specs to the codec are openly documented.   Now lets stop bitching
about one person's oggenc when the source is right there; make your own
version and release it as whatever you'd like.   

<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-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.