[vorbis] Vorbis native Ripper-decoder-encoder

[Gregory Maxwell]

On Sat, Aug 04, 2001 at 12:42:46AM -0400, Claudio Bustos wrote:
> Hi:
> 
> I think the problem confuse the motivation of the insdustry, the
> developers, and the end-users.

The industry and developers are irrelevant here. They can easily transcode as
it stands today. They usually understand the trade-offs and can usually
manage to 'do the right thing'.

End users on the other hand, in general, don't think and are not willing to
become informed. If you hand them a stupidly simple transcoding tool, they
will do the wrong thing.
 
> Yes, re-encoding from Mp3 to Ogg "can't improve the quality. It wouldn't
> improve seamless playback. It wouldn't improve the tagging and it wouldn't
> improve the patent situation.". But the only thing of this that a end-user
> take care is the first. And, one of you says, a Ogg file of a small size
> can have a similar quality of the original Mp3. Actually, the quality will
> be poor, but in a $10 speakers noone can tell the difference.

They will take their Xing or Blade encoded 128/160k mp3s (or old lame) and
recode to OGG. Users will download the OGGs and then see that they sound
like crap, of course they do, the original did even before the transcoding
loss (which is very real, but it's easier if we ignore it for the
discussion).

After discussing downloaded mp3 quality with a 'regular joe', a while back:
I came to the conclusion that the perceived poor quality MP3's downloaded
from the Internet came from the presence of all the garbage encoders out
there. Users who are savvy enough to care about quality can comprehend the
bitrate field as an indicator of quality, but they have trouble coping with
a metadataless cause of poor quality (varying encoders) and were dismayed
when their 160kbit/sec downloaded file sounded like it was underwater.

Of course, Vorbis will get better and there will be OGGs out there made by
different coders and even with different codecs, but they should never be of a
lower quality per bit then the 1.0 encoder (the garbage mp3 encoders existed
primarily because of the intellectual property issues surrounding mp3
encoding). 

Whatever happens, if someone produces a drag-and-drool transcoding tool that
doesn't indicate transcoding was used in a OGG comment of some consistent
form, I will personally track them down and shoot them. :)

> So, we don't need it, but is good take a look of the posibilities. Remember
> the posible inclusion of Monkey's Audio on the specification of Ogg. Lossy
> and lossless compression in the same package... nice!

Possibilities are good, but we must always track the possible failure modes
as well. :) 

--- >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.