Summary arguments: Ogg-Vorbis v. mp3

pro Ogg-Vorbis:

*smaller file size for comparable audio quality
*generally better fidelity at any low/medium bit rate
*extreme freedom with metadata--no restrictions of information size
*open-source and patent-free (actually the least important fact, but a plus)

pro mp3:

*ubiquitous
*people don't want to download new software if they don't have to.
*The iPod doesn't do Ogg-Vorbis (perhaps the strongest argument)

Judgeing from the above arguments, which even the pro-mp3 camp seems to agree
are the major arguments on both sides, isn't it clear that Ogg-Vorbis wins,
hands down, if your primary concerns are file-size and sound, and freedom
with metadata? (caveat below)

caveat: For some people (Jon?), the iPod argument is enough to be wary of not
providing mp3 files. Not enough for me, since there are competitive players
that do offer Ogg-playing capabilities out there right now...maybe what Apple
needs to do is start releasing players with Vorbis ability.

mp3's by the way, are shitty with metadata: If I'm a co-composer with someone
else of a work, and we both want to write our full-names to be scrolled by
WinAmp, XMMS, or whatever, we are SOL. I would have to be "ArnKrstrJhnsn"
instead of "Aaron Krister Johnson", if even that, depending on my
collaborator's name(s). And I hate that I can't write the date "February 2nd,
2005" and have to write "2005" only in mp3. Plus, what if I want to write the
time of day for any reason? Sorry, again, SOL.

I think mp3s are terrible in more ways than one.

Best,
--
Aaron Krister Johnson
http://www.akjmusic.com
http://www.dividebypi.com

>pro mp3:
>
>*ubiquitous
>*people don't want to download new software if they don't have to.
>*The iPod doesn't do Ogg-Vorbis (perhaps the strongest argument)

The LAME mp3 encoder has been tested far more rigorously on a
far wider range of source material than any other lossy encoder
in the world.

>mp3's by the way, are shitty with metadata: If I'm a co-composer
>with someone else of a work, and we both want to write our full-names
>to be scrolled by WinAmp, XMMS, or whatever, we are SOL. I would have
>to be "ArnKrstrJhnsn" instead of "Aaron Krister Johnson",

Not true.

>if even that, depending on my collaborator's name(s). And I hate that
>I can't write the date "February 2nd, 2005" and have to write "2005"
>donly in mp3.

The problem with Ogg is that date formats are not restricted.

>Plus, what if I want to write the >time of day for any reason?
>Sorry, again, SOL.

Comment field, which is just as un-restricted as the Ogg date
field.

-Carl

On Saturday 19 February 2005 11:11 am, Carl Lumma wrote:
> >pro mp3:
> >
> >*ubiquitous
> >*people don't want to download new software if they don't have to.
> >*The iPod doesn't do Ogg-Vorbis (perhaps the strongest argument)
>
> The LAME mp3 encoder has been tested far more rigorously on a
> far wider range of source material than any other lossy encoder
> in the world.

And it still sucks. It added artifacts (a crescendo, in fact) to my Bull "Ut
Re" rendering.

> >mp3's by the way, are shitty with metadata: If I'm a co-composer
> >with someone else of a work, and we both want to write our full-names
> >to be scrolled by WinAmp, XMMS, or whatever, we are SOL. I would have
> >to be "ArnKrstrJhnsn" instead of "Aaron Krister Johnson",
>
> Not true.

I can only add 30 characters to the composer field. Including white space. So,
If the sum of my co-composer's name and mine is > 30, we're SOL....

No such limit in ogg. I don't know where our info is coming from.

>
> >if even that, depending on my collaborator's name(s). And I hate that
> >I can't write the date "February 2nd, 2005" and have to write "2005"
> >donly in mp3.
>
> The problem with Ogg is that date formats are not restricted.

Ok, if you want to think of that as a problem, be my guest.

> >Plus, what if I want to write the >time of day for any reason?
> >Sorry, again, SOL.
>
> Comment field, which is just as un-restricted as the Ogg date
> field.

So, by your logic, limitations on metadata are good? Ok, mp3 is for you then.
I can't argue with you there....

Aaron Krister Johnson
http://www.akjmusic.com
http://www.dividebypi.com

>> >pro mp3:
>> >
>> >*ubiquitous
>> >*people don't want to download new software if they don't have to.
>> >*The iPod doesn't do Ogg-Vorbis (perhaps the strongest argument)
>>
>> The LAME mp3 encoder has been tested far more rigorously on a
>> far wider range of source material than any other lossy encoder
>> in the world.
>
>And it still sucks. It added artifacts (a crescendo, in fact) to my
>Bull "Ut Re" rendering.

With lame at preset-standard, it's widely believed, by people who know
more about this than you or I, that ogg has more "problem cases" than
mp3. This is because mp3's problem cases have been rigorously tracked
down and solved, whereas the development focus for ogg has been aimed
at lower bitrates.

>> >mp3's by the way, are shitty with metadata: If I'm a co-composer
>> >with someone else of a work, and we both want to write our full-names
>> >to be scrolled by WinAmp, XMMS, or whatever, we are SOL. I would have
>> >to be "ArnKrstrJhnsn" instead of "Aaron Krister Johnson",
>>
>> Not true.
>
>I can only add 30 characters to the composer field. Including white
>space. So, If the sum of my co-composer's name and mine is > 30, we're
>SOL....

I just tested 72 characters and it worked fine. In fact I have something
like 20 GB of mp3s, all the tags of which I've edited by hand, and I've
never run up against the limit.

>No such limit in ogg. I don't know where our info is coming from.

Your info is indeed mysterious. I tried it before I spoke, you either
haven't tried it, or haven't taken a breath long enough read that I'm
talking about ID3 v2.

>> >if even that, depending on my collaborator's name(s). And I hate that
>> >I can't write the date "February 2nd, 2005" and have to write "2005"
>> >donly in mp3.
>>
>> The problem with Ogg is that date formats are not restricted.
>
>Ok, if you want to think of that as a problem, be my guest.

Thanks! Glad to be here.

>> >Plus, what if I want to write the >time of day for any reason?
>> >Sorry, again, SOL.
>>
>> Comment field, which is just as un-restricted as the Ogg date
>> field.
>
>So, by your logic, limitations on metadata are good?

That's the whole point of metadata. Otherwise, filenames would be
sufficient. It's nice to have extensibility, but if you don't have
good standard fields your metadata implementation is bogus.

-Carl