[LAU] LAM -2008 mix online : update

[hollunder at gmx.at]

On Thu, 13 Nov 2008 12:41:07 +0100
<hollunder at gmx.at> wrote:

> On Thu, 13 Nov 2008 10:45:42 +0700
> Patrick Shirkey <pshirkey at boosthardware.com> wrote:
> 
> > Hi,
> > 
> > I have adjusted the sample rate for a couple of tracks and added a
> > couple more that I missed from the original release.
> > 
> > The mix is now 100 minutes long. Still no Hiphop tracks and no
> > classical pieces either.
> > 
> > http://djcj.org/audio/lam/lam-2008.ogg
> > 112,379MB - ogg
> > 
> > - If people can have a listen and let me know if the levels are
> > right that would be helpful. It sounds fine on my headphones but
> > may be too quiet at the start on a speaker system.
> > 
> > - I have compressed it with q5 this time. That makes the file size
> > about 40MB bigger. The previous version was q3 which is the default
> > for oggenc. Please let me know if that improves the audio quality. 
> > 
> > 
> > Here's the full playlist:
> 
> It would be nice to have that available in the file somehow.
> I know that it's possible with vorbis, I just don't know how.

Excuse me for replying to myself.

I found out how to do it, it's as simple as:

cat song1.ogg song2.ogg song3.ogg > song123.ogg

The problem is that very few players handle it in an acceptable way,
and none that I tried showed the tag of anything but the first file, so
it's a no-go, unfortunately.

Sadly, the only player I know of that handles this correctly is still
foobar2k, a proprietary player for windows. It works in wine but that's
no use really..


> And another thing, usually the recommended ogg encoder is the
> aoTuV-version which one can get here as .deb for example:
> http://rarewares.org/ogg-oggenc.php
> Afaik the aoTuV-changes are going to be merged into mainline oggenc,
> I don't know if this has already happened.
> Benefits should be: faster, same or better quality vs. filesize

I just heard from a dev that normal oggenc should be fine as long as
one doesn't go below 80kbps.


One thing about levels: they are imho not optimal. One thing you could
try: wavgain a copy of the files (trackgain), make the vorbis file and
check if the levels are fine. theoretically it should work reasonably
well.

Best regards,
	Philipp