oggenc weirdness

[Geoff Shang]

On Sun, 24 Feb 2002, Kerry Hoath wrote:

> Since we're all talking about anything and everything at the moment, here's
> one for the list to think on:
> I am currently sampling some 4-track material I own to keep it
> safe in case the tapes age or snap.
> I get the wav files as I want them; (I can summarize to the list if
> anyone cares with the scripts I wrote yesterday to make it all work)
> and these aren't Jim's scripts, I wrote these ones with comments to aid
> in maintainance.
> I have a .wav file 22050 samples per second 16-bit little endian mono.
> I try to oggenc this file with the following command:
> oggenc -b32 side01.wav
> and get a message that "mode initialization failed"
> This used to work fine in 1.0rc2 but appears to have broken in 1.0rc3
> Anyone else running 1.0rc3 and seen this behaviour?

Hi Kerry:

Excuse me if you solved this.  firstly, make sure you're running RC3 of
everything, libs and oggenc.  Theory says that -b32 should work without
giving you errors.  But!  This is probably not what you want.

RC3 changed how some things were done.  Vorbis until recently didn't give
you the rate you asked for, it picked a predefined mode that somewhat
aproximated the rate you entered.  As of version 1.0RC3 however, it will
actually give you the rate you want, but this mode isn't as good as the
predefined modes.  The modes are specified as a quality value that ranges
between 0.0 and 10.0 and can run to several decimal places.  As a rough
guide, here's the aproximate average bitrate for the various VBR quality
settings for a 44.1khz stereo file:

-q 0 - 64 kbps
-q 1 - 80 kbps
-q 2 - 96 kbps
-q 3 - 112 kbps (default if not specified)
-q 4 - 128 kbps
-q 5 - 160 kbps (dramatic reduction in channel coupling at 5 and above)
-q 6 - 192 kbps
-q 7 - 224 kbps
-q 8 - 256 kbps
-q 9 - 320 kbps
-q 10 - 498 kbps

You will have noticed that -q 3.0 (i.e. aprox 112 kbps) is the default.  It
is important to note that these are vorbis bit rates, not MP3 equivalents.
This should be obvious, but the true impact of this might not be.  112 kbps
is, in my experience, very good, and is generally considered to be
equivalent to 160 kbps MP3.  On the occasions I've used RC3, I haven't been
able to hear any degradation at -q 3 and the smaller file size is nice.

These quality values don't yet scale properly for lower sampling rates
(this will finally be done in RC4), but the same principle applies.  So, no
matter what you're
encoding, -q 0 will get you the lowest possible rate.  Using RC3, -q 0 on
22khz mono will give you aprox. 32kbps.  If quality matters to you, you
might want to go higher, perhaps to 1.0 or so.  But hey, experiment and see
what you think.

Geoff.