[vorbis] An Analysis of the c't Codec Test

[James Snook]

I did a week of blnd testing with WinABX, although my equipment is not
the best going it's ok(SBLive 5.1, Sony amp, Wharfedale speakers and
namelss headphones used), thing is I find that all my music encoded with
LAME --r3mix and a maximum of 256vbr that I can almmost never pick mp3
from wav. However it took untill Vorbis -q 6.15~22 for me to stop
picking the Vorbis from wav, using 4 albums I concider high stress due
to the guitar work, or the range of noise in the tracks..  And sadly im
my opinion (and multiple ABX logs) I think that in order to BEAT --r3mix
LAME the Vorbis files will always need to be bigger in file size (on my
CD's ofc, I think more sample based music compresses better).
I don't have my logs anymore, as it was all purly personal testing for
myself and some friends, but the fact was there that I could distinctly
hear high freq' flange and "washy sound" in Vorbis up untill about 7.00,
although I admit you had to really listen hard to even tell, the fact
remains for my music that MP3 is just plain better still for file size
to sound quality ratio =/
All the same, I choose to use ogg at 6.20 now dispite these ABX results,
purly because I belive Vorbis has great potential, and the ability to
improve well beyond even r3mix tweaked mp3's.

Visit www.r3mix.net for very good information on MP3 encoding, it helped
me test Vorbis fairly vs MP3 at its best, not some awfull Xing or
default LAME setting.
Just though I'd say what I found :)

<p>-----Original Message-----
From: owner-vorbis at xiph.org [mailto:owner-vorbis at xiph.org] On Behalf Of
Per Wigren
Sent: 01 October 2002 16:28
To: vorbis at xiph.org
Subject: Re: [vorbis] An Analysis of the c't Codec Test

<p>Isn't the purpouse of using quality instead of bitrate that it will
always be 
the same quality? So very-easy-to-encode sound should sound as good as 
very-hard-to-encode sound, but the easy sound uses 60kbit/s and the hard
one 
380kbit/s ?

// Wigren

Tuesday 01 October 2002 15.09 skrev Ross Vandegrift:
> On Mon, Sep 30, 2002 at 10:48:27AM -0100, fungus wrote:
> > One the one hand this is a great achievement but on the other hand 
> > normal people aren't going to figure out how good it is because 
> > they'll never push the limits. Who's going to set "quality 0.5" when

> > they've got a range from zero up to ten.
>
> This is very true.  I can easily hear MP3 distortion up to about 
> 256kbps on most music.  I've heard some songs where I can tell at 
> 256kbps.
>
> I have never heard compression artifacts from Ogg Vorbis. 
> Unfortunately, I'm scared that one day I'll be able to, and my entire 
> CD collection, encoded at the economical q=3.0 or less will become 
> sick sounding.  So I kinda wuss out and just set q to 5-ish, so any 
> difficult to encode music has some headroom.

--- >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.

<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.