[vorbis] Testing of Merged GT3b1 and 1.0.1 oggenc

[Stephen So]

Hi everyone,

Recently I made an attempt at merging the tunings of GT3b1 into the 
official Xiph.org 1.0.1 libvorbis code.   While 99% of the GT3b1 tunings 
are independent of 1.0.1, there is one particular value 
(_psy_noiseguards) which affects the whole quality range.  I've chosen 
to use the 1.0.1 values since they changed since 1.0.  Therefore, files 
encoded in GT3b1 mode (q 5 and above) are slightly larger than normal 
GT3b1 files.

I would appreciate it if some people could test the quality of this 
encoder and make sure that not only are there no breaks in quality, but 
that this encoder behaves (quality-wise) the same as both 1.0.1 and 
GT3b1 in their respective quality ranges.  Also, try encoding at lower 
sampling rates.

At q -1 to 4, oggenc will use the Xiph.org 1.0.1 tunings
At q 5 to 10, oggenc will use the Sjeng.org GT3b1 tunings, which has 
better pre-echo and transient handling.
(Between 4 and 5, there will be a mix between the two)

The thread at hydrogenaudio.org, where I announced this is at 
http://www.hydrogenaudio.org/show.php/showtopic/17824

Windows and Linux binaries are available at Rarewares 
(http://rarewares.hydrogenaudio.org), with special thanks to John for 
verifying, compiling, and uploading. :)

Please note that this is experimental and the quality of the files is 
uncertain.  The vendor tag has the word EXPERIMENTAL so that files are 
tagged appropriately. :)

<p>Best regards,

Steve.


-- 
--------------------------------------------------
Stephen So, BEng(Hons)

PhD Student, 
Signal Processing Laboratory,
School of Microelectronic Engineering,
Faculty of Engineering and Information Technology,
Griffith University, Nathan Campus,
Brisbane, QLD, Australia, 4111.

Phone: +61-7-3875 3754
E-mail: s.so at griffith.edu.au
--------------------------------------------------

--- >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.