[vorbis] Participate in listening tests

ff123 miyaguch at eskimo.com
Wed Oct 31 07:14:30 PST 2001

Moritz wrote:

>In order to do it right, I will need quite some time. When is the
>deadline? I think I missed that one on your website, or overread it.

There is no deadline.  The tests can run indefinitely unless answers or 
comments leak out.  I will periodically release statistical results, but 
the comments on the sound will remain undisclosed until the tests are closed.

So here are some statistical results for wayitis.wav.  There are very few 
listeners at this point, and the type of analysis makes some assumptions 
which may not be warranted, so I'd interpret these results as being a 
suggestion of what the final results might look like rather than swallowing 
it whole without qualification.

FRIEDMAN version 1.21 (Oct 16, 2001) http://ff123.net/
Blocked ANOVA analysis

Number of listeners: 7
Critical significance:  0.05
Significance of data: 4.02E-02 (significant)
ANOVA Table for Randomized Block Designs Using Ratings

Source of         Degrees     Sum of    Mean
variation         of Freedom  squares   Square    F      p

Total               41          54.00
Testers (blocks)     6          30.85
Codecs eval'd        5           7.16    1.43    2.69  4.02E-02
Error               30          15.99    0.53
Fisher's protected LSD for ANOVA:   0.797

          ogg      aac      lame     wma8     xing
mpc        0.11     0.40     0.43     1.00*    1.10*
ogg                 0.29     0.31     0.89*    0.99*
aac                          0.03     0.60     0.70
lame                                  0.57     0.67
wma8                                           0.10

mpc      ogg      aac      lame     wma8     xing
   4.56     4.44     4.16     4.13     3.56     3.46

mpc is better than wma8, xing
ogg is better than wma8, xing

--- >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.

More information about the Vorbis mailing list