No, I don't think so. Was: [vorbis] Vorbis 1.0 binary - Beta4 still the best?

[Zvezdan Dimitrijevic]

>> One must never use graphs to look for artifacts.  Graphs are useful
>> only as a means of visually explaining what your ears tell you.  What
>> you've pictured above isn't an artifact unless you can hear it.  

I knew that you will incorrectly interpret my intention when I sent
these graphs. I was wrong because I didn't say that I came to my
conclusion after extensible listening test first. I generated these
graphs tonight (GMT), short time before I sent they to this forum, and
long time after mentioned listening tests.

I came to idea to generate these graphs because I recalled to my mind
that they can have a connection with my previous observation. My action
was affected by you message where you wrote that you demonstrated some
people with spec analyzer that WMA's lowpassing at 10kHz.

>> There's nothing in the diagram that we don't all already know.
>> starting with rc2, the encoder subtracts phase information from a
>> sample that it deems inaudible,

I am sorry, I didn't know precisely what is all information your encoder
throws from source sound. I knew for throwing high frequencies with
low-pass filter and so, but subtracting phase information is new for me.

OK people, I should agree with all of you. If I am only person who can
hear those differences in phase, it is obviously to me that all of my
previous conclusions is totally unnecessary and worthless. It will be
nonsense to return on algorithm founded in old version like Beta4 for
only one person. I give you a word that I will not bother you anymore.

BTW, maybe I should say only one more thing. After Aleksandar's late
night suggestion, I was downloaded so called Garf Tuned RC2 version, and
I spent some time to test it. My preliminary conclusion is that it is
like Beta4, that is it has not behavior which have RC2, RC3 and 1.0,
about I was talking.

Zvezdan

<p><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.