On Mon, 25 Oct 1999, Leonardo Stern wrote:
> Let me explain what I want isn't a program that compares 2 codecs
> I want a program that analize the frequency response of a wav (ripped
> from cd) and a mp3 or other and display both response graphics
No you dont. This is a completely contrived and invaild way of compairing
perceptual codecs. Please read up on how they work, good freq responce
means nothing!
I can whip up an encoder that will have teriffic results in that kind of
test, have super low bitrates, and sound completely unintelgeiable.
(spectrally shaping whitenoise would probably be good enough).
[snip]
> I encoded 4 types of music and always the wma sounds better even in 96
> kbps. But I want to do more tests with different coputers, comparing
> methods, musics, etc.
'sounds' define sounds?
Spacialazation?
Artifacts?
Intelegiblity?
Is it indistinguishable? To who? How do you know?
There is no such thing as an algrithmic objective score of natural
human-percieved sound quality.
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
