On Sun, 24 Oct 1999, Leonardo Stern wrote:
> Hi all,
>
> where I can find a program for windows 9x that compares mp3, wma, vqf,
> etc with the wav ?
>
> I've done some tests with wma x mp3 in diferent freq��ncies but I need
> some kind of analizer that could help.
You already have a very good analizer. It's in between your ears. :)
No software currently comes close to true analysis. This is one of the
things that annoy me about those sites that clame to 'prove' bladeenc
sounds better.
Simply, on a pure S/N basis, all the above encoders perform a fair amount
worse then would be possible with a fair simple waveform encoder.
The differnece is, the above methods SOUND a lot better then a much
simpler encoder with better S/N. This is because they take advantage of
the behavior of the human audio perception.
Even if you use something like a sophicated PQAM test, you are just
testing how good it sounds to an imperfect mathimatical model, not human
hearing.
It should be a crime to try to compair modern perceptual audio codecs
using automated methods.
The only way to test is to perform very carefully conducted listening
tests.
I've never conducted a test refined enough to pass for scientific rigor,
but I have used multiple people with good equipment, and the differnece
between blade and lame is noticable, even at high (160-224+) bitrates.
Lame is simply a lot better from my tests. If someone has seen a real (not
synthetic) test that shows blade better, I'd really like to know.
(I'd also like pointers on any synthetic tests using psy models differnt
and more advanced then the model used in mp3, for doing VBR tuning)
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
