Hi,
You may remember my tests with Lame3.25 and sox 12.16 at 24kbit/s.
Though I've not investigated the `polyphase' effect of sox fully, it
seems to be sounding better than the earlier versions and...
using a command line such as
sox INFILE.wav -r 22050 OUTFILE.wav polyphase -cutoff 0.45
to generate a 22.05kHz sample-rate file, filtered at -90dB/oct,
turnover 4961Hz, to feed into lame3.27, it actually SOUNDS BETTER
THAN FhG on speech at 24kbit/s. I don't know whether or not the -Y
-Z and -X n options make a difference, but I put them in anyway. No
other forms of audio have yet been tested.
I think that's amazing, and can't wait to see if the ISO code error
(some factors in the wrong order?) that was lately discovered might
allow me to open up the treble a little more without the "tinkling"
effect.
If you'd like me to post MP3 files demonstrating how (I think)
lame3.27 is sounding better than FhG (3.1) please say so.
John Hayward-Warburton
[EMAIL PROTECTED]
--
MP3 ENCODER mailing list ( http://geek.rcc.se/mp3encoder/ )
