On Sun, 14 Sep 2003 02:43:09 GMT, ff123 wrote:
Roberto's current group listening test is open. This test is at 64
kbit/s, and updates the original test I organized over a year ago:
http://audio.ciara.us/test/
Test ends on September 21st.
Test results:
http://audio.ciara.us/test/64test/results.html
Lame mp3 at 128 kbit/s and FhG mp3 at 64 kbit/s are used as the high
and low anchors, respectively.
1. No codec at 64 kbit/s can claim to be as good as mp3 at 128
kbit/s.
2. High efficiency AAC (AAC with spectral band replication, similar
to mp3 with SBR, aka mp3pro) makes a good first showing, being the
first among equals (he-aac, mp3pro, and ogg vorbis).
3. WMA9, Real, and Quicktime AAC make poor showings. While better
than the hoary mp3 at 64 kbit/s, they are definitely worse than the
top three 64 kbit/s contenders.
The test introduces schnofler's java version of abc/hr, the blind
listening utility:
http://rarewares.hydrogenaudio.org/files/abchr-java.zip
ff123