"KikeG" wrote in message
om...
"Keith G" wrote in message
...

My own initial feeling (in what was ostensibly a 128 Kbps CBR/MP3 vs.
110 -150 Kbps VBR/MP4 comparison) is that the MP4s sounded a little
'fresher' (more open in the treble?), but I haven't yet had the
opportunity
to do this blind (using my own Electro-Acoustic Replay Scrutinizer),
where
the tracks will be selected for me in random pairs......

At 128 Kbps and below, formats such as Ogg Vorbis and AAC are clearly
better than MP3, specially using a good AAC implementation such as
Apple iTunes one. At 128 Kbps and over, there's another very good
performing contender called Musepack or MPC. At high bitrates, MP3 is
not that bad, specially if compared to Ogg Vorbis, but at around 160
Kbps, Ogg Vorbis and even more AAC, tend to still be better, and MPC
is probably the best with difference. At bitrates of 200 Kbps MPC is
the winner without doubt, achieving perceptual transparency with
virtually all kind of sounds, and of the formats left, probably only
AAC is still beter than MP3.


Interesting. I've discounted MP4 for the moment on the grounds that it's a
bit of a red herring. I've also come to a bit of a conclusion on the MP3/WMA
front and have posted a couple more tracks to compare, if anyone is
interested. (Ain't like we're busy in here, or anything.....)


http://www.apah69.dsl.pipex.com/keit...t1-track01.mp3

http://www.apah69.dsl.pipex.com/keit...t1-track02.wma

and

http://www.apah69.dsl.pipex.com/keit...t2-track01.mp3

http://www.apah69.dsl.pipex.com/keit...t2-track02.wma


I've picked music to contrast with the earlier samples and would recommend
anyone to download them and burn an audio CDR so they can be flicked about
on an ordinary CDP. If anyone has got a 'sudden death' preference, I would
be interested to hear it......

"Keith G" wrote in message ...

Interesting. I've discounted MP4 for the moment on the grounds that it's a
bit of a red herring.

Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

At the end of the page is the overall score of each codec. At the
beginning of the page there is a link to the codecs and options used.

"Keith G" wrote in message ...

Interesting. I've discounted MP4 for the moment on the grounds that it's a
bit of a red herring.

Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

At the end of the page is the overall score of each codec. At the
beginning of the page there is a link to the codecs and options used.

On 18 Dec 2003 02:01:05 -0800
(KikeG) wrote:


Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

What an appaling test. the bitrates are way off (at least the guy documents that)

only three of the codecs are nominally 128kbit, LAME is substantially lower and ogg somewhat higher.

if you look st the results knowing that, it shows the test up as useless.

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

On 18 Dec 2003 02:01:05 -0800
(KikeG) wrote:


Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

What an appaling test. the bitrates are way off (at least the guy documents that)

only three of the codecs are nominally 128kbit, LAME is substantially lower and ogg somewhat higher.

if you look st the results knowing that, it shows the test up as useless.

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.

"KikeG" wrote in message
m...
"Keith G" wrote in message
...

Interesting. I've discounted MP4 for the moment on the grounds that it's
a
bit of a red herring.

Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

At the end of the page is the overall score of each codec. At the
beginning of the page there is a link to the codecs and options used.




Interesting, but not quite what I'm looking to compare (MP3 vs. ordinary
WMA) - thanks for posting anyway.

"KikeG" wrote in message
m...
"Keith G" wrote in message
...

Interesting. I've discounted MP4 for the moment on the grounds that it's
a
bit of a red herring.

Some months ago there was a listening test over various codecs at 128
Kbps, performed by a multitude of people, using double-blind methods.
You can look at the results at he

http://audio.ciara.us/test/128extension/results.html

At the end of the page is the overall score of each codec. At the
beginning of the page there is a link to the codecs and options used.




Interesting, but not quite what I'm looking to compare (MP3 vs. ordinary
WMA) - thanks for posting anyway.

Ian Molton wrote in message ...

What an appaling test. the bitrates are way off (at least the guy documents that)
if you look st the results knowing that, it shows the test up as useless.

The fact is that these codecs do achieve a average bitrate of 128 Kbps
when used to encode full albums. The actual bitrate used just for
these small test samples is not representative of realistic usage when
encoding full albums. That's the nature of VBR encoding. Please read
the presentation of the test (the first link at the page), this is
explained at there.

This issue was thoroughly discussed at the Hydrogenaudio forums before
the test was launched, and it was agreed that the final procedure used
was the most realistic way of calculating bitrates.

Ian Molton wrote in message ...

What an appaling test. the bitrates are way off (at least the guy documents that)
if you look st the results knowing that, it shows the test up as useless.

The fact is that these codecs do achieve a average bitrate of 128 Kbps
when used to encode full albums. The actual bitrate used just for
these small test samples is not representative of realistic usage when
encoding full albums. That's the nature of VBR encoding. Please read
the presentation of the test (the first link at the page), this is
explained at there.

This issue was thoroughly discussed at the Hydrogenaudio forums before
the test was launched, and it was agreed that the final procedure used
was the most realistic way of calculating bitrates.

On 18 Dec 2003 08:30:14 -0800
(KikeG) wrote:

The fact is that these codecs do achieve a average bitrate of 128 Kbps
when used to encode full albums. The actual bitrate used just for
these small test samples is not representative of realistic usage when
encoding full albums.

Pure nonscience.

if the average for the tracks is 128kbit, the avergae for the album is 128kbit and vice-versa

if you have some exceptionally hard to encode tracks it doesnt mean the other tracks should suffer in the comparison just to make it fair. and thats going to be different for ever album anyway.

either compare all at the same average bitrate for a given piece, or dont bother at all.

--
Spyros lair: http://www.mnementh.co.uk/ |||| Maintainer: arm26 linux

Do not meddle in the affairs of Dragons, for you are tasty and good with ketchup.