[Mr.T]

"Dave Platt" wrote in message
...
I've read a number of opinions stating that with a good MP3 encoder,
most audio material encoded at a 256 kbit/second data rate is audibly
indistinguishable from the original. It's still "lossy" - the output
audio stream is not bit-identical with the original - but the
differences are small enough that psychoacoustic masking makes it very
difficult to detect the differences reliably.

Repeatedly encoding and decoding audio material (i.e. multiple passes)
with MP3 will result in an increasing amount of sonic degradation,
even at these relatively high bit rates.

While this is true, many people simply quote this as akin to re-recording
cassette tapes. In my testing the first encode creates more loss than
another 5 encodes, simply because data has already been thrown away, and no
longer needs to be encoded anyway.
So editing high bit rate MP3 is not as bad as it first seems IME.


Doing a truly lossless encoding of high-quality audio, and achieving a
high compression rate, is far from trivial. As a rule of thumb,
today's lossless audio encoders (e.g. FLAC, Apple Lossless, MPEG-4
ALS, WavPack) seem to achieve a reduction in data size of roughly 50%,
with quite a bit of variation which depends on the characteristics of
the music. Repeated encode/decode passes with these algorithms do
*not* result in any sonic degradation.

If it did they would not be lossless would they!

MrT.