View Single Post
  #10   Report Post  
Posted to rec.audio.tech,uk.rec.audio,rec.audio.pro,sci.physics
Dave Dave is offline
external usenet poster
 
Posts: 8
Default How can I tell music has been an MP3? Quantitative Measurement of Fidelity

On 23 Jun, 03:43, "Richard Crowley" wrote:
"Dave" wrote ...

, "Richard Crowley" wrote:
1) SNR is not "an accepted measurement of fidelity"
2) There is no specific "accepted measurement of fidelity"
"Fidelity" is a combination of many things. Some subjective.
3) It would be difficult-to-impossible to actually meausre SNR
on a commercial CD because of the way they are mastered.
(i.e. there is no "baseline" because it is usually muted)

Have there been no bright PhD students sponsored by the music
industry, or are they too busy with their revenue stream? The noise
is what is not the notes. For a symphony you have an idea of what the
notes should be, because you have the sheet music, and you know what a
violin, flute etc should sound like. You may be able to measure
something more because that is what the brain does.


That is just impossibly simplistic. There is a great deal of stuff
"between the notes" besides noise. Have you done much
recording yourself?

It doesn't sound as complicated as fusion, and that is having billions
spent on it. Besides I thought banks and defence liked graduates with
in-depth signal analysis experience.

I have listened to plenty of CD (about 650), so I think I can tell a
good recording from a bad one. There may be problems of course with
quantative measurement, in that the recording could be done to get the
measurement
high, and it could just sound clinical.

The point was that if I thought a CD sounded poor quality I think
there should be a computer program to confirm this, instead of just
asking someone else.