mini-disc as source?

[Dan McDonald]

Re: mini-disc as source?

Thanks, Rick. This is interesting. Looks like the MP3 format at 128 cuts out nearly all the information above 14k, or am I misinterpreting the graph?

For a lot of MP3 stuff when I used the spectrum analyzer, it looks like it drops off even lower - around 10-12k. I imagine that the dropoff is associated with low encoding fequencies as Peter mentioned. Is there a formula for figuring out how much is lost based on the encoding frequency?

I still wonder how MD would compare to a decent portable cassette recorder of about 10 years ago.

Dan

[Rick]

Re: mini-disc as source?

Dan
You are correct, the bandwidth is down around 14 to 16kHz, but notice that it is not a constant. The MP3 compression algorithm will try to preserve signals that are above the perceptual threshold if possible.
There is not a simple formula that will correlate bandwidth to bits/sec. It is obvious that lower bitrates will reduce the bandwidth, but how much depends on the music, the amount of stereo separation and the actual spectral content.

Rick

[Dan McDonald]

Re: mini-disc as source?

Thanks, Rick.

Also, as was mentioned before, I would assume that how much is lost depends on how much is present. That is, recording a rock concert, where everything is amplified and you are already limited by the band's p.a. system, microphones, etc., may give pretty decent results with a mini-disc or mp3 format, whereas unamplified acoustic music, such as a symphony orchestra or bluegrass band on the street, may lose a lot more because the full range of the instruments is available for recording?


Dan


[This message has been edited by Dan McDonald (edited 05-21-2002).]

[Rick]

Re: mini-disc as source?

I would generally agree with that. Dynamic range, bandwidth and musical complexity all play a part. You would also be much less critical of a rock concert recording because of the compression and bandwidth limitations.

Rick