[VAXman(Admin)]

Capturing a live performance? IMPOSSIBLE!

I've been at a large number of live events that have later been released as recordings. None of them capture the sound of the live event. Microphones are not human ears; they'll never capture the sound of the live event with the same "accuracy" of the human ear.

As for vinyl vs. CD/digital... C'mon... the vinyl suffered from myriad issues that make it "unfaithful" for sound reproduction. First, to record the vinyl, the sound undergoes dynamic compression and formant spectral equalization (known as the RIAA curve). Lows attenuated (lest the power of the low would cause the cutting stylus to cross grooves) and accentuation of the highs (as it takes more energy to cut at the high end).
During the recovery (playback) the compression and equalization must be reversed. The equalization can, with a fair bit of exacting filter design, be undone but the compression is usually varied by the recording engineer dependant upon the source dynamic. Soft passages are amplified and loud passages are attenuated but there is no information on the vinyl to inform the playback mechanism just how much attenuation and aplification was used and when and where it was used.

The digital medium, OTOH, does not suffer from this. I recall in the early days, people complaining that the dynamic range seemed over emphasized. Was it? Or was it that the ears of the listeners were accustomed to the myriad recording processes of vinyl?


The CD upper end is 20KHz+. Consider that the highest note on a piano is in about 4Khz, 20KHz is 2+ harmonics above that note. If the ear's spectral resonse is not much better, then what does is matter for the CDs?

If anyone here has an understanding of fourier analysis, they will appreciate that there are waveforms above a certain frequency that can not be reproduced.

A square wave, for example, is a function of:
∞
∑ 4/πk *sin (2πk/T)
k=0

If you know anything about the sin() funtion you will realize that even values of k yield a 0 term

Therefore, you get terms of:

4/π(T)$+ 4/3π(3T) + 4/5π(5T)... by the 3rd term the factors are already negligible. So a sq. way at 440Hz (middle A) woudl have harmonics at
1320 and 2200 and would be fairly represented on the CD. I doubt that a recording of a true square wave at 440 on vinyl would be very faithful... there you have to see _-_-_-_-_-_-_-_-_ in the groove. I'd bet it would wear (ie. distort) very quickly.