A real acoustic waveform of many instruments playing live is possibly a very beautiful sound but a very ugly waveform full of peaks and spikes.
When we say then human ear can cope from 0dB to 120dB we mean a difference (ratio really) of 12 0s. If 0dB were say 1W, 120dB is 1000,000,000,000W an extremely large number.
The difference does not sound as loud. One reason being the 3 tiniest bone in the human body that connect the ear drum to the rest of the ear-brain mechanism. The pivot points change at loud sounds doing apparently an instantaneous limiting of ear sensitivity.
If we record and reproduce a dynamic range of 120dB it may sound the worst thing. No amplifier can cope with this,
In fact we do not want a great dynamic range when we listen at home as the low levels will be masked by ambient noise and the high levels will be clipped.
How much dynamic range has a great recording like Kind of Blue - Miles Davis? One could figure out by looking at VU meters while it is playing. Yet it sounds subjectively very dynamic. The answer is on the AES paper by Russel o Hamm, Tubes vs Transistors.
Also electron tubes and analog recording eliminate the nasty peaks and concentrate on the music.
If one wishes to record the nasty peaks too one is left with very few digital steps for the main music signal. And because we wished to record the high peaks too the whole recording had to be reduced in level in order to fit thle digital finite headroom.
So the average level is low and the recording sounds low level, lacks body.
Analog recording "eats" the peaks by instantaneously clipping them. But because there is emphasis on recording (increasing treble while recording) and de emphasis during playback (decreasing treble while playback) this rounds artfully back the waveform peaks. The same principle is used in tune quitar amplifiers, one low pass filter in the chain being the loudspeaker itself. The same principle is used in FM broadcasting (see also Orban).
And we are left with a low peak to average ratio. Ie the average level can be high and the peaks will not clip an analog to digital converter.
This possibly explains why digital sounds nice when it re records an analog recording (eg Kind of Blue) but leaves something to be desired when recording real music.
Reference:
www.261.gr
Crafting loud mixes that sound great - Sound on Sound
Tubes vs Transistors is there an audible difference - Russel o Hamm - JAES
When we say then human ear can cope from 0dB to 120dB we mean a difference (ratio really) of 12 0s. If 0dB were say 1W, 120dB is 1000,000,000,000W an extremely large number.
The difference does not sound as loud. One reason being the 3 tiniest bone in the human body that connect the ear drum to the rest of the ear-brain mechanism. The pivot points change at loud sounds doing apparently an instantaneous limiting of ear sensitivity.
If we record and reproduce a dynamic range of 120dB it may sound the worst thing. No amplifier can cope with this,
In fact we do not want a great dynamic range when we listen at home as the low levels will be masked by ambient noise and the high levels will be clipped.
How much dynamic range has a great recording like Kind of Blue - Miles Davis? One could figure out by looking at VU meters while it is playing. Yet it sounds subjectively very dynamic. The answer is on the AES paper by Russel o Hamm, Tubes vs Transistors.
Also electron tubes and analog recording eliminate the nasty peaks and concentrate on the music.
If one wishes to record the nasty peaks too one is left with very few digital steps for the main music signal. And because we wished to record the high peaks too the whole recording had to be reduced in level in order to fit thle digital finite headroom.
So the average level is low and the recording sounds low level, lacks body.
Analog recording "eats" the peaks by instantaneously clipping them. But because there is emphasis on recording (increasing treble while recording) and de emphasis during playback (decreasing treble while playback) this rounds artfully back the waveform peaks. The same principle is used in tune quitar amplifiers, one low pass filter in the chain being the loudspeaker itself. The same principle is used in FM broadcasting (see also Orban).
And we are left with a low peak to average ratio. Ie the average level can be high and the peaks will not clip an analog to digital converter.
This possibly explains why digital sounds nice when it re records an analog recording (eg Kind of Blue) but leaves something to be desired when recording real music.
Reference:
www.261.gr
Crafting loud mixes that sound great - Sound on Sound
Tubes vs Transistors is there an audible difference - Russel o Hamm - JAES
No comments:
Post a Comment