Re: What is a "high resolution audio system"?
Reply #9 – 2023-01-31 17:20:34
The "system" appears to have, in those words, a "resolution" or "resolving power". Hence my question: how does one measures or quantify this "resolution"? And how can an analog system be smart enough to hide anything? I think you're being too dismissive. An analogue system clearly can't be "smart", but it certainly can hide stuff! Ignoring the transducers for a moment, there is an electrical signal path from input to output which is ideally linear, but actually never is. By "linear" we mean that the graph of output voltage plotted agains input voltage is a straight line. In practice, the transfer function depends on the input frequency, so for an input containing a broad spectrum of frequencies the output is the sum of all the input frequencies multiplied by their respective transfer functions. For any particular frequency, the transfer function may also include a phase shift. Then there is the problem of the transducers: taking acoustic pressure waves and converting those to an electrical signal, using the amplified electrical signal to waggle a needle (or whatever) cutting a groove in a master disc, then at the reproduction end picking up the motions of the stylus and converting those to electrical, and finally the amplified signal into a loudspeaker to recreate the acoustic signal. The fundamental purpose of the electrical stages is to boost the power so that the input transducer can drive the output transducer with sufficient force, but is generally "shaped" to compensate for performance shortfalls in the transducer. This shaping may not be perfectly matched to each individual transducer, so this introduces response errors. Also, in addition to the actual wanted signal, there is an unavoidable random noise signal (hiss) injected into the signal path by the components in that signal path. Consequently, even if it were possible to design an amplifier to have a perfectly flat and phase zero response over the entire audio spectrum (it isn't), the transducers would still impose variations in the frequency and phase response. Any lack of linearity WILL HAVE an effect on the output, and COULD HAVE an audible effect on resolving the detail of the sound (for those with genuine ability to hear it - which does not include me). If the fine detail of the signal is of a similar level as the hiss, it will definitely be hidden even if the system were capable of reproducing it. (Note that transducers can also have a response which is modulated by the signal which has been present immediately before, which adds another level of complication.) Digital avoids a lot of this, but introduces its own problems in terms of analogue to digital conversion linearity and resolution, sample clock jitter etc. And it still has to have a microphone at the input and a loudspeaker at the output which (as noted) need shaping.