I love music, all kinds. I have played in everything from jazz ensembles to blues or beach music bands to symphony orchestras. Likewise I listen to a broad range of music, though I lean toward the more acoustic performance genres than electronic.
There are both empirically measurable physics and unavoidably subjective qualities involved in the design and application of audio components, especially for analog transducer elements at each end of the chain, microphones, pickups, loudspeakers, analog to digital converters at both ends.
I'll address the digital realm first since that is the easiest to deal with. Digital encoding is simply applying a scale of discrete numeric values to a varying analog voltage. Think of a simple yard stick. The physical length of the stick is represents the range of analog values. The incremental numeric marks the digital values. The smallest measurement increment is the scale resolution. For a yard stick perhaps the highest practical resolution is 1/32" and the limiting resolution beyond which it becomes useless is perhaps 1/64th inch. Digital audio resolution has progressed from the original CD format 44.1KHz/16bit through the typical studio values of 48KHZ-96KHz 24bit to current DSP software and internal sampling values as high as 192KHz with full 32 bit float, i.e infinite fractional value encoding.
Once an audio signal is in the digital realm, the only critical thing is that those numeric values get transferred between AD and DA converters with absolute mathematical accuracy. The hardware at its best is accurate to one bit error in billions.
To quote Don Davis again, you don't empirically know something in physics until you can assign it a number. The digital audio realm today is the realm of applied mathematical physics. The only effect it can have on audio quality is that which is deliberately introduced for desired reasons depending on application.
Two important things to understand about physics.
1. Science does not deal in absolutes. It deals in probabilities and approximations. The gold standard measurement constant in physics is the speed of light in a vacuum. After nearly 200 years physicists still quibble over the exact measurement of this value depending on which set of theoretical equations are used... at the 17th decimal place!
2. Beyond a reasonable point empirically related to the task at hand mathematical precision becomes irrelevant.
To be continued...