The fundamental difference is the ability to accurately represent the original incoming analogue waveform digitally.
The quality of the individual components and they way the are put together is quite complex and requires expensive high quality components.
Just consider the powersupply and its shielding. Have you ever had your cellphone ring while you are at the computer. You hear the interference through the speakers before you hear the phone rings.
That is an example of what radiation of various kinds do to audio signal paths.
In technical terms we have to sonsider theese components of the puzzle:
amplitude, frequency and phase modulations: time and frequency domain characteristics, modulation and demodulation principles; multiplexing; noise, power density, signal-noise ratio.
Sampling theorem. Impulse modulations:
amplitude, position, duration impulse modulations, code-impulse and delta modulations; multiplexing; quantisation noise, decision error probability.
Digital signal transfer:
intersymbol interference, Nyquist's first criterion, optimal filter, modulations.
If you want to read how this affects us in the audio world you should look at post by Dan Lavry on this site. Some of the debates get quite heated at times and are quite educational.
Maybe someone else will get into how the design of audio ad/da units affects the final result.
There is is a huge difference in quality between the best and the worst.
I can say this much:
No one at creatives research lab will be getting jobs designing AD converters for the nuclear accelerator in Cern anytime soon.
Here are some reviews of a benchmark that touches on some of the issues http://www.digitalprosound.com/artic...fterinter=true http://www.benchmarkmedia.com/news/dac1PARreview.html
hope that helps