[Puma Cat]

Originally Posted by audio bill

This has been explained many times, but I'll try again... First of all digital data is not transmitted in digital format (ones and zeros as many believe) over a cable, but as an analog signal representing the digital data. As such the signal is not a perfect square wave... in reality it exhibits overshoot, settling time issues, jitter, judder, etc. which are all types of distortion related to the timing of the signal transitions (when changing between the levels representing zero and one).

When transmitting computer data only the recovered 'digital' data is of meaning, but with audio signals the timing of the data is also of significance. The digital clock signal is embedded in the timing of those transitions, and can be directly affected and corrupted by the types of distortion mentioned above. So a digital cable transmitting an audio signal via the SPDIF standard will have the resultant analog signal altered from the original analog signal after D to A conversion from such timing related distortions directly impacting accurate recovery of the digital signal's clock.

That's the best I can do to explain the difference between computer data transmission and digital audio transmission and why there is a difference. Hope it helps!

This is really a great post that explains the key performance attributes impacting digital playback, especially when streaming from music servers. The bottom line is that EVERYTHING matters and has an impact on perceivable audio quality. It doesn't matter if a printer or computer display has to wait for a few micro- or miiliseconds for a data packet to be re-sent. However, for music reproduction, timing is absolutely critical. This is why folks spend big bucks for digital (femto)clocks with rubidium crystal oscillators, for example.

Ethernet and USB cables are also subject to a lack of galvanic isolation, EM, RFI radiation, induction from EM radiation from wide-bandwidth processors, e.g. CPUs, GPUs, vibration, and other sources of noise. Key parameters here are not only the timing of the signals, but the quality of the power supplies as well. And these noise components are discernable and audible. A big source of garbage noise are the very cheap switch-mode power supplies in computers used as music servers; these are incredibly "dirty". One of the best things can do is to get a computer out of the audio rack and as far away as possible from the amplfication components (inverse-square law and all that...)

There is a really good video on this on Hans Beekhuzen's YT channel: https://youtu.be/n26CMGX_yrk