You guys are seriously contorting yourselves out of shape.
[1] A square wave does not have infinitely fast rise and fall times. [2] The difference between a 1 and a 0 is infinite, and can only exist in our imagination. We pretend that a voltage change across a given threshold inbetween slices of time is a 1 or a 0. It's a useful approximation. This just an educational statement I'm not trying to make point.
[3] Also, why are high grade clocks great in pro audio, but snake oil for audiophiles?
1. It's really not so difficult to understand. Look at this image:
At the leading edge the negative half (lower value) of the cycle is connected to the positive half (higher value) by a vertical line and as the x-axis is time, this means that the amount of time taken for the transition between the negative and positive halves of the square wave is zero, IE. The end of the negative half and the beginning of the positive half occur at the same moment in time. It should be obvious that this is a physical impossibility in the real world: A speaker cone cannot be in two different places at the same moment in time, the air molecules through which the wave is travelling cannot be in two different places at the same time and of course the human ear drum cannot be in two different places at the same time! There has to be some amount of time for the speaker cone/air molecules/ear drum to move from the end of the negative half to the positive half of the square wave but of course that means they would be connected by a sloping line rather than a vertical line which in turn means the square wave is no longer "square".
2. Who is the "we" you're referring to? I presume it must be audiophiles similar to yourself because it's certainly not scientists or those who created the USB specs. The fact it's an approximation (rather than perfect square waves) is completely irrelevant because the 0's and 1's can be perfectly reconstructed from these approximations, regardless of how approximate they are (up to a pretty broad limit). So what are you saying? The CPU in my computer is doing billions of calculations per second across billions of transistors communicating with each other using electrical currents to represent 0's and 1's. So you're saying that either a CPU has an "imagination" or that it can't possibly work because all the billions of approximations per second would lead to constant errors? Err, thanks for that "education" but it educates/says more about you personally than it does about how the transference of digital data actually works.
3. As has been explained in other responses, there are circumstances in pro audio where external clocks are unavoidable. There have of course been examples of excessively "high grade" clocks in the pro audio world. The difference between the pro audio and audiophile worlds is that pros generally understand and therefore accept the valid measurements/science used to identify snake oil products and reject the marketing/testimonials/etc which contradicts it. This is pretty much the opposite of the more extreme audiophiles, who believe ONLY their (marketing biased) perception and therefore MUST reject the measurements and the science (or pervert it, to "fit"). The Rubidium atomic clock used in the SOS article linked to by Spruce Music is such a snake oil product because the measurements demonstrate and the science dictates that the clock recovery circuitry is the defining factor of ultimate clock accuracy, not the clock generating the signal entering that circuitry. My clocking scheme probably cost around 10 times more than yours but the components which originate the clock signal probably cost somewhat less.
I did know a fellow who had extreme visual acuity.
Maybe "golden ears" has a different meaning in the audiophile world than in the pro audio world? (serious question, I don't know). In the pro audio world the term is a euphemism, most/all of those described as "golden eared" have ears which actually perform relatively poorly! What sets them apart from others isn't what their ears are capable of responding to but their ability to dissect what they are hearing and consciously identify factors which others are not consciously aware they're hearing. This takes natural ability, skill and considerable experience and most/all of those described as golden eared are therefore at least middle aged (and therefore their ears have deteriorated and are certainly less acute than those of say a normal/average child).
G