Originally Posted by VirusKiller
Um no, completely incorrect
Perhaps I didn't explain myself very well (or maybe you didn’t understand me)... it's been a long week.

Originally Posted by VirusKiller
For SPDIF and SpeakerLink (which uses a form of SPDIF), there is a single, ideally square-wave, binary signal with no predefined clock rate and biphase marking to indicate the bit transitions.
While there is no *predefined* clock rate (in much the same way as Amex has no preset spending limit), there is always a clock, the rate being set by the transmitter. The whole point of biphase is to allow the clock to be embedded into the signal (Wikipedia says about self-clocking signals such as biphase encoding: 'If the embedded clock signal is isochronous, it gets sent simultaneously with the data.', which is what is happening here).

However (unlike ethernet), it does more than just allow the transitions to be read. Every D/A converter needs a clock and I'm 99% certain on DSP speakers, that the clock that drives the D/A convertors is recovered from this timing signal. If the speakers had their own clocks, you'd have at least three clocks, all drifting apart.

Originally Posted by VirusKiller
True, degradation of the signal can occur. The worst form of this (according to Richard Hollinshead who invented MHR to mitigate the problem) is non-75ohm connections leading to reflections in the cables which soften the ideal square waveform, reducing uncertainty in the exact timing of the transitions - this is jitter. MHR, by randomizing the data, removes correlated jitter (the most audible form), but not uncorrelated jitter.
Exactly, this was pretty much the point I was trying to make (except I was focusing more on the cable capacitance and inductance that alters the signal rather than reflections caused by a change in impedance).

Originally Posted by VirusKiller
The question remains as to whether vibrations and RF interference can cause a meaningful (i.e. audible) difference at the transmission frequencies involved. *Personally*, I call BS on this as I think that - if there is an effect - it’s going to be a 2nd (1/10th) or 3rd (1/100th) order effect compared to the primary problems inherent in SPDIF; naturally it follows that I believe that certain manufacturers are having a laugh at their customers’ expense.
This wasn't a point I was trying to make (at least in the message you replied to). From my perspective, I wouldn't rule anything out, but neither have I heard a conclusive difference (yet). I can see how it might make a difference with components such as the Shunyata power conditioners, by the very nature of how they work.

Prolific music listener / Hi-Fi addict!