Gmorris -- interesting point. Your comments got me thinking: An electret capacitor is made by charging a capacitor and subjecting the dielectric to a high enough temperature to allow the molecules to align with the static EM field. After cooling the dielectric, a residual voltage appears across the capacitor terminals due to a net alignment of electric dipoles (net field = nonzero voltage). In the case of interconnects there is no net EM field (we're dealing with AC signals), however, one would expect that an alternating EM field would cause the molecules to vibrate in sympathy, resulting in dissipative losses. Many plastics under mechanical stress exhibit "creep", which is nothing more than the gradual shifting of molecules to minimize and distribute stresses through the bulk material. Perhaps the vibratory stresses allow the molecules to shift slightly, so that they settle into alignments least responsive to the imposed AC field(this is consistent with the idea that systems tend to settle towards their lowest-energy state). As a result of this settling in, there would be reduction in dissipative losses in the cable. This is all speculative, of course. While I'm not yet convinced that break-in is a real, physical effect (as Scrooge said of Marley's ghost, "The mind is easily fooled. You could be a splot of mustard or a bit of underdone beef. Why there's more gravy than grave about you!"), neither am I willing to dismiss it out of hand -- where there's persistent anecdotal evidence, there may be fire. Relative to the physical processes I describe above, I have no idea of the magnitudes of losses involved or the connection between such losses and possible sonic effects. Anyone have further thoughts on the matter?
