People don't argue much over design issues on pc boards. It's pretty much accepted wisdom to round off corners, keep spacing even, etc. From that point of view, there doesn't seem to be nearly as much thought given to the solder connection, at least around here. I've tried to stay away from contributing on whether or not you'll detect a sonic difference in favor of just making some technical arguments.
limpidglitch,
The wavelength of a 10 kHz sound is about 34 mm. 22 gauge wire has a diameter of 0.644 mm. Assuming the contact to be a point, if the high frequency travels along the edge, it traverses half the diameter on entering a trace, and half again when leaving it. After 26 points, that part of the signal is completely inverted relative to lower frequencies. My assumption above is that adding solder to a joint will change the geometry and have an impact on this effect. In this case, a joint isn't just a joint. In addition, the geometry can change the frequency-dependent resistance--i.e. capacitance and inductance. Needless to say, this is a very odd thing to consider in design & construction since every point of contact, whether a solder connection, or contact between materials in a component, would experience the same effect, and trying to keep track of it all is... let's just say it would be very difficult. But I have to stress that the same kinds of effects in other areas, such as corners and trace widths, are examined in the design of every pc board. Perhaps this specific one, like may others, is handled by rules of thumb about soldering and designing traces.
Again, I'd rather not make any claims one way or another as to whether any of this has a sonic impact. I only mean to point out that there are some relevant factors here that are measurable and have a demonstrable and measurable impact on electrical circuits.
As far as offering an opinion, yes I believe cables and other things can impact the sound. In particular, I have a marked preference for amplifiers that can deliver very high levels of current, and using fatter than necessary cables. The problem is, the whole subject is way too complicated. For one thing, the speed of sound isn't quite linear, and neither is a speaker, nor indeed does every sound have the same linear properties. For example, a gunshot isn't linear because it compresses too fast, causing turbulence, while a similar sound with similar frequencies but a different envelope and perhaps less energy might be more linear. This issue is complicated much further by our ears and the environment which provide even more nonlinearities. If sound propagation isn't linear, the whole point above about how the skin effect--or anything else--can change the phase is pretty much swamped by the environment. That's a good reason not to bother much about it.
By the way, my original post was meant more as a fun thought experiment--almost tongue in cheek--than a serious defense of the importance of using a particular solder. But you got that point by now. Still, there's some valid theory here, and I for one would not discourage anyone from following up on it. Over time, the line between engineering and black art moves. Perhaps over the next 100 years there will be enough advances in this area that we're no longer arguing this particular point. Perhaps one of the pioneers is a member of Head-Fi. But not to worry. Any such advance will bring with it new points to argue over.
