Audio, but especially high-end audio comes with its own strong following, and unlike with computers, where absurd performance claims can easily be debunked by running a set of benchmarks, there are no such benchmarks in audio. That is unfortunate as many manufacturers of audio equipment fool people into buying equipment or accessories based on claims that are simply false and in many cases blatant lies. Granted, that is a pretty bold claim to make on my behalf; but rest assured that I don not make such claims lightly. Unlike many of these manufacturer's customers I have a sound grasp of the underlying physics as I took the trouble of completing a Master's degree in electronics engineering.
Take cables for example, although there is some truth to buying good quality cables of sufficient gauge instead of hooking all your loudspeakers up with telephone cable, that is about as much as there is to it. The simple truth is that resistance (R), capacitance (C), and inductance (L) per foot and the length of cable used are the only parameters that have any effect in the audible spectrum, 20Hz to 20KHz, for which these cables are used. Other parameters that could affect the signal as it propagates through the cable, such as the often mentioned skin-effect, only come into effect at frequencies several magnitudes above this range, hence have no effect at all. The capacitance and inductance of generic speaker cables is neglectable, so only resistance will play a role. Therefore a good rule of thumb is that for runs up to 10-feet a 2x1.5mm^2 cable will do just fine, up to 50-feet 2x2.5mm^2 is sufficient, and for longer runs 2x4.0mm^2 is required.
And that is just for speaker cables, do not even get me started on interlinks, and especially digital interlinks meant to connect the digital output of a CD/DVD player to an amplifier. The digital signal is comprised of 0s and 1s, and all that you should care about is that these arrive at the other end of the cable in the same order as they were sent, regardless of whether you use a short two foot cable, or a 100-foot cable reel. That is the beauty and strength of digital; it takes a lot of parameters out of the equation, the quality of the cable being one of them. Therefore 0s and 1s are inherently incapable of being affected by cables in the signal path. But how about analog interlinks? Well, in essence the same applies as with speaker cables. A good rule of thumb is to use double shielded coaxial cable, with both a wire mesh and foil and a solid copper or tinned copper conductor. These are the same cables used for TV or satellite reception and hence work well up to several hundred MHz.
So the next time you are shopping for a set of speakers cables and the sales clerk recommends these brand X cables that feature pure silver strands, 99.999% pure oxygen free copper and Teflon insulation which have gotten rave reviews in a number of magazines think twice. You can also invite the salesman to do a double blind test, where someone else switches the cables for generic 2x1.5mm^2 cables and nothing else is changed about the system, not even the volume it plays at. You will notice that neither you nor he will be able to tell the difference, and that is not because you need a more expensive system to appreciate the qualities of these cables, but simply because there is no difference, period.