<font color='#000000'>Let's get to first principles here:
The design of any audio gear, let alone cables, is purely an engineering problem.
Engineering is based on science.
Science tells us that the only meaningful parameters affecting cable performance at audio frequencies are inductance [l], capacitance [c] and resistance [r]. The length and gauge (thickness) of the wire are the main things that account for differences in those three parameters. Exotic materials like silver conductors and Teflon insulation (dielectric), or the shape of the cable's cross section have no audible effect at audio frequencies.
In practical terms, any reasonably well-made set of RCA interconnects of normal lenghth for domestic use is low enough in l, c, and r to function properly, i.e., to have no audible effect (and minimal measurable effect) on an audio signal. Same for plain ol' copper zip cord speaker wire from about 16 ga. on up. If you're doing long runs, all you need is fatter wire, like Zipper says.
Skin effect (often cited by vendors) is a real phenomenon but just barely begins to play a role at the very highest audio frequencies, if any. For most of us over 30, those highest frequencies are out of range anyway (I'm 50 and feel lucky that I can still hear up to 15kHz). In any case, even the most acute human hearing is not maximally sensitive at those very high frequencies above 10kHz or so and the small differences (a small fraction of a dB) between cables having reasonably low l, c, and r is unlikely to be audible. Skin effect is only a real consideration in the design of cables that carry radio frequencies (i.e., transmission lines for antennas) and other very high frequency signals.
Everything else is B.S. as far as audio goes. This is not to say that other paramaters affecting cable design do not exist for applications such as RF transmission or high speed data transmission, but just that they are not a concern at audio frequencies.
Some very expensive cables do indeed sound "different" (but not necessarily better) because they are high in l, c, or r and cause frequency response anomalies. In other words, they muck up the sound by functioning as a non-adjustable tone control! Specifically, many of them roll off the high frequencies. Why anyone would want to do so for all program sources, all the time, is beyond me, but you can do the same thing with a $0.50 or so resistor wired in series at your speaker terminals.</font>