One of the problems with comparing the change of any part of a system to its previous state is this: our brains are not fixed-reference analytical devices. Our brains are somewhat ‘auto-correcting with drifting reference’ when it comes to sensory input. If I brought 101 audio enthusiasts to a listening room, ask them to listen carefully for a while, ask them to step out while I put in a different set of cables when actually I really leave everything unchanged, then brought the people back for a second listen, how many would say the ‘change’ made the sound better, worse or made no difference?

To really have a fair A-B (or A-B-C-etc.) comparison, one has to have an equivalent run and environment for each cable, be able to switch between cables rather quickly (about 1 second?) and have the listeners blind to which set of cables are being used at a given moment. Then one needs to run through a long pseudo-random, but carefully planned, series of selections between the cables being compared and see if a particular set is ranked consistently better or worse than another.

Then, as harp795 points out, you may not find out which cables give the most accurate transference of the signal, but what type of transference do most people prefer for a given setup/environment.

I tend to think that cables that cause the least modification of the signal are the best. If 3 of 5 compared styles that did that, then I would think that any comparison of the 3, with a large enough statistical sample, would show no preference between them. Interestingly enough, some very esoteric cable can be shown, via properly engineered testing, to alter the signal more so than simple, low-gauge, good quality copper cable.

In the February/March 2004 and April/May 2004 issues of the absolute sound, 16 kinds of speaker cable are compared with interesting results.