Yes I can hear differences between some brands of speaker cables and interconnects.

I was disappointed in the first article referenced above by Larry (Roger Russell's informative writing about this subject for McIntosh).
The problem is that a significant issue regarding wire and cable design was overlooked.
The issue is that a capacitive load on the output of most amplifiers and preamplifiers slows down the output current amplifier section, causing a large high frequency correction peak to arrive at the summing point of input and feedback in the device, often arriving late. This causes the voltage amplifier to both attempt to generate a correction signal that overloads the circuit, or, in a bad case, generate a high frequency spike and decreases overall stability as the negative feedback actually goes positive.
This can be easily observed on the test bench watching how the device under test behaves. Square wave response goes to flanders as the capacitive load increases.
In some cases the output signal still looks good, because a slow output current amplifier section can actually roll off the overstressed signal from the voltage amplifier section and hide the results from the test equipment. However if you move the test probes to the output of the voltage amp section then you can see all the damage before the current amplifier has the chance to hide it.
Way too many "testophiles" overlook the fact that the amp or preamp is not a black box. It is a voltage amplifier connected to a current amplifier (even the insides of a linear IC have this construction - its just that you can't get inside them to see how badly the signal is being abused).
Anyway, the point is that an interconnect cable or speaker wire of a construction type that trades off inductance for capacitance can easily present a load to an amp or preamp that significantly increases the distortion of the musical signal. A good example is using a multi-strand "woven or braided" type cable, which almost always will significantly increase the capacitive load, a very bad idea.
A worse case example of this occurred several years ago. Threshold came out with a new series of amplifiers, and they eliminated the normal output inductor from the design. Common good engineering practice is to use this element to protect the amplifier from the effects of an adverse capacitive load. The signal rolloff caused by this is way way way out of the hearing range, with a pole point no lower than 100K Hz or so. Threshold assumed that the inherent series inductance of a speaker wire would be enough protection, so that the output inductor would be redundant. Bad assumption.
At the same time Polk audio sprang their very fancy Cobra Cables on the audiophile world. These were multistranded braided very fine speaker wires, beautifully done, with hundreds of interwoven strands of tiny wires. This essentially eliminated the dreaded inductance and instead presented a huge capacitive load to the output of the amplifier.
Unfortunately, connecting the output of a Threshold amplifier to the speakers with Cobra Cables caused the amplifier to go into full bore oscillation and self destruct! Could anyone then hear the difference! Of course. With new Cobra Cables, now you had no sound at all. Actually you could even smell the difference

This of course was a worse case situation. Consider however that even with a smaller capacitive load, and even with an amplifier that could better tolerate that, the musical signal was still being damaged long before the problem was bad enough to completely kill the amplifier.
Capacitive loads modify the operation of an amplifier and or preamplifier! The modification has a name - - - DISTORTION!
The effects may be subtile, but are almost always audible. One thing we can see on the bench is a leading edge spike on a square wave generated by the capacitive load. We suspect that audiophiles hear this as "increased detail". Actually it is the same effect as turning up the "sharpness" control on you TV set. This generates a leading edge spike on the transitions between dark and light sections of the picture. A little bit can even make you think you are seeing more detail, too much and it is obvious eyestrain. They same is happening with the transient spikes a capacitive load generates in your audio components. A little bit may actually fool you into believing that the "increased detail" you hear is a good thing. However, it was not part of the original musical signal input.
So, can I hear the difference some speaker cables and interconnects generate? Sure I can. I can hear the garbage that is not supposed to be there. No problem. Would I pay extra to get this nifty garbage? You got to be kidding.
We design our amplifiers and preamplifiers to be as tolerant of capacitive loads as possible. This is the reason some people think they are not good enough because they cannot hear significant differences in cable brands when they play the magic cable game. They assume the components do not have the resolving power to differentiate the differences. Actually they simply have such good stability in the face of a capacitive load that they are not distorting and are playing the input signal properly in spite of the abuse. Oh well.
To repeat our recommendations. 14 gauge speaker wire from Home Depot terminated in dual banana plugs and their $5 a set interconnects are just fine. Paying more is simply buying into fraud. If you absolutely have to have premium wires and cables to feel good, look into Bluejeans Cables
http://www.bluejeanscable.com/. At least they are built to good engineering standards.
Regards,
Frank Van Alstine