I don't see any reason to buy a cable that has a serious problem with oxidation (both copper and silver) and certainly not one that costs quite a few Benjamins when the surface is acting like a semi-conductor and a diodic barrier.
Supra:
The diodes so formed between the strands are not seen by steady-state signals, but look like the plates of a high value capacitor to transient signals. This causes low-level energy storage and release after transients, that is invisible to steady-state testing yet nonetheless is perfectly audible with many music recordings.
Another incentive for choosing a cheaper cable is measurement and time domain scope pictures ->
http://www.jenving.se/?p=ply and by that I can remove one variable and just view it as it should be.. a passive device. The active devices in the chain yields greater gain in sound when going up the price-ladder, heck.. I'll rather choose EQ as tone-control than exotic headphone cables.
Having to many variables to take into consideration is for me a pest, let's say I settle with a good and expensive headphone and view an amplifier as just that, choosing a neutral one that adds as little of distortion as possible within a certain price frame matching somewhat the expense of the headphone. Then I could put those Benjamins to good use, clock-locking the transport to the master clock in the DAC with it's good and clean PSU.
Which leaves me just to worry about the plethora of analogue stages after the DAC, with an endless rolling of opamps, discrete designs and tubes for that matter, that would keep me busy for at least a couple of years.
Such a solution is more rewarding, but the best about the whole darn thing.. I can read up on it (white-papers, forums etc.) and gain valuable knowledge and experiences that can be applied and thereby makes it all more worthwhile on (in fact) many different planes.
One path to thread makes good headway and to me it makes that more sense directing ones effort to concentrate on the tangible principles, but your mileage may vary.. of course.