So says high-end cable maker AudioQuest (go here, enter the site, choose ‘Cable Theory’ and then page through to ‘2/10’ — oh I hate sites that rely on Flash Media for their entire content, rather than easily linkable HTML). ‘Necessarily’ is a weasel word. The fact is that an S-Video output implemented (on a DVD player) with no more than basic engineering competence will always outperform a composite video output. Some very well-engineered composite video processing circuits on display devices can do remarkably good jobs, producing quite watchable results, but they do this by dint of correcting and covering up the basic flaws in a composite video signal. See here to read how all this works.
Naturally, AudioQuest also offers power cables. The need for these is justified by the following:
The Challenge: The cable carrying AC power from a wall outlet adds distortion, and even the most sophisticated power supplies and filters cannot completely eliminate these upstream irregularities.
Every word of this is true. But that does not make it at all meaningful. The amount of distortion added by a few metres of power cable is immeasurably tiny, most likely even if you generated a pure 50 or 60 hertz supply with which to test it.
Of course, your wall outlet does not deliver a pure sine wave. Mine delivers power from a cable that connects my office to the house switchbox, through which it runs through a circuit breaker, which in turn is fed (via a supply fuse) from a wire hanging in the air to a power pole, which is in turn bolted to an exposed cable made of anything other than long-grain copper. This cable is also attached to thousands of electrical devices in other houses. It runs after a few kilometres to an 11,000 volt to 240 volt stepdown transformer (which, incidentally, is wired in delta configuration on the high tension side and star configuration on the 240 volts side).
That 11,000 volt line runs, after a few more kilometres of exposed wires, to a largish electrical distribution plant, where it is fed from another transformer which steps the supply down 330,000 volts. That is fed, in turn, by more than a hundred kilometres of cables hanging from large metal pylons to a hydro-electric dam which provides the bulk of the power. But at various points the grid to which all this is attached and with which it is synchronised is attached to other power sources: coal fired generators, gas fired generators, probably even some windmills out there somewhere.
And that’s only the ‘active’ cable. The ‘neutral’ cable is directly shared with three times as homes as the active cable because by using three-phase power, all the ‘neutral’ signals tend to more-or-less cancel out.
Somehow our ‘sophisticated power supplies and filters’ manage to cope with all this crap feeding into the power line, yet AudioQuest has the gall to suggest that reducing the immeasurably small amount of distortion created by a metre of cable between the wall socket and my amplifier will make some kind of audible difference. Yeah, sure. And all for just $US149.95 (one metre NRG-2 three core cable).
UPDATE (Thursday, 18 September 2003, 2:40 pm): The author of Number Watch emails:
What you say is true, but I think that you are too generous to them. They seem confused between distortion and interference, among a lot of other things. The term distortion cannot really be applied to AC power. A switching power supply works on square waves. The output of a full wave rectifier has an infinite-spectrum harmonic content, which is adequately dealt with by the tank capacitor and filter. Other sources of harmonic distortion are negligible in comparison.