A previous post featured a short film about members of the Audiophile Club of Athens and the rather extreme sound systems their members have created. Some members spent in excess of $300,000 to build their systems. You may be wondering just what manner of gear that sort of money would buy, and would it really sound that much better than a more modest (yet still comparatively “high end”) system of say, several thousand dollars. Before we go any further, let me state that in no way am I making fun of the way people spend their money. Heck, I’ve been known to drop some coinage on musical instruments and Kevlar kayaks, things some people find frivolous. No, I’m just interested in whether this gear is sonically superior or simply audio woo.
Normally, these sorts of questions can be illuminated through the use of double-blind testing of components. If you follow high end audio though, you know that aficionados often complain that something as simple as the switching networks employed introduce too much coloration. As there are numerous variables that interact, let’s just look at a single element: the loudspeaker cable.
There are two general schools of thought regarding cabling. The first contends that there is no sonic difference between cables and the second that there may be huge differences. The first group is happy to lay out some 20 gauge zip cable and have at it, while extremists with deep pockets in the second group may fork over literally thousands of dollars for their cables. You read that correctly.
I think both groups are being unrealistic. It is easy to show the limits of the first group and a practical impossibility to verify the audibility of supposed limits for the second group. A good overview of the some of the issues can be found in the recent series from Audio Design Line covering amplifiers and cabling. For example, while it is true that the power loss in a typical length of loudspeaker cable is minimal, it is relatively easy to show its effect on the damping factor of the system. (A poor damping factor will lead to indistinct and “muddy” bass). The 20 gauge zip cable is not going to cut it.
Of course, an audio cable is not just a simple DC distribution system. At higher frequencies, the impedance of the cable can vary a bit from its simple DC resistance value. Now any well designed amplifier will be able to deal with the impedance variations produced by typical cables, but some of the esoteric designs are not so robust:
In terms of professional use, a marginally stable amplifier has little to justify its use. In fact, a marginally stable professional amplifier is almost an oxymoron – a contradiction in itself. However, in the world of hi-fi, some designs exist which are so esoteric that practicality and justifiable engineering are not high on the list of design priorities.
Some of these are more works of art than works of science, and they are designed to be pampered and appreciated rather than to be bolted into a rack and forgotten about. Nevertheless, without doubt, some of these specialised hi-fi amplifiers do perform extremely well under the limited circumstances of their intended use, but some are so minimalistic in their design, (even if not in their price) that they can sometimes be only marginally stable (although very few), and often their unbalanced input circuits and high sensitivity (100 mV as opposed to 1 volt) make them more prone to disturbance than the less sensitive and often more robust professional designs.
The article concludes that a so-called esoteric design could be adversely affected by a cable impedance variation that a more robust design would take in stride. Thus, it is quite possible that swapping out cables may lead to sonic differences, but I’d have to ask if this isn’t really just an issue of a poorly designed amplifier rather than a sonically superior cable. In part five the authors discuss impedance plots of cables. While it is clear that there are measurable differences between cables, this does not mean that there must be sonic differences. For example, it would be trivial to alter a 100 Hertz sine wave input to an amplifier from 100 millivolts to 99 millivolts using a quality digital multimeter, but the output level reduction would be beyond human hearing acuity (down less than .09 dB). Of course, the converse is also true. Just because things measure identically does not mean that they will sound the same (after all, you may have performed an incomplete series of tests). Unfortunately, some of the graphs are nearly worthless (note that some of the Z plots do not include a calibration for the amplitude scale).
One thing is for sure, and that is that all parties agree that the shorter the cable and less complex the loudspeaker impedance, the better. Part six discusses multi-cabling but some of it I just find plain goofy. The basic idea is to use separate cables to the various components (woofer, tweeter, etc.) rather than a common cable. It seems to me that the esoteric audio folks could learn a trick or two from the sound reinforcement and recording studio monitoring people and just chuck the whole concept of passive loudspeaker crossovers in favor of up-stream active crossovers and multiple amplifiers. This topology, while expensive for the average home listener, overcomes numerous problems with single amplifier designs, and seems to be a natural for those with deep pockets (the authors mention this topology but should have discussed it in greater detail).
“But amplifiers are expensive” I hear people say. Sure, but if they’re designed as part of a loudspeaker system it won’t be as bad as going out and buying individual amplifiers off the shelf. Besides, just how much are those loudspeaker cables anyway?
Let’s start with Stealth Audio Cables. As you can see, their top of the line Dream cables are $11,600 for a three meter length. You can buy some pretty serious amplification for that kind of change. You might want to peruse their site for “technical articles” on their cables. You might also be interested in an overhead view of their facility (zoom in and select aerial view). Not to be outdone, Transparent Cable offers their Opus series. A 10 foot cable will set you back a mere $34,000.
As crazy as these prices are, I get an even bigger kick out of the analog interconnects. Analog line-level cables connect one component to another and deal with much lower power levels and “nicer” impedances than loudspeaker cables. The Indra RCA interconnects are $11,500 for a two meter length, a relative bargain compared to the two foot Opus cable at $12,000. But wait, it gets better. What about the AC power cord that your amplifier or other gear use to hook up to the wall socket? Once again, you can get a Dream series three meter power cable for just $4200. I can’t help snicker when someone writes about how important a quality AC power cable is in order to insure good power delivery to audio components. Did they forget that on the other side of the AC socket there might be 75 feet or more of ugly Romex power cable leading back to their house’s AC entrance?
In closing, it seems to me that if you really want to hear what the artists and producers had in mind when they recorded a piece of music, you might want to consider looking at the sort of gear that they used, and that would include the single most overlooked piece of the home audio playback chain, the room.
(Some commentary on general audio woo can also be found at Randi.org.)