Voodoo vs. Snake Oil – Audio Musings by Gary Alan Barker
Anyone who has spent a few decades in the audiophile community will know that High-end audio is 50% Voodoo and 50% Snake Oil, and it is the job of the reviewer, and to a lesser extent the audiophile, to determine which is which. Any observable phenomenon must have a reason, whether or not that reason is obvious, or indeed the reason put forth by the person pointing out that phenomenon. There is, of course, the question of whether that phenomenon is truly observable or if it is an illusion brought on by suggestion.
Blind AB tests, or if you prefer Double-Blind AB tests (preferably where the person conducting the test, doesn’t know whether the switch is actually taking place) are an excellent way of determining whether an observable phenomenon actually exists. What they won’t tell you is whether that phenomenon is an improvement or detriment.
This is where the trained ear comes in. Like wine tasting, not everyone has the skill and ability to judge audio quality. Several factors go into creating a trained ear. Preferably you start with better than average hearing, repeated testing by groups like the AES has proven that only a small percentage of the people can actually hear the differences in certain aspects of audio reproduction. (A case in point, a few years ago, at a high-end amplifier company for which I was the technical services manager, a processor came in with complaints of making a high pitched noise. The repair tech claimed that there was no noise, but when an associate and I listened to it, we could clearly hear the noise. This prompted the tech to connect the unit to an analyzer, where he discovered a fluctuating tone between 20 and 25 kHz.) Exposure to live orchestral and acoustic music is essential in properly training the ear, as it gives a real-world reference to judge from. This is the reason for the popularity of using classical audiophile recordings when testing audio gear, for without a solid reference, how can you determine if what you are hearing is correct. (This is also why we often like to test only one component at a time in a reference system that we already know the sonic aspects of.) Mentoring is also an important step in training one’s ear; if you don’t know what to listen for, you will most likely miss it. And possibly most critical, is lots of listening. The more different equipment you listen to, the more you learn the individual characteristics of sound each piece produces.
Of course, personal knowledge and research are also necessary to sift the Snake Oil from the Voodoo. While listening needs to be performed with a completely open mind, the explanations as to the cause of the effect should be taken with a grain of salt. Take the example of the green pen, the disc stabilizer, and the black CD. All of these were technical band-aids aimed at reducing read errors in CD transports. Read errors are normally handled by error correction, a process that determines what the missing data is by postulation based on the previous and following bits of data. Error correction by its nature is a loss of fidelity. The green pen is fairly straight forward: A CD is basically a clear polycarbonate disc, with a thin layer of etched aluminum attached to it, making the disc itself a prism that refracts ambient laser light in all directions. Green is a filter for red light, and CD transports use red light lasers. By reducing the ambient laser light inside the transport, read errors are reduced. Disc stabilizers were designed to compensate for the fact that CDs were often poorly manufactured and out of balance, causing them to vibrate, which in turn caused read errors. The drawback to disc stabilizers is that they made the CD too heavy, adversely effecting the transport’s speed control, occasionally going so far as to damage the transport. Most high-end transports incorporate some form of disc stabilization into their design, making aftermarket stabilizers obsolete. The black CD works on the same basic principle as the green pen. All of these work pretty much as advertised, so where does the Snake Oil come in? The proponents of the black CD, recommend ripping a standard CD to a computer then burn it to a black CD to improve the sound. The logical inference being that the $30 CD drive in the computer is better able to read the CD, than a $3000 CD transport. And there is good reason for this. The high-end CD transport manufacturer approaches the design of their device from the point of view of an audio component, so they use conventional audio techniques of better isolation, better stabilization, and better quality electrical components, while the manufacturer of the computer CD drive approaches the problem from the point of view of a data storage device, that needs to read the data at 50 or more times the speed of a CD transport. Of course, back to the Snake Oil, at least 1 high-end CD transport manufacturer saw the wisdom of basing their unit around a $30 computer CD drive.
As I mentioned before, not all change is improvement, and careful long term listening is required to determine this. Case in point TIMD (Transient Intermodulation Distortion): In blind AB tests, people tend to choose amplifiers that produce high levels of TIMD. The reason for this is that TIMD gives the illusion that you are hearing something extra, that you don’t with the amplifier that produces low levels of TIMD, which of course you are, you’re hearing the TIMD. In prolonged listening, TIMD causes extreme listening fatigue, and a trained ear tells you that what you are hearing is not part of the music, or at least, not part of the natural sound of the instruments (obviously TIMD can be introduced in the recording process, and may account for a bright harshness in some recordings).
The other side of Snake Oil is that even if the promoter’s premise is correct, it doesn’t necessarily follow that their conclusions are valid. As an example, I cite my introduction to high-end cable. As an audiophile, I am quite the skeptic, and when people started pushing expensive designer cable, I wasn’t buying (figuratively and literally). As an electronics person, I could see that heavier cable carries more current, but given a sufficient gauge wire, I wasn’t willing to believe that it could otherwise affect the sound. So one day, a friend gave me a set of rather expensive speaker cables (produced by one of the most popular cable manufacturers), insisting that I test them out, in comparison to what I was currently using (12 AWG zip cord). After extensive listening, I determined that there was indeed a difference, it sounded worse (which did mean that cable was as much a component as the electronics and lead me to do a great deal of research into the subject of cable). The manufacturer’s white paper on the subject, claimed that different frequencies, traveled at different rates down differing gauges of wire (hence their cable was made up of several different gauges), which described exactly what I was hearing, the sound was bloated and spongy, totally lacking phase coherency and linearity of bandwidth, which you would be what you would expect if the same note was reaching the other end of the cable at a multitude of different times based on it passing down the different strands at different rates.
So, in the long run, Voodoo and Snake Oil work hand in hand, the Voodoo is the magic that makes one product sound superior to another, while the Snake Oil is the marketing that brings it to the public, and the reviewer is there to hopefully point out the relative value of both.