January 2008

First Things First

  I've just returned from the fourth annual Rocky Mountain Audio Fest (full report next issue; highlights on-line now at AVguide.com) and was struck by the sheer number of loudspeaker manufacturers I'd never heard of showing new products. I must have encountered more than a dozen new companies (at least they were new to me) exhibiting loudspeakers. (Added: Full RMAF show report on Enjoy the Music.com can be seen by clicking here).

In fact, there are more than 400 loudspeaker manufacturers who can be considered competitors in the high-end arena. The vast majority of these companies buy the same raw drivers from the same driver manufacturers, put those drivers in MDF boxes, and, with different levels of skill, create crossover networks. The results, as you might expect, vary widely. I saw and heard a number of loudspeakers that seemed to violate generally accepted principles of good design practice — and most of them had extremely colored tonal balances. A few managed to do some things well — dynamics, for example — but one had to listen past the tonal distortions. This phenomenon of a product being optimized for a single performance parameter is driven by designer "tunnel vision." The designer values one aspect of the sonic presentation so much more than other aspects that he optimizes the design to emphasize his priority to the exclusion of all else. That's fine if you happen to share the designer's idiosyncratic view but not so good if you're looking for a product that delivers balanced performance.

A related trend, also on vivid display at the Rocky Mountain Audio Fest, is the use of ultra-tweaky products and techniques for optimizing sound quality in a system that starts with gross loudspeaker-induced frequency-response errors. I'm not arguing that you can tell a good speaker from a bad one by looking at a frequency-response graph. Far from it. But if listening to instrumental timbres becomes the aural equivalent of looking at the instruments in a fun-house mirror, you know something is wrong that cable elevators, cones, dots, and CD treatments won t fix.

As evidenced by the Letters section in the past three issues, many readers are skeptical of the efficacy of some of these tweaks, and even of interconnects and loudspeaker cables. Many tweaks — some of which are improbable and seem to defy the laws of physics — do indeed improve the sound of a stereo system. Others are pure snake oil. But one must be careful in labeling a product a hoax purely on the basis of the product's compatibility with established dogma. The truth is that some aspects of audio design are a black art; there is no formalized method for inventing and evaluating discoveries. Rather, the designer, guided by some vaguely understood principles, often stumbles on a device or technique that demonstrably improves the sound, even though he cannot explain the mechanism behind the effect. When that happens, one of two unfortunate scenarios unfolds. The first is that the designer ascribes a "false interpretation" to the effect. He's not quite sure how it works, but comes up with an explanation that seems to fit the facts — an explanation that can be completely wrong.¹

The second scenario is that the designer (or more likely, the company's marketing department) invents a scientific-sounding theory of why the product improves the sound. These pseudo-scientific "explanations" seek to legitimize the product by couching the product's effect in scientific terms that usually involve some combination of quantum theory, electrons, and vibration. The company thinks the product won't sell if they 'fess up and say that the product works, but they don't know how or why.

However, just because the explanation is absurd doesn't mean that the device doesn't improve the sound. We have much to learn about audio, and many important phenomena wouldn't have been discovered if high-end designers were forced to conform to existing viewpoints.

I'm reminded of a quote from the great audio thinker Richard Heyser: "One of the most belittling experiences is to deride the 'black art' of a craftsman who gets consistent results by a certain ritual which he cannot explain and then to discover that his actions in fact held a deeper technical significance than we understood from our simplified model."

Some aspects of audio involve black art — but the fundamentals of good loudspeaker design isn't one of them.

1. The impact of "false interpretation" on the history of science is examined in detail in Michael Poianyi's Persona/Knowledge: Towards a Post-Critical Philosophy [University of Chicago Press, 19581.