You've reached the new home for Jay Rose's “dplay.com” articles and tutorials!

Dr. Barry Blesser, one of the fathers of modern digital audio,
at the AES/Boston Annual Banquet, June 2000

Blesser Headshot
Introduction Jay Rose 1:57 Stream & Download (231kB)
Speech* Dr. Blesser 22:02 Stream & Download (2.5mB)
Q & A   membership   34:20 Stream & Download (3.9mB)
* - © 2000, Dr. Barry Blesser

Our banquet speaker, Dr. Barry Blesser, offered a preview of some papers he'll be publishing in the AES Journal... and a challenging view of our future.

Dr. Blesser has been a leader in the development of digital audio. While a professor at MIT, he designed the EMT 250 -- the first practical digital reverb. He wrote what is considered the seminal papers on using DSP for audio, participated in the birth of Lexicon, served the AES as President, and received its Bronze, Silver, and Governors' Award. His latest venture combines engineering, management, and historical disciplines to help companies cope with technical change.

In his talk, Dr. Blesser showed how the audio industry is no different from any other. But within that simple statement is a bombshell implication: The world has changed in the past decade or two -- technology is becoming just another commodity, with mass distribution the predominant business strategy -- and it may be that few audio manufacturers or engineers will survive the shift.

The best model of this new orientation is the software industry, where there's virtually no manufacturing cost; all of the investment is up-front. If a program cost a half-million dollars to develop, and you sell ten thousand units at $100 each, you make a profit. But sell only five thousand units, and you're out of business. For these companies, maximum sales following tightly-controlled development is the only business model that works. The design goal is, inevitably, "give the customer just enough to motivate a purchase".

Audio and many other industries have typically followed a very different path. A technology would first show up in the highest-end professional products, hand-assembled and sold to a handful of users. Then it would be developed into general studio or broadcast versions for a somewhat larger market, and high-fidelity versions for the music lovers. Finally, if warranted, mass-produced consumer versions would appear. Digital reverb has been one example of this... but so has mechanical reverb (remember spring-based "spatial enhancers" for car radios?), the tape recorder, and even the class AB tube amplifier.

Dr. Blesser calls this a standard development thread. Previously, a designer engineer could "ride a thread" for their entire career, becoming an expert early on and selling that expertise when the product reached consumer levels. Occasionally -- such as when reverb switched from mechanical to digital -- an engineer would have to jump threads and learn a new technology.

But, he maintains, computer-based technologies have tied standard threads in knots. Products are conceived and designed with mass markets in mind, and the professional or studio version is strictly a spin-off. The compact disc was the first such technology. DVD is another. Because of their mass-quantity scale, these technologies tend to draw heavily on concepts and hardware from the computer industry. Even in an audio product, the contribution of individual audio designers may be very slight. Consider the portable CD player: with all of its analog circuits on a pre-engineered chip set, creating a new product might not require an audio engineer at all.

Consumers tend to ignore this "thread inversion" because the overall product category doesn't change. To a music buyer, compact discs are merely a smaller, quieter version of the vinyl recording. To a radio announcer, a station's touch-screen hard disk player is just an instant-cueing, jam-free cart machine. But to the industry, the effects are earth-shaking: How many turntable manufacturers made the transition to CD players? How many cart companies are sill in business?

The result, in audio and other industries, is the disappearance of a mid-level market where mid-sized companies and moderately-talented engineers can be successful. There are niche manufacturers, hand-assembling expensive products for aficionados; and there are mass producers. But there's nothing in between, virtually no way for a small company to ever get big, and very few threads left that an engineer can ride for a career.

[Dr. Blesser pointed out a current exception in the audio industry. There is an unquenchable thirst for entertainment, which by its nature has to be custom crafted by talented engineers along with performers and writers. But this, too, may change with new technology.]

The other result is a shift in the definition of "quality". When the goal of a mid-level audio manufacturer was acceptance by professionals, product quality could be defined in technical terms. You could measure it on a meter. Now that the goal is shifting to mass consumption, quality is whatever the consumer considers a good deal: price and availability of program content count more than those last few percent of THD.

Engineers -- even MIT-trained ones like Dr. Blesser -- may find it hard to survive using the old models.

Needless to say, a vigorous question-and-answer period followed. Some addressed the future viability of the AES, as audio manufacturing is swallowed up by a few giant companies. Others held there'd always be a natural desire for technical excellence, even if it isn't in the business plan. In all it was an exciting and challenging evening, concluded with the Boston Section presenting a plaque to Dr. Blesser, commemorating his contributions to the industry and the technology.

--Jay Rose, July 10, 2000

11