The Well-Tempered Synclavier

Areal-time, digital synthesizer that is portable and costs under $15,000? Anyone familiar with electronic music, who knows what all that means, is bound to say that it doesn’t exist. And, to be sure, it didn’t before last July, when the combined five-year efforts of Jon Appleton, Sidney Alonso, and Cameron Jones resulted in the Synclavier, propelling the technology of electronic music a vast leap forward.

The electronic music world learned about it only a few weeks ago, when music professor Jon Appleton delivered a paper outlining the characteristics of the Synclavier at the 1977 International Com- puter Music Conference in San Diego. Since then the leaders in the field have been making their way to New England Digital Corporation in Norwich, Vermont, and plunking down their money. Thirteen thousand dollars for a synthesizer, com- puter and keyboard ready to plug into your own audio system ... up to $15,000 if you want extras.

What does it do? Well, if you want a trumpet, you just play a note on the keyboard, dial the knob and push buttons until you’ve designed a trumpet sound, then enter your trumpet in the computer’s memory and play. Because of the time- varying harmonic qualities available from the specially designed digital oscillator cir- cuitry, you can get the sound of a “real” trumpet or of any other instrument. The computer will store up to 16 instruments and give them back instantly. If you want more, wait about two seconds and the computer will retrieve them from a magnetic storage disc, in batches of eight instruments at a time if you want.

With your orchestra ready, you start playing the keyboard, which is like that on any standard 64-note piano. If you have just played a trumpet solo and want the notes repeated as a bell, just press a but- ton. The computer remembers the sounds as you play and will give them back in- stantly in the form of any instrument in its memory. If you want to vary the characteristics of the sounds, press a but- ton, dial the knob, and create whatever you want. By playing over something already recorded, you can build up your orchestra to any size.

The explanation of how this marvel operates lies in a vast array of numbers and voltages that are rapidly changed into three different electronic languages between the keyboard, computer, and syn- thesizer. The Synclavier employs several breakthroughs in computer-driven syn- thesizer technology, several of which are being patented. The software or method of programming the computer is among the most flexible and advanced in the business. The methods used in solving the problems of synthesis are equally in- genious and include an interesting form of simplification that could be called “faking it.” The 16-bit computer a bit is a binary digit, and 16-bit means the processor or computer is accurate to five decimal places is, in itself, an extremely versatile and sophisticated instrument with numerous applications in the sciences and medicine.

But why bother?. Why not leave well enough, in the form of handcrafted Stradivarius violins and intricately machined horns, alone? Composer Appleton’s answer is that, day or night, with the flick of a switch, he has an in- finitely patient orchestra that remembers everything he tells it.

Appleton, a Reed College graduate who earned his master’s degree in music at the University of Oregon, began his career in electronic music in the early sixties, work- ing with tape. This method of recording natural sounds and cutting and splicing them into musical compositions, the earliest form of electronic composing, originated in France in 1945. Making music with electronic machines by syn- thesizing began in the early sixties and became widely known with the release of the album “Switched On Bach,” the biggest-selling single album of classical music ever.

The Moog synthesizer, the instrument used in producing “Switched On Bach,” is one of at least 50 such instruments on the market today, all of which have at their heart analog oscillators.

Appleton was never satisfied with analog systems even though he used them, in addition to taped natural sounds, in his musical compositions until 1973. He has made three albums of electronic music since 1969, each of which has won an award from ASCAP, the American Society of Composers, Authors and Publishers.

The major problems with analog systems are, in simplest terms, that the tuned os- cillators drift out of tune with time and temperature changes and, in addition, can’t produce sounds with the time-varying har- monic content of natural sounds. Not that the industry isn’t trying. According to a re- cent issue of Synapse, the electronic music magazine, “Sennheiser has released the VSM 201 Sound Effect Vocoder. The unit features twenty channels of filters with a spectral range of 100 Hz. to 8000 Hz. Stevie Wonder is rumored to have purchased the prototype. The unit lists for $10,800.”

When the string of a violin is plucked, the sound resonates inside the instrument case and all sorts of pleasing things called harmonics and partials are generated. As the note decays with time, the sounds vary. On an oscilloscope, a television-like instru- ment that can picture sound waves, natural sounds look extremely complex and im- perfect. The harmonics accompanying each note are scattered all over the screen.

In 1972, Appleton began to talk with Sidney Alonso, a graduate student at Thayer School, and Cameron Jones ’75 about digital synthesizers. Appleton liked what he heard.

The earliest digital synthesizers were put together at the Bell Telephone Labora- tories in New Jersey, in the early seventies. Because of certain characteristics that result from frequency modulation of wave- forms generated by digital oscillators, it is fairly easy to simulate natural sounds with time-varying harmonics. Digital systems make representations that are composed of many little bits of information or numbers which, when seen together, seem like a smooth flow. Because so many little mea- surements are needed to simulate a natural waveform, it takes a digital computer to control a digital synthesizer. An advan- tage of digital systems is that they operate very quickly, so by rapid sequencing of many independent calculations, these figures appear to be occurring simulta- neously. Thus, one digital oscillator cir- cuit can do the job of 16 analog oscil- lators. At the heart of a digital synthesizer is an extremely accurate quartz crystal resonator which provides a reference fre- quency for several oscillators. Such digital systems can be used for measuring, or sampling, natural waveforms and repli- cating them with a great deal of accuracy.

With financial help secured by President Kemeny, Appleton, Alonso and Jones began work on a digital system that could be used for musical teaching, composition and performances. There was no such thing available commercially at that time, nor aside from the Synclavier is there now.

The first prototype, developed in 1973, was a music instruction system tied to the Dartmouth Time Sharing System, which was then powered by a Honeywell 635 computer at Kiewit Computation Center. The system included a commercial 16-bit minicomputer, a digital synthesizer designed by Alonso, the DTSS via a telephone hookup, and an ordinary com- puter terminal with a set of earphones added. Jones and a team of un- dergraduates wrote the software. A student wishing to practice his or her music lessons merely talked to the com- puter by typing a request for a certain lesson program and answered the questions the computer asked, which in- cluded the identification of musical tones that came through the headset. In addi- tion, by typing instructions to the com- puter, students were able to compose and play back their compositions, testing their skills without having to gather together several musicians and their instruments.

The system was so successful 300 music students used it the first year that more terminals had to be added, and Alon- so set out to improve the quality of the syn- thesizer. Other schools became interested, and in 1973 the team received part of a grant from the Sloan Foundation to create an exportable system, not tied to a large time-sharing computer, costing about $20,000.

Francois Bayle, director of the Groupe De Recherche Musicales in Paris, where electronic music originated, came to Norwich a few weeks ago and spent the better part of two days playing the Synclavier. In the end, he purchased what is, in many ways, a beautifully crafted in- strument embodying creativity, simplicity, and ingenuity. The Synclavier’s several capabilities mean that the thing can be run by playing notes on a piano-like keyboard soon to be keyboards rather than by typing instructions through a computer terminal. Its “real time” operation, made possible by having a computer right inside rather than down the street, gives one the advantage of instant recall and makes live performances possible. Large time-sharing computers, by contrast, have the disadvan- tage of turn-around-time, which occurs when an operator has to wait while the computer shares part of its capabilities with someone else.

There are no digital synthesizers presently on the market, and not even the custom synthesizers built in various sound labs are combined with a portable 16-bit computer. According to Alonso, the average large computer is designed for use by a bank. “The big computer companies are locked into old designs that aren’t flex- ible enough,” he says. “Banks, all banks, make the same calculations day in and day out.” Large commercial computers generally have large circuit boards, mak- ing it difficult and costly to add a specific new function. In addition, if you happen to make a mistake working on a large com- puter, the repair costs can be phenomenal.

“In the computer business, the maxim is: Don’t bother to design your own processor,” says Alonso. “But we did it anyhow.” The result is a black box, 19 in- ches long, 12 inches wide and 5 inches high, with small circuit boards containing “mature” parts components that have been around for a while and are being produced relatively inexpensively operating at low voltages so a short will not prove disastrous. A lab technician us- ing Bliss, the computer’s trademark, can easily interrupt his experiment, redesign it, add or subtract components from the com- puter, and proceed with very little lost time and at low cost.

Although Alonso and Jones had nothing to do with it, one of the major breakthroughs that made Bliss and hence a relatively low-cost Synclavier possible occurred in computer disc technology in 1976. Before that, a moving- head disc memory system costing about $lO,OOO was the method used to store and retrieve information from a computer. It is now possible, for certain relatively small- scale applications, such as the Synclavier’s memory system, to replace a moving-head disc with a “floppy disc” system, which costs some $2OO.

The second digital synthesizer system developed in 1974 for Dartmouth music students eliminated the DTSS computer and employed a moving-head disc memory system with a commercial mini-computer, a digital synthesizer, three remote ter- minals, and a master studio terminal equipped with a primitive 64-note keyboard. Despite the fact that they weren’t satisfied with the mini-computer which ran the system and were anxious to design their own, Alonso and Jones spent most of their time over the next two years working with Appleton to develop im- proved software and a more sophisticated digital synthesizer.

The way the human brain responds to bits of sound is a characteristic that allows Alonso and Jones to pull off a sleight of hand in the synthesis of sound. Although it’s not entirely clear, Alonso explains, what goes on in the brain when the ear sends signals to it, we have no trouble recognizing or responding to such bits such as the leading edge, or first one tenth of a second, of a sound indicating danger. Taking advantage of this phenomenon, Alonso and Jones have instead of ask- ing the synthesizer to recreate the entire sound wave of, for instance, a trumpet note invented some techniques that get the job done by using less of the capacity of the instrument to synthesize one particular characteristic of the note, leaving it free to work on simulating other characteristics.

By the summer of 1976, Alonso and Jones had started New England Digital Corporation. They had a digital syn- thesizer they were fairly happy with and were perfecting the design of their own 16- bit mini-computer. Then a strange thing happened. That July Jon Appleton decided to take a job as director of the Stiftelsen Electronmusikstudion Stockholm, the Swedish National Center for Electronic Music.

“I had worked there before, as guest composer in 1970 and 1*971, and the job offer was a real good one,” says Appleton. Instead of proving to be a major step in Appleton’s life, the job in Sweden is a foot- note in the development of the Synclavier. He was back in Norwich by last January, working full-time with Alonso and Jones.

The job that remained was to turn the black boxes, the synthesizer and the com- puter into a' musical instrument. Essen- tially what they did was to eliminate the typewriter as a means of talking to the computer and replace it with a knob and some buttons mounted near the keyboard.

Cameron Jones, during his student years at Dartmouth, was known as a “Kiewit Jock.” He hung around the computer center with a devotion that another student might have reserved for the basketball court. He talks in a very concise and logical manner, and his way with com- puters makes him a key part of the Synclavier’s development.

Computer software means program- ming, and Jones has helped devise very creative ways of talking to the synthesizer via the computer that have greatly ex- panded the composer’s ability to use his imagination at the keyboard. “I like to think I am right inside that computer,” says Jones, “trying to get the synthesizer to do everything I want it to.”

The three men decided to divide music into three “time domains”: spectral, representing the characteristics of a par- ticular instrument; envelope, tracing the rise and fall of a single note in sound power; and volume, how loud the note is.

All of the composer’s decisions, then, about the desired characteristics of a note are transmitted to the synthesizer via the computer and Jones’ software, and the result, heard instantly, can be beautiful.

Alonso’s dreams for the Synclavier in- clude being able to plug in a microphone, sending natural sounds into the computer, and letting the synthesizer try to represent them. He would also like to set up the system so it will analyze a natural sound, synthesize it, listen to the result, and keep modifying it until something close to the natural sound comes out of the machine. The software does not yet exist for Alonso to realize either one of these dreams.

The Synclavier’s recent debut at Rollins Chapel had the audience alternately puzzled and delighted. There was Appleton playing at the keyboard and bells were heard. The bells that he had played a measure before came back over the speakers as horns and then reappeared as a harpsichord moments later. Appleton acted as if nothing unusual was happening. “I make the music, they make the machines,” he says, smiling.

Appleton now finds himself thinking about music and instruments as numbers and numerical relationships. The index of modulation for a French horn might be 25, for instance, and for a xylophone 60. “In this business a composer must guard against becoming merely a programmer,” he says. He has had to learn a great deal about acoustics, and he pays close atten- tion to the work of John M. Chowning at Stanford University, where a great deal of research has been done into the characteristics of natural sounds with an eye toward synthesizing them through digital systems.

“Alonso and Jones have the company, and the payoff for me is well-known: I get one of whatever they make, and I’m their music man,” says Appleton. Dartmouth students will have a Synclavier of their own in January.

Appleton is also one of the major sources of advertising for Synclavier at the moment. A record of his work, including two pieces written for and performed on the Synclavier, will be released in March 1978 on Wergo, a West-German label dis- tributed in the United States.

“The other night was my first live per- formance on the Synclavier, and I wasn’t used to an audience,” says Appleton, “but it went well. I even did some things I hadn’t thought of.”

Presto: trumpets!

Appleton observes music student Craig Harris '7B at the Synclavier keyboard. In thebackground of Dartmouth’s electronic music studio is an early Moog Synthesizer.

Woody Rothe is a free-lance writer andphotographer who specializes in theworkings of man and nature in NewEngland. He lives in Lyme, NewHampshire.