American Innovations in Electronic Musical Instruments

American Innovations in Electronic Musical Instruments

Samplers Although synthesizers, which directly generate an audio waveform algorithmically or in hardware, and samplers, which play back a recorded sound, have become intertwined in the current age of digital audio, their genesis took somewhat separate paths. One might be tempted to say that the roots of sampling lay in French tradition of musique concrète,… Read more »

Written By

Joseph A. Paradiso

Samplers

Although synthesizers, which directly generate an audio waveform algorithmically or in hardware, and samplers, which play back a recorded sound, have become intertwined in the current age of digital audio, their genesis took somewhat separate paths. One might be tempted to say that the roots of sampling lay in French tradition of musique concrète, pioneered by Pierre Henri and Pierre Schaeffer in the late 1940’s and later developed in the United States by such iconoclastic experimenters as Louis and Bebe Barron, Tod Dockstader, and Kenneth Gaburo. While there’s little doubt that today’s sampling techniques were foreshadowed by tape manipulation in the concrete studios, real-time performance samplers were in existence long before the term Musique Concrete was coined.

The first sample-playback synthesizers used glass or film disks, where a looped sound was encoded as an optical soundtrack, circling the disk in a series of concentric rings, one for each note. A DC lamp above the disk provided illumination and a radial bank of photocells below generated the audio signals as the disk rotated. As the waveforms weren’t actually recorded, the first units didn’t qualify as samplers, but rather wavetable synthesizers, with synthetic waveform amplitudes printed onto the transparent disk or encoded as slots cut into an opaque disk. These included Pierre Toulon’s Cellulophone (France, 1927), Emerick Spielmann’s Superpiano (Austria, 1927), Edwin Welte’s Light-Tone Organ (Germany, 1936), Ivan Eremeef’s Syntronic Organ and Photona (1934/35, USA), and the Radio Organ of a Trillion Tones (1931) and the Polytone Organ (1934) by A. Lesti and Fredrick Sammis in the USA. Some of these devices did make the bridge across to actual sample playback. These included the 1930 Hardy-Goldwaithe Organ (a keyboard instrument encoding 71 sampled notes onto a single optical disk) and Sammis’ “Singing Keyboard” in 1936, which played soundtracks stored on strips of 35 mm film. Optical sample players had a brief comeback in the early 70’s with the Optigan, a low-cost keyboard produced by Mattel that played sounds stored on interchangeable 12″ film disks and was sold at department stores like Sears. The disks encoded 63 separate tracks, 37 that were dedicated to the 3-octave keyboard, 21 to buttons sounding looped chords or instrumental vamps, and 5 to switches for selecting percussion loops or sound effects. Despite its mediocre sound quality, many Optigans were sold across the country, and the instruments are still prized for their camp appeal. Two descendants of the Optigan strived for higher quality sound and a professional market, namely the 1973 Chilton Talentmaker and the 1975 Vako Orchestron, the latter made famous by Patrick Moraz during his engagement with the British band Yes. After a brief commercial existence, these devices succumbed to rivals based on magnetic tape strips or digital storage.

Although tape samplers were explored in research studios (e.g., Hugh LeCaine’s 1955 keyboard-controlled “Special Purpose Tape Recorder”, which he used when recording his classic “Dripsody”), keyboard-driven tape instruments were being built and sold by California-based Harry Chamberlin from 1948 through the 70’s. Things really took off, however, when Chamberlain’s sales agent, Bill Fransen, brought two of these devices to England in 1962 to search for manufacturing and design improvements. He cracked a deal with Bill and Lesley Bradley of a Birmingham-based tape recorder company, Bradmatic Ltd, resulting in the first mellotrons. Over the next 15 years, the mellotron, particularly the M400 (released in 1970 and selling over 1800 units), had a major impact on rock music and is a trademark sound of the era’s progressive bands.

Digital keyboards were soon to overtake this path. The first real-time digital wavetable synthesizer was the 1971 Allen Digital Computer Organ, the result of a collaboration between the Allen Organ Company in Pennsylvania and a team led by Ralph Deutsch of North American Rockwell, who were exploring commercial applications of aerospace electronics. This design morphed into the RMI Keyboard Computer in 1977, a relatively portable device aimed at the rock market. Because of the limited amount of memory available at the time, the stored waveforms typically only spanned a single cycle, and additional signal processing was utilized for the attack and decay characteristics.

Although the RMI device used a kind of punched card to enable new waveforms to be input, no wavetable instruments of that time had any provisions for allowing a user to define a waveform or sample a sound. This changed with the 1979 introduction of the Australian-built Fairlight CMI, a high-end digital synthesizer built around the notion of user sampling. The effect on mainstream music was dramatic; although this technology was familiar to academic composers, the resulting sounds were entirely new to popular music, and quickly permeated the charts. Sampling capabilities were promptly adopted into the other high-end commercial digital synthesizer of the time, New England Digital’s Synclavier.

Less expensive systems were eventually produced for more affordable sampling. The first was the grainy-sounding Emulator I released by California-based E-Mu Systems in 1981, followed by the highly successful Emulator II in 1984. Likewise, 1984 produced the device that brought digital sampling to the average keyboardist, the low-cost Mirage by Ensoniq, a Pennsylvania-based spinoff of Commodore Computer Corp. Perhaps the minimum cost point in sampling keyboards was reached in 1986 with the SK-1 by Japan’s Casio Corporation. This was a mini-keyboard with crude sampling capability aimed at the toy market, retailing around $100.

Subsequent years saw an explosion of wavetable synthesizers; devices that played back looped waveforms stored in digital memory, but not normally able to sample sounds (although in some cases a sampling upgrade could be purchased). As making one’s own quality samples can be a delicate operation, most musicians opted instead to use preset samples from the device’s resident ROM or loaded from a library disk. The first good wavetable synthesizer was Wolfgang Palm’s 1980 German-made PPG Wave 2 Computer, which developed the WaveTerm sampling extension a few years later. The ways in which waveforms are switched and combined in the PPG devices were extended in the 1986 Prophet VS from California-based Sequential Circuits, a device which enabled envelopes to dynamically crossfade four evolving audio sources via what Sequential termed “Vector Synthesis”. After Sequential ceased operation, the VS was taken over by Korg, where it evolved into the popular Wavestation in 1990.

Wavetable synthesizers directly targeted acoustic instruments in 1983 with the K250 by Massachusetts-based Kurzweil Music Systems. This was the first commercial, real-time performance keyboard with a bank of ROM-based sounds that were sampled across many different notes and dynamics (termed multi-sampling), paving the way to realistic-sounding renditions of pianos and other concert instruments. Kurzweil released the inexpensive K1000 wavetable synthesizers in 1987, stocked with similar high-quality ROM sample sets, including many acoustic instruments. E-Mu followed suit in 1989 with the Proteus 1, drawing on the many sounds assembled for the then vast Emulator library.

Today, nearly every synthesizer on the market has a built-in wavetable that contains such acoustic or electronic samples; as memory is so inexpensive, devices routinely come with hundreds or thousands of waveforms stored in ROM. The nature of the “sampler” as a discrete unit is also blurring and becoming subsumed into the broader definition of digital audio. As recording studios are becoming all digital and based around personal computers, sampling is now often a PC or Macintosh software option, perhaps with an add-in computer card or module to handle high-resolution analog-to-digital conversion. Many synthesizers feature firmware hooks that give an external computer access to their wavetables. With appropriate sample-management software, sounds can be loaded directly from almost any source, such as audio CD’s or sound files from the web.

Several musicians, however, didn’t wait for the arrival of affordable samplers and took conventional recording and playback equipment onto the stage to work sampled sounds into their performances. Playing live music atop taped sound effects or backing tracks was fairly conventional in many venues by the late 60’s. Academic composers who blended electronics and live musicians frequently needed to have the electronic sounds played from tape (particularly if they were computer-generated), with the live musicians racing to keep time (this was termed “music-minus-one”). Taped effects likewise became an important part of concerts by Pink Floyd and other soon-to-be-popular bands that would go on to define “arena rock.”

The San Francisco group Negativland worked extensively with tape collages in a cross-over rock context in the early 70’s; while this wasn’t unusual in itself, when Negativland toured a decade later, they brought their samples along in a myriad of cartridge tapes, selecting, cueing, and playing soundbytes with amazing dexterity, in synch with live musicians playing more conventional instruments. At the dawn of the 80’s, the turntable also became a very expressive sample playback instrument used live on stage with famous DJs, scratchers, and rappers such as Grandmaster Flash and Afrika Bambaata, growing eventually into the highly-sample-leveraged hip-hop and urban styles of today. Perhaps the most extreme of turntable artists was Christian Marclay, who would paste slivers of different LP’s together to form “composite” records that he would play in his public performances and installations. Another approach to incorporating found sound into modern performance can be seen in the work of the British artist Scanner (a.k.a. Robin Rimbaud), famous for using a radio scanner during his shows to drop in live samples from sources such as police radio, cab dispatchers, and random, unsuspecting people talking on cell phones.

From American Innovations in Electronic Musical Instruments
by Joseph A. Paradiso
© 1999 NewMusicBox