A synthesizer (spelling var. synthesiser) is an electronic musical instrument designed to produce artificially generated sound, using techniques such as additive, subtractive, FM, physical modelling synthesis, or phase distortion.
Synthesizers create sounds through direct manipulation of electrical currents (as in analog synthesizers), mathematical manipulation of discrete values using computers (as in software synthesizers), or by a combination of both methods. In the final stage of the synthesizer, electrical currents are used to cause vibrations in the diaphragms of loudspeakers, headphones, etc. This synthesized sound is contrasted with recording of natural sound, where the mechanical energy of a sound wave is transformed into a signal which will then be converted back to mechanical energy on playback (though sampling significantly blurs this distinction).
When natural tonal instruments' sounds are analyzed in the frequency domain, the spectra of tonal instruments exhibit amplitude peaks at the harmonics. These harmonics' frequencies are primarily located close to the integer multiples of the tone's fundamental frequency.
Percussives and rasps usually lack harmonics, and exhibit spectra that are comprised mainly of noise shaped by the resonant frequencies of the structures that produce the sounds. The resonant properties of the instruments (the spectral peaks of which are also referred to as formants) also shape the spectra of string, wind, voice and other natural instruments.
In most conventional synthesizers, for purposes of resynthesis, recordings of real instruments can be thought to be composed of several components.
These component sounds represent the acoustic responses of different parts of the instrument, the sounds produced by the instrument during different parts of a performance, or the behaviour of the instrument under different playing conditions (pitch, intensity of playing, fingering, etc...) The distinctive timbre, intonation and attack of a real instrument can therefore be created by mixing together these components in such a way as resembles the natural behaviour of the real instrument. Nomenclature varies by synthesizer methodology and manufacturer, but the components are often referred to as oscillators or partials. A higher fidelity reproduction of a natural instrument can typically be achieved using more oscillators, but increased computational power and human programming is required, and most synthesizers use between one and four oscillators by default.
One of the most salient aspects of any sound is its amplitude envelope. This envelope determines whether the sound is percussive, like a snare drum, or persistent, like a violin string. Most often, this shaping of the sound's amplitude profile is realized with an "ADSR" (Attack Decay Sustain Release) envelope model applied to control oscillator volumes. Each of these stages is modelled by a change in volume (typically exponential). The attack is the initial run-up of the sound level. The decay is the run down immediately after the attack. Sustain is the volume when the note is held. The release is the volume profile when the note is released. Exponential rates are commonly used because they closely model real physical vibrations, which usually rise or decay exponentially.
Although the oscillations in real instruments also change frequency, most instruments can be modelled well without this refinement. This refinement is necessary to generate a vibrato.
Overview of popular synthesis methods
Subtractive synthesizers use a simple acoustic model that assumes an instrument can be approximated by a simple signal generator (producing sawtooth waves, square waves, etc...) followed by a filter which represents the frequency-dependent losses and resonances in the instrument body . For reasons of simplicity and economy, these filters are typically low-order lowpass filters. The combination of simple modulation routings (such as pulse width modulation and oscillator sync), along with the physically unrealistic lowpass filters, is responsible for the "classic synthesizer" sound commonly associated with "analog synthesis" and often mistakenly used when referring to software synthesizers using subtractive synthesis. Although physical modelling synthesis, synthesis wherein the sound is generated according to the physics of the instrument, has superseded subtractive synthesis for accurately reproducing natural instrument timbres, the subtractive synthesis paradigm is still ubiquitous in synthesizers with most modern designs still offering low-order lowpass or bandpass filters following the oscillator stage.
One of the easiest synthesis systems is to record a real instrument as a digitized waveform, and then play back its recordings at different speeds to produce different tones. This is the technique used in "sampling." Most samplers designate a part of the sample for each component of the ADSR envelope, and then repeat that section while changing the volume for that segment of the envelope. This lets the sampler have a persuasively different envelope using the same note. See also: Sample-based synthesis.
There are also many different kinds of synthesis methods, each applicable to both analog and digital synthesizers. These techniques tend to be mathematically related, especially frequency modulation and phase modulation.
- Subtractive synthesis
- Additive synthesis
- Granular synthesis
- Wavetable synthesis
- Frequency modulation synthesis
- Phase distortion synthesis
- Physical modelling synthesis
- Digital sampling
The start of the analog synthesizer era
Early synthesizers used technology derived from electronic analog computers and laboratory test equipment.
In the 1950s, RCA produced experimental devices to synthesize both voice and music. The giant Mark II Music Synthesizer, housed at the Columbia-Princeton Electronic Music Center] in New York City in 1958, was only capable of producing music once it had been completely programmed. The vacuum tube system had to be manually patched to create each new type of sound. It then used a paper tape sequencer punched with holes that controlled pitch sources and filters, similar to a mechanical player piano but able to generate a wide variety of sounds.
In 1958 Daphne Oram at the BBC Radiophonic Workshop produced a novel synthesizer using her "Oramics" technique, driven by drawings on a 35mm film strip. This was used for a number of years at the BBC. Hugh Le Caine, John Hanert, Raymond Scott, Percy Grainger (with Burnett Cross), and others built a variety of automated electronic-music controllers during the late 1940s and 1950s.
By the 1960s, synthesizers were developed which could be played in real time but were confined to studios because of their size. Modularity was the usual design, with standalone signal sources and processors being connected with patch cords or by other means, and all controlled by a common controlling device.
Early synthesizers were often experimental special-built devices, usually based on the concept of modularity. Donald Buchla, Hugh Le Caine, Raymond Scott and Paul Ketoff were among the first to build such instruments, in the late 1950s and early 1960s. Only Buchla later produced a commercial version.
The first playable modern configurable music synthesizer was created by Robert Moog, who had been a student of Peter Mauzey, one of the engineers of the RCA Mark II. Moog designed the circuits used in his synthesizer while he was at Columbia-Princeton. The Moog synthesizer was first displayed at the Audio Engineering Society convention in 1964. Like the RCA Mark II, it required hours to set up the machine for a new sound, but it was smaller and more flexible. The Moog synthesizer was at first a curiosity, but by 1968 it had caused a sensation.
Micky Dolenz of The Monkees bought one of the first three Moog synthesizers and the first commercial release to feature a Moog synthesizer was The Monkees' fourth album, Pisces Aquarius Capricorn & Jones Ltd., in 1967. Pisces Aquarius Capricorn & Jones Ltd. also became the first album featuring a synthesizer to hit #1 on the charts. Also among the first music performed on this synthesizer was the million-selling 1968 album Switched-On Bach by Wendy Carlos. Switched-On Bach was one of the most popular classical-music recordings ever made. During the late 1960s, hundreds of other popular recordings used Moog synthesizer sounds. The Moog synthesizer even spawned a subculture of record producers who made novelty "Moog" recordings, depending on the odd new sounds made by their synthesizers (which were not always Moog units) to draw attention and sales.
Moog also established standards for control interfacing, with a logarithmic 1-volt-per-octave pitch control and a separate pulse triggering signal. This standardization allowed synthesizers from different manufacturers to operate together. Pitch control is usually performed either with an organ-style keyboard or a music sequencer, which produces a series of control voltages over a fixed time period and allows some automation of music production.
One major innovation by Moog was in 1970, when they made a synthesizer with a built-in keyboard and without modular design--the analog circuits were retained, but made interconnectable with switches in a simplified arrangement called "normalization". Though less flexible than modularity, it made the instrument more portable and its use much easier. This first prepatched synthesizer, the Minimoog, became very popular, with over 12,000 units sold. The Minimoog also influenced the design of nearly all subsequent synthesizers.
In the 1970s miniaturized solid-state components let synthesizers become self-contained and movable. They began to be used in live performances. Soon, electronic synthesizers had become a standard part of the popular-music repertoire, with Giorgio Moroder's "Son of my Father" the first #1 hit to feature a synthesizer (Shapiro, 2000).
By 1984, Raymond Kurzweil, on suggestion from Stevie Wonder, created the first synthesizer that could duplicate the sounds of orchestral instruments. It was based on recorded samples of actual instruments. Trained conductors and musicians were incapable of distinguishing the Kurzweil synthesizer from the real thing.
Roll your own
During the late 70s and early 80s it was relatively easy to build your own synthesizer. Designs were published in hobby electronics magazines and complete kits were supplied by companies such as Maplin Electronics in the UK.
Electronic organs vs. synthesizers
All organs (including acoustic) are based on the principle of additive or Fourier synthesis: Several sine tones are mixed to form a more complex waveform. In the original Hammond organ, built in 1935, these sine waves were generated using revolving tone wheels which induced a current in an electromagnetic pick-up. For every harmonic, there had to be a separate tonewheel. In more modern electronic organs, electronic oscillators serve to produce the sine waves. Organs tend to use fairly simple "formant" filters to effect changes to the oscillator tone--automation and modulation tend to be limited to simple vibrato.
Most analog synthesizers produce their sound using subtractive synthesis. In this method, a waveform rich in overtones, usually a sawtooth or pulse wave, is produced by an oscillator. The signal is then passed through filters, which preferentially remove some overtones to obtain a sound which may be an imitation of an acoustical instrument, or may be a unique tonality not existing in acoustical form. An ADSR envelope generator then controls a VCA (voltage controlled amplifier) to give the sound a loudness contour.
Other circuits, such as waveshapers and ring modulators, can change the tonality in non-harmonic ways or create distortion effects which are often not found in natural sound sources. In spite of the popularity of modern digital and software-based synthesizers, the purely analog modular synthesizer still has its proponents, with a number of manufacturers producing modules little different from Moog's 1964 circuit designs, as well as many newer variations.
Microprocessor controlled and polyphonic analog synthesizers
Early analog synthesizers were always monophonic, producing only one tone at a time. A few, such as the Moog Sonic Six, ARP Odyssey and EML 101, were capable of producing two different pitches at a time when two keys were pressed. Polyphony (multiple simultaneous tones, which enables chords), was only obtainable with electronic organ designs at first. Popular electronic keyboards combining organ circuits with synthesizer processing included the ARP Omni and Moog's Polymoog and Opus 3.
By 1976 the first true music synthesizers to offer polyphony had begun to appear, most notably in the form of Moog's Polymoog, the Yamaha CS-80 and the Oberheim Four-Voice. These early instruments were very complex, heavy, and costly. Another feature that began to appear was the recording of knob settings in a digital memory, allowing the changing of sounds quickly.
When microprocessors first appeared on the scene in the early 1970s, they were expensive and difficult to apply. The first practical polyphonic synth, and the first to use a microprocessor as a controller, was the Sequential Circuits Prophet-5 introduced in 1977. For the first time, musicians had a practical polyphonic synthesizer that allowed all knob settings to be saved in computer memory and recalled by pushing a button. The Prophet-5 was also physically compact and lightweight, unlike its predecessors. This basic design paradigm became a standard among synthesizer manufacturers, slowly pushing out the more complex (and more difficult to use) modular design.
Synthesizers became easier to integrate and synchronize with other electronic instruments and controllers with the invention in 1983 of MIDI, a time-coded serial interface cable. MIDI interfaces are now almost ubiquitous on music equipment, and commonly available on personal computers (PCs).
The so-called General MIDI (GM) software standard was devised in 1991 to serve as a consistent way of describing a set of over 200 tones (including percussion) available to a PC for playback of musical scores. For the first time, a given MIDI preset would consistently produce an oboe or guitar sound (etc.) on any GM-conforming device. The file format .mid was also established and became a popular standard for exchange of music scores between computers.
John Chowning of Stanford University is generally considered to be the first researcher to conceive of producing musical sounds by causing one oscillator to modulate the pitch of another. This is called FM, or frequency modulation, synthesis. Chowning's early FM experiments were done with software on a mainframe computer.
FM uses sine-wave oscillators (called operators) which, in order for their fundamental frequency to be sufficiently stable, are normally generated digitally. Each operator's audio output may be fed to the input of another operator, via an ADSR or other envelope controller. The first operator modulates the pitch of the second operator, in ways that can produce complex waveforms. FM synthesis is fundamentally a type of additive synthesis and the filters used in subtractive synthesizers were typically not used in FM synthesizers until the mid-1990s. By cascading operators and programming their envelopes appropriately, some subtractive synthesis effects can be simulated, though the sound of a resonant analog filter is almost impossible to achieve. FM is well-suited for making sounds that subtractive synthesizers have difficulty producing, particularly non-harmonic sounds, such as bell timbres.
Chowning's patent covering FM sound synthesis was licensed to giant Japanese manufacturer Yamaha, and made millions for Stanford during the 1980s. Yamaha's first FM synthesizers, the GS-1 and GS-2, were costly and heavy. They soon followed the GS series with a pair of smaller, preset versions - the CE20 and CE25 Combo Ensembles - which were targeted primarily at the home organ market and featured four-octave keyboards. Their third version, the DX-7 (1983), was about the same size and weight as the Prophet-5, was reasonably priced, and depended on custom digital integrated circuits to produce FM tonalities. The DX-7 was a smash hit and may be heard on thousands of pop recordings from the 1980s. Yamaha later licensed its FM technology to other manufacturers. By the time the Stanford patent ran out, almost every personal computer in the world contained an audio input-output system with a built-in 4-operator FM digital synthesizer -- a fact most PC users are not aware of.
Samplers and sampling
One kind of synthesizer starts with a binary digital recording of an existing sound, which is then replayed at a range of pitches. This is called a sampler.
Sampling can also be used in combination with other synthesizer effects. Some popular software synthesizers take a sampled sound and process it with software-based filters, reverbs, ring modulators and the like.
Sampling started out as the purview of academic researchers with access to mainframe computers. The appearance of the Fairlight CMI in 1979, the first well-known digital instrument capable of sampling, started a revolution. The Fairlight was used on scores of popular recordings by artists such as Jean-Michel Jarre, Kate Bush, Peter Gabriel and Art of Noise. The costly, complex and rare Fairlight (and an equally costly competitor, the New England Digital Synclavier) caused California company E-Mu to introduce their Emulator I in 1981, a lower-cost sampling keyboard which could save sound recordings to floppy disk. Ensoniq followed suit in 1985 with an even lower-cost sampling keyboard, the Ensoniq Mirage, which cost about $1,500 USD compared to the $7,900 USD price tag of the Emulator I.
The physical modelling synthesizer
Physical modelling synthesis is the synthesis of sound by using a set of equations and algorithms to simulate a physical source of sound. When an initial set of parameters is run through the physical simulation, the simulated sound is generated.
Although physical modelling was not a new concept in acoustics and synthesis, it wasn't until the development of the Karplus-Strong algorithm, the subsequent refinement and generalization of the algorithm into digital waveguide synthesis by Julius O. Smith III and others, and the increase in DSP power in the late 1980s that commercial implementations became feasible.
Following the success of Yamaha's licensing of Stanford's FM synthesis patent, Yamaha signed a contract with Stanford University in 1989 to jointly develop digital waveguide synthesis. As such, most patents related to the technology are owned by Stanford or Yamaha. A physical modeling synthesizer was first realized commercially with Yamaha's VL-1, which was released in 1994.
The modern digital synthesizer
Most modern synthesizers are now completely digital, including those which model analog synthesis using digital techniques. Digital synthesizers use digital signal processing (DSP) techniques to make musical sounds. Some digital synthesizers now exist in the form of 'softsynth' software that synthesizes sound using conventional PC hardware. Others use specialized DSP hardware.
Digital synthesizers generate a digital sample, corresponding to a sound pressure, at a given sampling frequency (typically 44100 samples per second). In the most basic case, each digital oscillator is modelled by a counter. For each sample, the counter of each oscillator is advanced by an amount that varies depending on the frequency of the oscillator. For harmonic oscillators, the counter indexes a table containing the oscillator's waveform. For random-noise oscillators, the most significant bits index a table of random numbers. The values indexed by each oscillator's counter are mixed, processed, and then sent to a digital-to-analog converter, followed by an analog amplifier.
To eliminate the difficult multiplication step in the envelope generation and mixing, some synthesizers perform all of the above operations in a logarithmic coding, and add the current ADSR and mix levels to the logarithmic value of the oscillator, to effectively multiply it. To add the values in the last step of mixing, they are converted to linear values.
The earliest digital synthesis was performed by software synthesizers on mainframe computers using methods exactly like those described in digital synthesis, above. Music was coded using punch cards to describe the type of instrument, note and duration. The formants of each timbre were generated as a series of sine waves, converted to fixed-point binary suitable for digital-to-analog converters, and mixed by adding and averaging. The data was written slowly to computer tape and then played back in real time to generate the music.
Today, a variety of software is available to run on modern high-speed personal computers. DSP algorithms are commonplace, and permit the creation of fairly accurate simulations of physical acoustic sources or electronic sound generators (oscillators, filters, VCAs, etc). Some commercial programs offer quite lavish and complex models of classic analog synthesizers--everything from the Yamaha DX-7 to the original Moog modular. Other programs allow the user complete control of all aspects of digital music synthesis, at the cost of greater complexity and difficulty of use.
Commercial synthesizer manufacturers
Notable synthesizer manufacturers past and present include:
- Electronic Music Studios
- Kurzweil Music Systems
- New England Digital (NED)
- Palm Productions GmbH (PPG)
- Roland Corporation
- Sequential Circuits
- Waldorf Music
For a more complete list see Category:Synthesizer manufacturers
Classic synthesizer designs
This is intended to be a list of classic instruments which marked a turning point in musical sound or style, potentially worth an article of their own. They are listed with the names of performers or styles associated with them. For more synthesizer models see Category:Synthesizers.
- Moog modular synthesizer systems (Wendy Carlos, Tomita, Tonto's Expanding Head Band, Emerson, Lake and Palmer, The Beatles)
- Mellotron (Ray Buttigieg, Tangerine Dream, Moody Blues, Barclay James Harvest, T-Rex, The Beatles)
- ARP 2600 (The Who, Stevie Wonder, Weather Report, Edgar Winter, Jean-Michel Jarre)
- ARP Odyssey (Ultravox, Styx, Herbie Hancock)
- Minimoog (Rush),(Yes, Emerson Lake and Palmer, Stereolab, Devo, Ray Buttigieg)
- EMS VCS3 (Roxy Music, Hawkwind, Pink Floyd, BBC Radiophonic Workshop)
- Fairlight CMI (Jean-Michel Jarre, Jan Hammer, Peter Gabriel, Pet Shop Boys)
- NED Synclavier (Michael Jackson, Stevie Wonder, Laurie Anderson)
- Sequential Circuits Prophet 5 (Berlin, Phil Collins, The Cars, Steve Winwood)
- E-mu Emulator (The Residents, Depeche Mode, Deep Purple, Genesis)
- Roland Jupiter-8 (Duran Duran, OMD)
- PPG Wave (Rush),(The Fixx, Thomas Dolby)
- Roland TB-303 (Techno, Acid House)
- Roland D-50 ((Jean-Michel Jarre, Enya)
- WaveFrame AudioFrame (Peter Gabriel, Stevie Wonder)
- Yamaha DX7 (Steve Reich, The Cure, Brian Eno)
- Yamaha VL-1
- Yamaha SHS-10 One of the first "keytars" from the 1980s
- Korg M1
- Korg Karma
- Roland JP-8000 (The synthesizer that defined modern trance music; the warm sawtooth lead sound heavily used in modern trance music originates from the JP-8000.)
- Clavia Nord Lead (The Prodigy, The Weathermen,Jean Michel Jarre) The Nord Lead was the first modern analog modelling synthesizer using digital circuitry to emulate analog circuits
- Oberheim OB-Xa (Rush), (Styx, Supertramp, Van Halen)
- Lyricon First mass-produced wind synthesizer. (Michael Brecker, Tom Scott, Chuck Greenberg, Wayne Shorter)
- Alesis Andromeda A synthesizer with modern digital control of fully analog sound producing circuitry