My focus on the Yamaha DX7 synthesizer in my research may seem overly niche or specialized, but in this essay I establish the magnitude of this synthesizer’s impact on popular music in the 1980s. While the average person is unaware of the synthesizer by name, the DX7 changed popular music as we know it today. To say that the DX7’s arrival was earth-shaking would be no exaggeration: the affordable price, sound palette, and physical feel of the instrument combined to make the DX7 the new must-have instrument in every studio, garage, and university music department in the U.S. and the U.K. My study of contemporaneous music periodicals has led me to construct a narrative that describes how the DX7 impacted the lives of working musicians in the U.S. and the U.K. in the 1980s.
The sonic landscape of popular music included keyboards long before the 1980s. Along with the piano, the standby keyboard instrument in the pop world, keyboardists might often play electric pianos such as the Fender Rhodes, the Hohner Pianet, the Wurlizer electric piano, or the distinctive Hohner Clavinet, or electric organs such as the Hammond B3 or the Vox Continental. A handful of keyboard legends emerge in popular music during the 1950s and ’60s—Little Richard, Stevie Wonder, Ray Manzarek, and Keith Emerson, to name a few—but prior to the 1970s, the role of the keyboardist was, generally speaking, relegated to a rhythm section role, rather than being featured as a solo instrument. In other words, while guitar heroes like Jimi Hendrix, Eric Clapton, and Keith Richards enraptured audiences with epic guitar solos, keyboardists typically remained out of the spotlight. The amplified and distorted guitar was able to produce loud, sustained sounds, which were simply better suited to virtuosic and crowd-pleasing solos. In those early decades, the keyboard instruments were no match.
The Moog modular analog synthesizer, developed in the early 1960s, was the first synth that rivaled the power of the electric guitar in volume and timbre. The Moog is an analog modular synthesizer, meaning that the sounds were generated from a sound wave oscillator—a physical (analog) device, controlled, in a physical sense, by voltage. To produce different sounds on the Moog, one needed different modules that would further generate or modify sounds. Sounds were created using quarter-inch cables to “patch” modules together (this is the etymology of the term “patch,” which today refers to any sound or preset on a synthesizer). The resulting collection of modules was typically bigger than the keyboard itself. The size alone meant the Moog could not be taken on the road: this early synthesizer, not unlike the early computers, was a massive device that was prohibitively expensive for most musicians.
The invention of the Minimoog resolved the size issue, and made the price more affordable. Introduced in 1970, the Minimoog was capable of producing a similarly powerful analog sound without needing the bulky modules. The Minimoog had limitations, being a monophonic synthesizer; ironically, these limitations actually led to the liberation of the keyboardist from the background. The Minimoog was not functional as a pad or rhythm instrument like older organs and pianos, but it excelled at the indulgent, virtuosic, monophonic solos that had previously been the exclusive purview of the electric guitar. The Minimoog firmly established synthesizers as an important musical instrument within pop and rock music. The synthesizer rapidly became cool, and moreover, became synonymous with futuristic technology—a signification with important implications for aesthetics in 1980s pop music, which frequently used themes of futurism.
A handful of synthesizers dominated the sound of the 1980s. The Roland Jupiter-8, released in 1981, was one of these. The Jupiter-8 is an analog synthesizer so immensely powerful that it can seem like the teleological goal of all prior analog synthesizer development. The Jupiter-8 is responsible for many iconic ’80s sounds, including Michael Jackson’s “Thriller” (recorded in 1982) and music for the movie The NeverEnding Story (1984).
Analog synthesis never really disappeared from the sound of popular music, but most other quintessential ’80s synthesizers introduced the listening public to other, newer methods of sound synthesis, all of which were made possible by digital computing and microprocessors. The Fairlight CMI, first released in 1979, and the E-MU Emulator, first released in 1981, are two of the first keyboards to use digital sampling. Sampling made any sound imaginable available to keyboardists—or at least, to keyboardists who could afford them. Both the Fairlight and the Emulator were prohibitively expensive for the average keyboardist, with list prices in the tens of thousands of dollars at their release (1979 and 1981, respectively). Successful artists like Kate Bush and Duran Duran, who were able to pay for studio time and access to them, used these instruments to bring an entirely new flavor to their music. Even while samplers were out of the reach of many musicians, they left a major mark on the music of the 1980s.
Also new in the 1980s was wavetable synthesis, famously deployed in the PPG Wave synthesizers. The PPG Wave processes analog waveforms through the digital microprocessors to produce new, more dynamic sounds. Both sampling and wavetable synthesis represent a merging of analog and digital technologies: the source of the sound is an analog source—a recorded analog sample in the case of the sampler, and an analog waveform in the case of wavetable synthesis—but the sound is processed and delivered digitally.
The other classic ’80s synths were purely digital, using frequency modulation (FM) synthesis technology first developed by John Chowning at Stanford University in the 1960s. The New England Digital Synclavier, introduced in 1977, was the first commercial instrument to solely use digital FM synthesis. The Yamaha GS-1, released in 1981, was Yamaha’s first FM keyboard—technically not a synthesizer at all, because it consisted entirely of stored sounds from the factory. Digital FM synthesis remained less popular than other formats until one pivotal moment: the unveiling of the Yamaha DX7 at the National Association of Music Merchants show in the summer of 1983.
The Yamaha DX7 was only produced from 1983 to 1986, and in this short period, Yamaha sold approximately 150,000 of them; the DX7 today remains one of the best-selling synthesizers of all time. Keyboardists across Europe, North America, and Asia were enamored with the DX7, to the point where it was seen as a replacement for antecedent workhorse synthesizers. Many musicians were smitten with the possibilities, and not only because they were dazzled by a brand new technology. Two years after the release of the DX7, when the newness would already have worn off, jazz and R&B keyboardist Patrice Rushen said, “[The DX7] is such a great instrument, such a versatile instrument. I think we’re still at the tip of the iceberg of what it’s capable of.” Some musicians even insinuated the DX7 would make other synthesizers obsolete. Roy Bittan, a keyboardist for Bruce Springsteen’s 1986 tour, put this into practice when he transferred all the synthesizer sounds that he could from the older Yamaha CS-80 analog synthesizer to the DX7: “I realized at once that [the DX7] was going to be very valuable in the future. … [T]he CS-80 is harder to control, and the sound is not as clean.” Similarly, Jimmy Jam (Timmy Harris), a songwriter and producer for Janet Jackson and other singers, said he never used a real Fender Rhodes anymore after the DX7 and other synthesizers began providing similar electric piano sounds: “No [I never use a real Rhodes sound], I just go direct into the board with a Rhodes synth sound. With all the companies having Rhodes patches, it’s easier to use the variations.” Film composer Jerry Goldsmith gushes, “The Yamaha DX7 is amazing; some people feel that if you have a rack of three or four of them, you don’t need anything else.” In the eyes of many musicians, the DX7 was simply more practical than an array of other synthesizers.
“The Synth that Changed Everything”—a title bestowed upon the DX7 in a 30th anniversary retrospective in Keyboard—enjoyed its immense success for two reasons. First of all, the DX7 made use of several cutting-edge technologies, and secondly, it sold at a relatively affordable price. Perhaps the most truly groundbreaking feature of the DX7 was its envelope generator. The old standard for an envelope generator was a four-step envelope, or attack-decay-sustain-release (ADSR) envelope, that allowed for control of the rate of the attack (but not its level), the rate of decay, the level of the sustain, and the rate of the release (the level of which was always 0). The DX7 introduced the eight-step envelope, which allows the user to define four different levels, as well as the rates of change between each of those levels, totaling eight steps in the definition of the envelope, and opening up the possibility of all sorts of unusual envelopes. Example 1 compares the envelopes of a Jupiter-8 with that of a DX7. 2.1a is a representation of the envelope interface on the Jupiter-8. Each of the four sliders corresponds to a “step” in the four-step ADSR envelope. A adjusts the rate of the amplitude’s increase from 0 to peak; D adjusts the rate of the amplitude’s decay from peak down to S; S is a user-specified amplitude; R is the rate of the amplitude’s decay from S down to 0. There are two switches: one for “key follow,” which adjusts the envelope according to the frequency of the sound to mimic an acoustic instrument, and one for “envelope inversion,” which reverses the profile of the ADSR envelope. By contrast, on the DX7, since both levels and rates of change are specified by the user, envelopes no longer needed to conform to these prototypical rules. Example 1b illustrates each of the eight steps in the eight-step envelope. While this figure uses a typically shaped envelope for its demonstration, crucially, the envelope can take on absolutely any shape, since all rates and levels are customizable.
This flexibility alone would provide plenty of options for most people, but the DX7 goes further: each of the six operators in the tone-generation process may be assigned its own unique amplitude envelope, once again multiplying the number of the DX7’s potential sounds, and allowing for a dynamism in the timbral profiles of these sounds that was never before possible. Some of the DX7’s presets emulated non-musical sounds, a capability made possible in part by these complex envelope options. For example, TAKE OFF seems to simulate an airplane or spaceship engine starting up and then launching the craft—something which does not have an amplitude profile conforming to that of a traditional or inverted four-step ADSR envelope. If desired, this same envelope generator can also be used to control the pitch, through an additional seventh envelope called the pitch EG.
The timbre of a sound on the DX7 can change further depending on how hard and/or how fast the player strikes the key: in other words, the keys are also both pressure- and velocity-sensitive. Jay Chapman, a writer for Electronics & Music Maker, states, “It’s my opinion that the velocity/touch-sensitivity feature … is a major contributory factor in putting the DX7 into a class of its own when compared to the other polyphonic synths available in the same price range.” The DX7’s keyboard action was one of the best in the synthesizer market, and this technology was incorporated into the design of the sounds themselves. Another non-musical preset, TRAIN, would clang a train bell when the key was pressed normally, but if the player then presses harder onto the key, the sounds of a train engine chugging gradually crescendo while the bell reverberates. Sounds like TRAIN and TAKE OFF might sound quaint or gimmicky to modern ears, but this was the first time such a sound had ever been synthesized from scratch (rather than sampled).
Other features were not unique to the DX7, but were still on the vanguard of innovation. The DX7 was one of the earliest keyboards to make use of the now-commonplace Musical Instrument Digital Interface (better known as MIDI) technology. The DX7’s MIDI capability was limited—it only transmits on MIDI channel 1—but effective. MIDI sends data to and from the DX7, which means that users could use MIDI to exchange preset sounds, or to hook the DX7 up as a controller for other synthesizers. Coupled with the DX7’s excellent velocity and pressure sensitivity, the DX7 was a popular choice of MIDI controller. More radical was that the DX7 could be hooked up via MIDI to a breath controller, which could better simulate the sound of wind instruments. As with velocity and pressure sensitivity, Yamaha specially designed several preset sounds (those ending in “BC”) to be used with a breath controller. In 1983, these capabilities were very hard to find in other synthesizers.
FM synthesis is really what lies at the heart of the DX7’s innovative status. FM is a highly versatile method of sound synthesis, capable of producing a wide range of timbres, if one knows how to program with FM. Although a complete discussion of the processes involved in FM synthesis is beyond the scope of this project, a short overview is helpful in understanding what makes the FM synthesis process of the DX7 so revolutionary compared to subtractive synthesis on analog machines. FM synthesis on the DX7 begins with the generation of sine waves from the DX7’s digitally controlled oscillators, called operators. An operator can be either a carrier wave or a modulator wave in the FM sound synthesis process (as in FM radio), and these carriers and modulators can feed into one another through various arrangements, known as algorithms (see Example 2). In an algorithm with only two operators, often one operator would be the carrier and the other the modulator. Relating this abstract mathematical concept to musical sound, the frequency of a carrier will determine the pitch of the note played, while the frequency of a modulator will affect the timbre. The DX7 has six operators, each of which might be turned on or off, and the programmer may arrange these operators into one of thirty-two different algorithms, dictating which operators are modulators or carriers and the relationship between them all. Factoring all these variables together, the number of possible sounds is immense. This vast universe is quickly made apparent to the user of a DX7 through the visual map of all 32 algorithms painted directly onto the face of the instrument (see Example 3).
The innovation of FM synthesis was the greatest strength of the DX7, but also, paradoxically, its greatest failure. The DX7 rapidly gained a reputation as being difficult to program, a reputation that endures to this day, due to the relative complexity of FM synthesis. Using an algorithm to modulate carrier sine waves with other sine waves is not as easily understood by a layperson as the subtractive synthesis technologies used by most analog machines. The daunting new FM digital system motivated many artists’ rejections of the DX7. Alan Howarth, for example, said,
I’m not a real DX7 fan … I never warmed up to it, and along came the [Sequential Circuits Prophet VS, a hybrid digital-analog synth]. It gave me some pretty neat sounds. The digital waveforms were nice and bright, and they gave me all the [Hohner] Clavinet and [Fender] Rhodes things, and I could still deal with it in an analog fashion, which is what I’m most familiar with.
This reveals that although Howarth was interested in the sound of digital synthesis, he was reluctant to spend time stumbling on the learning curve of this entirely new system of sound creation. Midge Ure similarly states, “The DX7 … I still think is impossible to program! I can’t be bothered with it.”
Further obfuscating the process of programming the DX7 is its unintuitive user interface. On virtually every other synthesizer before the DX7, timbres were created and adjusted by fiddling with a number of knobs. All these knobs were laid out plainly before the programmer’s eyes; one knob controlled one parameter. The programmer could press and hold a key on the synthesizer, simultaneously twist knobs back and forth, and hear the effect of these adjustments on the sound. The process of creating an analog sound was fairly intuitive, and one did not necessarily need to know about acoustics or anything else mathematical in order to create nice sounds.
A DX7 programmer, on the other hand, interacts with the FM synthesis algorithms through a series of membrane buttons and one slider that increases or decreases numbers. What the slider and the buttons do might change depending on what “mode” the programmer is using—which may seem simple enough, but this simple barrier puts a significant conceptual filter between the visible interface and the technology. Even the conceptually straightforward slider does not provide the programmer with real-time feedback while adjusting the numbers with the slider: before the programmer can hear what they have done, the sound has to be saved, the DX7 switched over from “edit” mode to “play” mode, and only then can the programmer press a key and hear the results. Listening to the sound is Step 10 in the step-by-step guide to creating a sound in the DX7 owner’s manual. There is no way for programmers to experience the effect of their changes to the algorithm in time as they make the changes.
All of this information is only communicated to the DX7 programmer on a tiny, non-backlit LCD screen, capable of displaying only two rows of sixteen characters each. As Ted Greenwald, an editor for Keyboard magazine, summarizes, “Admittedly, the difficulty of DX programming stems as much from the differences between analog and FM synthesis, and between [four-step] and [eight-step] envelopes, but there was clearly a need for a front panel that related, in graphic terms, to the structure of any given patch.” The combination of the more abstract mathematics behind sound generation, the substantial differences between analog and digital technology, and the difficult interface led to a total avoidance of programming the DX7 by the vast majority of keyboardists.
Given the degree of obfuscation in the process of editing a timbre on the DX7, the instrument’s status as a best-selling synthesizer may seem completely surprising, but the DX7 had one crucial redemptive feature: the pre-programmed factory preset sounds. Thirty-two factory presets were saved into the synthesizer’s internal memory. A cartridge reader gives users access to even more sounds, read from either ROM or RAM cartridges. Every new DX7 shipped with two ROM cartridges from Yamaha with sixty-four sounds each, thirty-two of which were the same sounds programmed in the internal memory, yielding a total of 128 different sounds that came with the synthesizer.
The DX7 was one of the earliest synthesizers to come equipped with such pre-programmed sounds, referred to as “presets.” In older synthesizers, before the advent of computing and digital memory, there were no such thing as presets—presets are created with digital memory, and digital computing was still new in 1983. On synthesizers like the Minimoog, instead of presets, players had to recreate sounds anew each time, by fiddling with knobs and sliders. To keep track of these sounds, people would write them down on charts. To switch sounds back and forth, one had to physically move the knobs back and forth—or one could take after Rick Wakeman, who is said to have bought thirteen Minimoogs and surrounded himself with them at his shows. When digital memory was available, people used it to save their own pre-programmed sounds, not to use other people’s sounds. Where factory presets were available, for example, on the Jupiter-8, they were frequently saved over with the user’s own sounds, or otherwise not much used. As Paul Théberge notes, the reason older synthesizers did not have presets was not, as one might assume, because the technology did not previously exist, but rather because instrument makers did not believe there was a demand for the feature. Synthesizers had only ever existed as a specialized and exclusive kind of instrument, and so in the early days of the synthesizer, people that wanted to play synthesizers were the same people that wanted to learn how to become synthesizer programmers.
Synthesizers became more integral to the sound of popular music throughout the course of the 1980s. Many everyday musicians, such as those playing in garage bands, wanted to buy their first synthesizer, but had no familiarity with the basic concepts of synthesizer programming. By 1983, the availability of presets made the DX7 exceptionally attractive to such keyboardists. Instead of needing to learn about subtractive synthesis, with its oscillators, filters, and waveforms—or worse, needing to learn about FM synthesis!—the salesperson could show buyers that they simply needed to press one of the thirty-two buttons on the face of the DX7 to quickly access any sound. Bass, harpsichord, Rhodes, brass, organ, strings, bells, and more: everything the buyer might need was available with the push of a button, no twiddling of knobs or comprehension of algorithms required. Sounds on the DX7 are literally at the player’s fingertips (Example 4).
Rather than wrestle with the unforgiving programming interface, most casual players of the DX7, and even many experienced session musicians, relied solely on the presets in the internal memory or cartridges. In October 1986, when the DX7 had been on the scene for three years, producer David Briggs expressed a sentiment shared by many music makers of the 1980s: “I haven’t had the need [to program the DX7 myself.] Most of the factory sounds that I use are great, and I haven’t been able to improve on them. … Everybody wants the brass, electric piano, and string sounds.” Théberge similarly reports, “By the end of [the 1980s], marketing departments were estimating that as few as 10 percent of users programmed their own sounds.” The DX7 was a large part of a paradigm shift in what a synthesizer was expected to offer.
What clinched the DX7’s place in synthesizer history was not just its offering all these revolutionary features, but that it offered them at a significantly lower price. Other comparably priced synthesizers, such as the analog Korg PolySix, used more derivative technologies, yet the DX7 competed with the PolySix in price while offering cutting-edge technical specifications. Dave Formula, a keyboardist for the bands Magazine and Visage, raved in April 1984 shortly after the release of the DX7, “… [the DX7] gives you so much for the price. You can compare it with things that cost six times as much, and I don’t see that much difference.” The DX7 was priced significantly lower than other revolutionary synthesizers, which played a large factor in the popularity and subsequent influence that the DX7 exerted over the popular music scene (Example 6). Bob Moog, the inventor of the Moog synthesizer, spoke of the revolution in synthesizer pricing, which the DX7 helped to spur, as a “democratization” of the synthesizer.
The first microprocessor-controlled synthesizers were strictly professional instruments, with price tags of $5,000 and up. … [Today], no matter how much or how little money you have in your pocket, you can almost certainly find a synthesizer that will give you some musical satisfaction.
As a result the number of musicians who own these instruments has increased dramatically. Early last year I went to catch a Korg clinic which was put on by Chuck Leavell at a local music store in Asheville, North Carolina. As his final demo, Chuck played the Korg Poly 800 and blew the audience away. Here was a portable 8-voice keyboard with programmability and MIDI that cost less than a plane trip to the Coast!
The fact that the event took place in North Carolina should not be overlooked. … Before then, you couldn’t really buy a synthesizer in Asheville. You had to travel three hours to Charlotte, North Carolina’s largest city. But by 1984, the salesman knew his way around keyboard synthesizers, Chuck Leavell demoed to a hundred or so musicians, and suddenly synthesizers were a musical presence in North Carolina. That’s what I mean by “democratization.”
As I discuss in my dissertation, the availability of the DX7 and other synthesizers meant that more musicians had access to unusual sounds in their music—to borrow Moog’s terminology, it “democratizes” the novelty sound in popular music. As technologies become cheaper and more accessible, the trajectory of popular music’s sound is altered.
|Synthesizer name (year)||List price (USD)||Adjusted for inflation (2017)|
|Fairlight CMI (1979)||$25,000.00||$83,100.00|
|New England Digital Synclavier II, 8-voice (1979)||$13,750.00||$45,700.00|
|Yamaha GS1 ( 1981)||$11,850.00||$33,150.00|
|PPG Wave 2.2 (1982)||$8,800.00||$22,000.00|
|E-MU Emulator II (1984)||$7,995.00||$18,500.00|
|Roland Jupiter-8 (1981)||$5,295.00||$14,000.00|
|Yamaha DX7 (1983)||$1,995.00||$4,800.00|
Example 5: List prices in USD of various 1980s synthesizers.
The DX7 is part of a collection of advances in synthesizer technology that together changed the rules for what it took to be a keyboardist. Digital technology also delivered sequencers and arpeggiators to the public. These electronic tools made the synthesizer an accessible choice even for untrained keyboardists. A sequencer is a computer that can store and recall patterns entered by the programmer. The programmer does not necessarily have to enter these patterns in real time, which introduces the possibility of programming something that would be quite difficult to perform if played live. An arpeggiator function allows a keyboardist to hold down a chord while the arpeggiator transforms the chord into a random, rapid arpeggio, hence the name for the technology. This of course makes complex and energetic arpeggios far easier to play, even while performing live. Bob Doerschuk, a prolific music journalist and regular writer for Keyboard magazine, wrote of The Human League, “Like electronic magicians, they dazzled their audiences by producing lots of music while playing as little as possible. The point of Dare  seemed to be that you didn’t even need to know how to play at all to be in a pop band—a point made many times since then.” Using these tools, one barely had to know how to play a keyboard to perform rhythmic and driving licks that would previously have been virtuosic.
This paradigm shift, unsurprisingly, motivated some music makers of the time to vocally oppose these technologies. Some saw the wild success of the DX7 as yet another example within a larger trend of moving toward simplicity and automation in keyboards and rock music. For their May 1986 issue, Keyboard magazine featured Duran Duran’s keyboardist, Nick Rhodes, as their cover story, a move that outraged those Keyboard subscribers who thought the magazine should stay focused on virtuosi (in turn focusing on the genres of classical, jazz, and progressive rock). Consider this letter to the editor, which was published in the following month’s issue: “How I miss the good old pre-DX7 days. I am so sick of reading about these Fairlight freaks and one-finger virtuosos. As a performer, Nick Rhodes is a joke. Why don’t you interview a real keyboard player, like Rick Wakeman?” The letter writer is an example of those that view technologies that facilitate performance and composition with suspicion; i.e., people who are concerned with the relationship between synthesizers, sequencers, arpeggiators, and the notion of authenticity (revealed in the letter-writer’s declaration that Wakeman is a “real” keyboard player). This letter-writer seems unaware that Nick Rhodes actually rarely used the DX7, a fact that Rhodes plainly states in the very same interview that the letter-writer complains about. The letter-writer’s misattribution of the decline of keyboard-playing society to the rise of the DX7 shows that the DX7 was the poster child for these developments, even when not directly responsible for the declinational performances.
The DX7 does not make the physical aspects of performance any easier; it has no built-in arpeggiators or sequencers to take away the hard work of moving quickly and adeptly around the keyboard. What really made the DX7 a target for criticism from music-makers concerned with “authenticity” was, instead, the factory presets. The presets meant that synthesizer players were no longer required to be experts in synthesizer programming, which lowered the bar of entry for aspiring keyboardists. Perhaps inevitably, reliance on presets was equated with inauthentic and low-quality music by many musicians. Roland Orzabal of Tears for Fears spits, “Stop using the bloody DX7 presets and invent something new of your own … People should stop pissing around with synthesizers and make fucking good music.” Philip Oakey of The Human League, on the other hand, expressed a more complex frustration with his and his bandmates’ continuing use of the DX7 presets as he reflected on the recording process during an interview with Doerschuk.
They get really bland after a while. And when we get into what I call the DX Sound Hunt, it drives me up the wall. Someone in the studio will say, ‘Okay! Let’s have a bell sound.’ Then we start going through the 128 sounds on our DX—we have the Sycologic MX1 expander board—and playing every one, including the whistle, the train, and the bombs. If we find something we like, it has probably turned up on 50 records that have been made over the past few years. But chances are we’ll give up after half an hour and say, “I guess we’d better [rent] a PPG.” Why couldn’t we get our own sound, like we did in the Dare days?
In this interview, Oakey seems to be regretful, or even in denial, of using the DX7 presets, yet DX7 presets saturate The Human League’s album Crash, released in September of 1986 and produced by Jimmy Jam and Terry Lewis, only eight months prior to this comment. Similarly, singer-songwriter Joe Walsh (formerly a member of the Eagles) publicly expresses a bit of embarrassment about using the presets during an interview with Keyboard magazine—“I very seldom go through presets. I don’t like anybody to see me sliding my finger up and down the presets on the DX7”—yet DX7 presets can be heard on tracks in his 1985 album The Confessor, released a little over a year before that comment. Walsh also uses some DX7 presets, albeit to a lesser extent, on Got Any Gum? (1987), released just over one year after the interview in Keyboard. The common attitude toward DX7 presets in the mid-1980s can perhaps be understood like that toward Auto-Tune today: everyone uses it, but no one wants to admit it.
The prevalence of the presets led musicians to believe that the DX7 was an inflexible instrument—the DX7 was sometimes thought of like a cheap Casio meant for performance only, rather than a robust FM digital synthesizer. For example, Joe Zawinul, keyboardist of the fusion band Weather Report, said, “… I don’t think [digital synthesizers are] as flexible as I want them to be. I played the DX7 for a couple of nights in Japan and I really like it. … I think Yamaha has some really good things, but I can’t deal with all the sounds they’ve got—after a while, they get stale.” Here, Zawinul seems to think that the “sounds they’ve got,” i.e., the presets, are an essential part of the instrument, rather than something that can be either used or discarded. This evaluation of the DX7 as an inflexible instrument is factually incorrect, since, as described earlier, the number of possible sounds is infinite, when factoring in the variables such as the thirty-two different algorithms and the possibility for different eight-step envelopes for each of the six operators.
But very few practicing musicians were able to successfully program the DX7 and make their own customized sounds. The most famous person to do so is probably Brian Eno. Eno was largely protective of the patches he programmed, like many programmers, since at this time patches were not copyrightable. Eno nevertheless would give small hints about his methods for making the DX7 sound more unique, as in this 1988 interview:
I’ve found ways to de-stabilize the DX7 a little bit to create interactions between it and other instruments that are more interesting … I don’t have very good voltage supply, for instance. Within the patches, I build in certain elements that don’t repeat. For instance, there’s something wrong with the programming of envelope generator four on the original DX7 and you can use that to create non-repeating patches. If you have that set to a value under 50, you’ll find that the synthesizer behaves unpredictably. Unfortunately they’ve sorted this out on the second generation of DX7s, so I still use the first one, and that’s an important element of quite a few of my patches.
Eno’s ambition and dedication to learning to program the DX7 was not the norm, however. The vast majority of DX7 players indeed seem to have relied on the presets, the Yamaha ROM cartridges, or even additional presets purchased from the cottage industry of DX7 presets for sale that germinated during this time.
A relatively small handful of presets became particularly widespread throughout the pop music of the 1980s. I have not done a rigorous statistical analysis of the prevalence of all the presets, but to me, E. PIANO 1—the DX7 version of the Fender Rhodes electric piano—seems to be the most ubiquitous; E. PIANO 1 is so remarkable a phenomenon that I have written an entire article discussing this particular preset and its effect on the “’80s sound.” BASS 1 is perhaps the next most common, and certainly at least as iconic as E. PIANO 1. BASS 1 mimics a funky slap bass, and frequently opens a track with an aggressive riff, as in “Danger Zone” by Kenny Loggins (1986). Other common groups of sounds include the DX7 “flute sounds,” such as FLUTE 1, CALIOPE, and VOICE 1; percussive sounds, like MARIMBA, COWBELL, and LOG DRUM; bell sounds, like TUB BELLS (tubular bells) and CELESTA; and plucked sounds, like CLAV 1 and HARPSICH 1.
As more and more DX7 players repeatedly returned to these presets on hundreds of records throughout the 1980s, the sound of the presets eventually became recognizable. This quickly diminished the appeal of the instrument, and I hypothesize that perhaps the DX7 sounds particularly dated today because of the existence of the factory presets; perhaps an analog instrument that was more easily customizable wouldn’t have had such an easily recognizable sound. The DX7 became less and less popular as a synthesizer toward the end of the 1980s and into the 1990s, until finally the DX7 had a truly bad reputation and was essentially synonymous with cheesiness. The Roland D-50 synthesizer, released in 1987, which blended sampled attack sounds with digitally synthesized sustain portions of tone, was the first blow to the DX7’s previously uncontested dominance. The Korg M1, an inexpensive sample-based synthesizer released in 1988, completely dethroned the DX7, even breaking the record for most units sold by any synthesizer, which had previously belonged to the DX7.
Now that digital audio workstations (DAWs), digital instruments, and virtual studio technology (VST) have made the sounds of vintage synthesizers more accessible to a larger section of the population, digital FM sounds have returned to the music scene. Modern technology also facilitates the cumbersome process of editing an FM sound, so contemporary music producers are more interested in learning about the process of FM synthesis than music makers of the ’80s or ’90s (Example 6). Even DX7 preset sounds have had a resurgence in popularity: for example, Bruno Mars’s 2016 album 24K Magic, which reached #2 in the U.S. and #3 in the U.K., uses several DX7 presets or sounds that are mimicking DX7 presets. Ultimately, whether or not 24K Magic uses actual DX7 presets or FM synthesis sounds that merely resemble DX7 presets is beside the point: the DX7 sound has made a comeback in recent years, after a slump in popularity in the ’90s and early 2000s.
While the average music consumer, or even most music enthusiasts or music theorists, may not be aware of the Yamaha DX7 as a musical phenomenon, every person who listens to 1980s Anglo/American popular music is familiar with the sound of the DX7 synthesizer. As I have shown through this constellation of remarks from music makers of the 1980s, musicians were constantly using and discussing the DX7 in their music.
 Cateforis 2011 details this evolution in Chapter 6, “‘Roll Over Guitar Heros, Synthesizers Are Here…’”.
 For a detailed history of the Moog and the Minimoog, see Pinch and Trocco 2004.
 By the late 1980s, sampling-based synthesizers had finally become more affordable, at which point they did become quite popular. Later in this essay I discuss the Korg M1, the most popular sampling-based keyboard.
 After 1986, Yamaha produced successors to the DX7, the DX7II and the DX7II FD. There were no further DX7II FDs produced after 1989, which likely solidifies the role of the DX7 as a quintessentially ’80s synthesizer. See Vail 2002, 130.
 Quoted in Frederick 1986, 46. As the interview title insinuates, Rushen played piano from a very young age. She was classically trained before she released her jazz albums, which used the DX7 and other synthesizers.
 Quoted in Doerschuk 1986c, 72.
 Quoted in Doerschuk 1987b, 85.
 Quoted in Darter 1985, 24.
 Keyboard 2003, 40.
 “Envelope” is a common term in synthesizer programming, referring to the amplitude profile of a sound. Typically, a sound begins at zero, peaks shortly thereafter during the attack (A), decays slightly (D) and then maintains a somewhat lower volume as the pitch sustains (S), and then releases (R) back down to zero.
 Chapman 1984, 70.
 The inventor of FM synthesis has published very complete and succinct explanations of synthesis via frequency modulation from a mathematical perspective, including concepts such as the modulation index, and how one might simulate various kinds of acoustic instruments. See Chowning 1973.
 Additive and subtractive synthesis both start from the understanding of timbre as made up of the fundamental and the partials above it. Different sounds are created by adjusting the number of partials and their volume.
 Quoted in Burger 1986, 16.
 This quote from Howarth also exemplifies the idea that people wanted the DX7 as a replacement for the Hohner Clavinet and the Fender Rhodes.
 Quoted in Gilby 1985, 39. Note however that despite this, Ure does use the DX7 in many of his own tracks, such as “Do They Know It’s Christmas?” by Band Aid, which I analyze in my dissertation.
 RAM (random-access memory) cartridges could be rewritten, so users could store their own custom sounds on RAM cartridges. ROM (read-only memory) cartridges were not able to be edited.
 Japan and Europe received Yamaha’s ROMs #1 and #2, while the United States received ROMs #3 and #4. The presets on ROM #1a and #3a were saved in the internal memory of Japanese/European and American synthesizers, respectively. Most of the presets were shared between 1/2 and 3/4, but some sounds appear only on one region’s ROMs.
 One precedent for this technology can be found in home organs, such as the Lowrey Berkshire Deluxe, which would have some preset sounds that were not user-programmable.
 Pinch 2005, 259. I enjoy this story, but some find it doubtful that Wakeman really had thirteen Minimoogs. I was only able to find photo evidence of him simultaneously using four.
 Théberge 1997, 75–8.
 Quoted in Armbruster 1986, 46. Briggs was primarily based in Nashville, Tennessee, and worked mostly with country music artists.
 Théberge 1997, 75. This number is calculated based on the number of Sequential Circuits Prophet-5 synthesizers received at the factory for repair that still had the factory presets saved to he internal memory. However, Théberge notes also that there may be another explanation for the prevalence of factory presets on the Prophet-5s: programmers may have purged their own custom sounds before sending the synthesizers back to the factory to avoid any infringement by the company on the programmer’s intellectual property. Sounds were not copyrightable at this time, so a company could conceivably steal a programmer’s sound and sell it themselves. Théberge notes that a personal informant attested to this.
 Quoted in diPerna 1984, 40.
 Moog 1985, 42.
 This is one of the senses in which new wave music, and eventually mainstream 1980s pop in general, grew out of the DIY aesthetic of the punk scene in the late 1970s.
 Steve Cross, letter to the editor, Keyboard (June 1986): 6.
 Doerschuk 1986b, 76.
 Quoted in Rue and Goodyer 1990, 53.
 Quoted in 1987a, 88–9.
 Jimmy Jam and Terry Lewis’s aforementioned preference for the DX7 no doubt have to do with the reliance on DX7 sounds on Crash. I hear BASS 1 and BRASS 1 on “Money”; BRASS 3 on “Jam”; VOICE 1, BASS 1, and KOTO on “Are You Ever Coming Back”; BRASS 1 and BASS 1 on “I Need Your Loving”; BASS 1 and SYNBRASS 1 on “Party”; E. PIANO 1 and SYNBRASS 1 on “Love on the Run”; BASS 1 on “The Real Thing”; and LOG DRUM on “Love Is All that Matters.” I think “Human” is likely using DX7 bell and plucked sounds, but I can’t identify which.
 Quoted in “Uses and Abuses of Synthesizers” 1986, 69.
 To my ears, BASS 1 is used on “I Broke My Leg,” FLUTE 2 on “Bubbles,” and CALIOPE (sic) on “Slow Dancing,” all tracks from The Confessor. GUITAR 4 I think is also used in “The Radio Song,” CLAV 1 in “Up to Me,” and XYLOPHONE in “Time” on Got Any Gum?. Walsh might also be using CALIOPE in “Memory Lane,” but produced somehow to sound less sharp—this sounds a bit Mellotron-ish to me, actually.
 Quoted in Armbruster 1984, 55. Zawinul was discussing his preference for the Sequential Circuits Prophet-5 over all types of digital keyboards, both samplers and FM synthesizers.
 Théberge 1997, 76.
 Quoted in Diliberto 1988, 54.
 “CALIOPE” is spelled as shown in the Yamaha ROM cartridge, rather than correctly (“calliope”). Colloquially, CALIOPE is frequently called a “pan flute” sound, but this should not be confused for the actual preset PAN FLUTE, a sound that was only available on the American ROM #4A.
 For example, see Brian McCalla, “What sonic qualities make a synth a cheesy ’80s synth?”, Quora, last modified January 8, 2017, https://www.quora.com/What-sonic-qualities-make-a-synth-a-cheesy-80s-synth.