Monotimbral instruments such as guitars, pianos, or even the voice can only produce one sound at a time. Multitimbral instruments, on the other hand, can produce multiple sounds simultaneously, opening up the scope for complex elements to be crafted together in ways that would be difficult or impossible using conventional monotimbral signal paths.
Significantly, this allows one person to do what previously had taken many people to do—or, at the very least, what it took many instruments to do. If you want a fat arpeggiated bassline underneath your silky chords without using two synths, you need a multitimbral instrument.
Back in the '80s and '90s, a synth's multitimbral nature used to be a much-hyped selling point, even when capabilities were relatively meager and expensive. Now multitimbrality is arguably a more powerful feature in synthesis and sound design than ever before, but it's not touted as often. The power of DAWs makes some of the great new functionality found in hardware synths seem limited by comparison. But as you'll read below, there's plenty of reasons to learn about and use multitimbral synths.
Multitimbrality made its name in the mid-'80s as synthesizers began to gain enough voices—and gain enough control over those voices—to make proper multitimbral synths possible and affordable. But the idea goes back much further.
To use a grandiose example, an orchestra could be viewed as a multitimbral instrument. When composers write a piece, they take advantage of the variety of tones contained within the singular structure of the orchestra in order to create their piece. As technology and keyboard instruments progressed, efforts were made to put the tones of many players under the control of one pair of hands.
From the 12th century, pipe organs offered multitimbrality, putting control over a great many sound generators into a keyboard instrument, with different timbres to be mixed and matched according to taste and purpose. Mozart called the organ "the king of instruments," hardly surprising considering their vast size and incredible range of sound. Organs could have thousands of pipes—the world record is around 33,000—and elaborate consoles with thousands of switches. The biggest could hit a fundamental note of 8Hz, which is subsonic, but sure to put the fear of god into your bowels.
In the early 1900s, symphonic and theater organs were designed to replace entire orchestras, and they could play back complex pieces automatically from paper piano rolls. Before the advent of electronic sound generation, these were simply the best solution that the technology of the time could offer when it came to putting many sounds under the control of one person.
Electronic organs and their silkier cousins, the string synths, carried these ideas into the '60s and '70s.
While a new breed of Minimoog-influenced synthesizers introduced a host of exciting sounds to the world, these were largely monophonic and monotimbral. However, plenty of electronic organs and string synths retained the polyphony, the voice-stacking architecture, and the nomenclature of their pipe-organ ancestors. They could create a richness in sound by combining different tones—and, in some cases, totally different voices—on top of each other.
The distinctive keyboard line at the top of "Lucy In The Sky With Diamonds" was courtesy of a Lowrey organ with a combination of its sounds that were designed to simulate stringed and struck instruments. And on the famous Solina string synth of this era, the sounds available to combine were contra bass, cello, viola, trumpet, and horn. Not that they really sounded like their namesakes—but the terminology was hinting toward the future.
It would be some time before the various branches of electronic instruments were fully realised in one keyboard. The main obstacles to overcome were cost, practicality, and reliability—and millions of dollars were spent trying to create the ultimate complex synthesizer.
Luckily, it was the '70s, and musicians had big money to spend. Korg's PS-3300 (like three stacked PS-3100s in one synth), launched in '77, was truly a product of its era.
It was a beast of a synthesizer, containing a staggering 144 VCOs, 144 VCFs, 144 VCAs, 144 envelope generators, and three bandpass resonator banks. These were divided among three identical blocks, each of which could be individually sculpted and then combined at the final output for some extremely impressive sounds.
I can honestly say I've never played anything quite like it. For example, the resonator section on each block gave you three bandpass filters, which could be swept and modulated to provide vowel and phasing sounds—something very difficult to achieve with other synthesizers, even today.
For Yamaha's part, the company had a bit of a trend toward the end of the '70s and into the '80s, putting two of the same thing in one box for their higher-end synths—as evident in the company's CS-30, DX1, and the infamous CS-80.
These all effectively had two of the same synth controlled from the same keyboard, with the ability to balance the two sections to taste using a simple slider, or even to output the two voices separately for individual processing.
When using a dual system such as this, we use the term bitimbral, but it's still effective in creating more interesting sounds than would otherwise be possible. The fact that this feature carried over into several of Yamaha's keyboards implies the engineers knew the value of such a design.
In the analog domain, along with more voices came more circuits, more weight, and more cost. It's important to remember that when dealing with multiple voices and true polyphony, there is an awful lot of control information needed—and this was early days for even basic microprocessors.
Analog synths, therefore, remained largely monotimbral or bitimbral through the '70s and '80s, and keyboard players—especially those who didn't have to load their own gear on stage—were happy to surround themselves by great castles of expensive keyboards.
Through the '80s, advances were made and some amazing analog polysynths were created, but they were largely monotimbral, looking to refine an existing idea. Then along came accessible digital synthesis, and a whole new paradigm was born.
In 1987, Roland's distinctive D-50 was built around a new kind of engine that combined synthesis with samples, putting multitimbrality at the very heart of its architecture.
Patches were constructed by layering together up to four sounds or "partials" and then using a joystick to change the balance between them, a wonderfully new and expressive way to manipulate sound. Partials included traditional synthesis waveforms as well as sampled sounds, setting a template for sound design that's still in use today.
As with Yamaha's DX7, the D-50 captured the attention of many keyboardists of its time and can be heard on everything from Enya and Michael Jackson to the theme tunes for The Simpsons and Star Trek: The Next Generation.
It has even experienced something of a resurgence over the last few years following its reissue in boutique form in 2017 as the Roland D-05.
It's no surprise that Eric Persing, one of the D-50's original sound designers, went on to found Spectrasonics, whose Omnisphere plugin is one of the leaders in contemporary sound-layering engines.
Just a year later, in 1988, Korg's M1 took the D-50 concept and ran with it. It had what for the time was a huge 4MB memory that contained exotic samples put together by an international team of sound designers, marking the first time a synth had been shipped to all countries with the same presets.
Again, the onboard samples could be combined with other samples and more standard waveforms for rich sound design—making it possible to do curious things like layer a woman's scream with a blown saki bottle-top and a saw wave. How about a selection of drum kits, too? Suddenly we were back in the world of the quirky theatre organ.
To top it all off, the M1 had an eight-track sequencer that allowed up to 10 songs to be assembled entirely within the keyboard. This was an incredibly liberating experience for musicians at the time, as it negated the use of recording equipment or an external MIDI sequencer to construct a song. It's no wonder it became the best-selling digital synthesizer of its time.
Where the D-50 set the template for sample and synthesis-based synth engines, the M1 was the model for multitimbral workstation keyboards. From this point on, digital reigned supreme and analog went into decline. Cheaper memory and more powerful processors meant that, suddenly, every conceivable sound—and many inconceivable sounds—were in easy reach of most musicians.
The list of workstations is long and fairly uninteresting, essentially leading us to our current paradigm, the do-it-all DAW. It's not that plenty of workstations didn't sound good. Many sounded great, and they were to be found all over the '90s and '00s hip-hop and R&B classics.
However, I'm sure I'm not alone when I look at those workstation behemoths and don't feel particularly enamored. They have their place, but as technology has progressed we've seen more interesting synths that use multitimbral techniques to deliver big and complex sounds.
Elektron is now synonymous with interesting synthesizer design. Almost all their machines could be described as multitimbral, and their synthesizers especially embrace the idea, often using four identical synth blocks independently or combined.
The Analog Keys (and its keyboard-less cousin the Analog Four) have four analog monosynths in one box, and you can play all four voices in unison with totally different sounds on each voice—and each voice can do as much as any modern monosynth. You can create a mono patch with eight oscillators (16 if you include the sub-oscillators and noise generators), four low-pass filters, four multimode filters, eight multimode LFOs, eight envelopes, and three bus effects.
Elektron's Digitone and Digitone Keys are capable of similarly monstrous sound stacking, but its sound generation comes from a very accessible implementation of FM synthesis.
Putting one eye on the future, Novation's recent AFX update to its Bass Station II keyboard turns the multitimbral concept sideways. Aphex Twin suggested putting a different sound on every key, rather than loads of sounds on one key, which its monotimbral architecture couldn't handle.
That allows you to do many things, such as making subtle variations in a sound in order to spice up a performance, or even creating analog drum kits that are spread across the keyboard. It's a fantastic and innovative development that takes advantage of modern processor power, and I do hope it finds its way into more synths.
While Elektron and Novation cater to the high-quality-but-affordable end of the market, there are a couple of recent high-end keyboards that are starting to revisit the multitimbral architecture of the D-50, now armed with 2019 processing power and memory.
Both the Dave Smith Instruments Prophet X and the Waldorf Quantum are designed around sampling and synthesis engines in a thoroughly 2019 fashion. Unsurprisingly, they've been lauded for their sound design potential.
Both are only bitimbral, but considering the different synthesis engines and huge sound libraries (the Prophet X packs a whopping 150GB of samples), even the potential to stack two patches will give you a massive sound. Many of Dave Smith's designs are at least bitimbral, and this feature has saved me many headaches on stage. The Moog One is admirably tritimbral, as it should be for such a high-end piece, and pretty much out-specs any analog polysynth that came before. It's an exciting time in keyboard land.
I've covered here the story of mainstream multitimbral synths, but there are a few interesting outliers worthy of your attention.
Yamaha's TX81Z is an authentic DX engine in a 1u rack and is eight-part multitimbral. I bought mine from a music store in New York, where the owner told me they used them just to fill up rack space.
Cheetah's MS-6 is a rare beast, but it offers six voices of analog synthesis, all of which can be stacked and detuned.
Roland's MKS-7 is interesting because it contains the brains of a Juno-106 and a TR-707 in one rackmountable module, amounting to a very cost effective way to get that vintage Roland sound without having to pay today's prices.
There's also the Roland JV-1080, which offers 16-part multitimbrality and a sound library from yesteryear, including many of the classic D-50 sounds, in a very affordable rack unit. It can still be heard today on records by artists such as Tame Impala.
While the all-singing, all-dancing DAW has somewhat negated the need for eight or 16-part multitimbral instruments, it's clear that some means to play with and stack different voices is a valuable part of any synth's sound-design capability. A singular signal path can only do so much, and there are great rewards to be found when you try different combinations of sounds.
In an age where most of us make music looking at the same machine that we use to watch TV or write emails, multitimbral instruments provide us with a welcome break away from the screen to build complex sounds and compelling starting points. They allow us to really delve into the instrument, to think about sound in an architectural way, and to do all that in a self-contained system.
About the author: Tom Furse is a synthesist, composer, and producer, best known for his work with The Horrors.