Software Synthesizers (Excerpt): Top Music Technology Pages Return to MusicWords Home

[The material below is excerpted from Software Synthesizers, published in 2003 by Backbeat Books.]

In the broadest sense, any musical instrument that makes its sounds electronically can be called a synthesizer. We'll want to narrow that definition a bit -- but first you need to understand the difference between "electrical" and "electronic."

An electric guitar is not an electronic instrument. Why? Because the sound originates in physically vibrating strings, which are excited (by the fingers or a pick) in a mechanical way. Put your ear close to the strings, and you'll hear the music even if the guitar is unplugged. The same is true of a genuine electric piano: Physical tines or reeds inside the body of the instrument are struck by physical hammers, thus producing tones. The tones of the strings, tines, or reeds are captured electrically with electromagnetic pickups so that they can be amplified. (Note, however, that most electrically operated pianos built today are electronic, not electric: They have no physical reeds or tines. Their "electric piano" sounds are electronically simulated.)

In an electronic instrument, the tones originate in an electronic circuit of some sort, be it analog or digital. If you were to listen to the circuit, you'd hear, at most, a faint background hum and the whir of the cooling fan. The tones become sound only when the output of the instrument is sent to an amplifier and emerges from a loudspeaker.

All synthesizers are electronic instruments, but not all electronic instruments are synthesizers. To qualify as a synthesizer, an instrument has to have one of two characteristics: (1) Its sound-producing circuitry offers a significant number of options to the user, enabling the user to shape the tone color as needed, and/or (2) it makes a class of sounds traditionally associated with synthesizers.

Not all of the instruments discussed in this book are synthesizers. In Chapter 2 we'll meet some software that doesn't conform to either of the definitions above, and in Chapters 5 and 7 we'll encounter instruments that do conform to definition 1, but that aren't usually called synthesizers.

Don'tcha just love hair-splitting arguments over terminology? Let's dig a little deeper, and see if we can clarify matters.

The first synthesizers that became widely known to the public, in the late 1960s, used analog electronic circuits -- circuits built of discrete resistors, capacitors, diodes, and so on. The Minimoog is perhaps the best known example, but more than a dozen similar instruments were in production by 1975. They had a limited, but very distinctive, palette of sounds. Today, any instrument that makes this type of sound arguably qualifies as a synthesizer, whether or not it has the types of sound-shaping controls associated with those early instruments and whether or not it uses true analog circuits.

In the early 1980s, some other types of synths appeared. Yamaha produced a series of synths that used FM technology, the leader of the pack being the DX7. (I'll have more to say about FM and other types of synthesis in Chapter 8.) The FM sound was quite different from that of analog instruments, but FM synths still provide plenty of user control over the sound, so they meet the definition: They're synthesizers.

At about this same time, the falling price of computer memory and the availability of faster digital circuits made it possible to store sampled sounds digitally. (A sample is a recording of an actual sound -- see Chapter 5.) Synthesizers such as the Roland D-50 and Korg M1 began to appear; they made sounds by playing samples stored on ROM (read-only memory) chips. For instance, a synth might have recordings of real trumpet, guitar, and electric piano notes, which could be played from the keyboard. These instruments were also known as synthesizers, but only if they offered the kinds of tone controls usually associated with their analog predecessors. An instrument that played samples from ROM but didn't have synth-type tone controls was not generally considered a synthesizer, except perhaps by the marketing department of the company that built it.

A year or two later, we began to see instruments such as the E-mu Emulator and Ensoniq Mirage, which could store sampled sounds in RAM (random-access memory). New samples could be loaded by the user from a floppy diskette, or recorded directly into the instrument using a microphone. Although these instruments typically had the kinds of tone controls associated with synthesizers, they weren't called synthesizers. They were called samplers. Samplers are still an important type of instrument; several software samplers are covered in Chapter 5.

We'll meet other variations on the formula as we go along. You won't go too far wrong, though, if you look at it this way: It's a synthesizer if it sounds like a synthesizer, or if it has the types of controls usually associated with a synthesizer. But if it lets you load the pre-recorded sounds of your choice from a disk or other external medium, it's a sampler, not a synthesizer.

A note for insiders: Yes, you're right. This handy rule of thumb pretty much ignores physical modeling synthesizers. The Yamaha VL-1, which was the first commercially available synth based on physical modeling, did not have most of the traditional types of controls, nor did it sound like a traditional synth, but it was a synth all the same.

Why the confusion? Basically, we're trying to hit a moving target here. Synthesizers are not the same today as they were ten years ago, or twenty, or thirty. Those fiendishly clever manufacturers keep coming up with new variations on familiar themes, hybrids that combine elements of disparate technologies, and (once in a while) radically new approaches to sound production. My advice: Relax and enjoy the ride.