What I learned about music when I started making it

Published 24 Mar 2021, last modified 24 Mar 2021

First off, you may notice it’s been a long time since I updated my blog. I’ve written only two other posts since the birth of my youngest child, when, as one might expect, all my habits were interrupted. Disorganized as I am, it’s taken me some time to resume some of what I had been doing. I haven’t touched my podcast in quite a while.

One habit I dropped when the child was born was guitar practice. I bought my first guitar (acoustic) last year, hoping that strumming chords would make accompanying my own singing come more naturally than it did when I was trying to play piano/keyboard, which I have done since I took lessons as a child. I can read sheet music, but I’ve always struggled with my clumsy hands when playing keyboard arrangements at any but the slowest of tempos. I didn’t really get to give guitar a proper chance before I started holding a baby all the time again, but I certainly didn’t find it immediately easy to make the sounds I wanted to make on the instrument. Any guitar player can tell you that it takes time to train one’s fingers to contort into the correct position, to develop the callouses that make it less painful to fret the strings, to translate chords into muscle memory quickly enough to play music at a consistent tempo.

Having shelved my attempt to learn guitar, I found myself studying other ways to make music. In particular, I started seeking out YouTube videos of musicians developing different voices on the Korg Minilogue, which I had first discovered through a video where Anodyne Institute demonstrates how to reproduce the lead voice of Boards of Canada’s “Turquoise Hexagon Sun.” This led to me studying the basic components of the synthesizer—audio-rate oscillators, low-frequency oscillators, low-pass filters, and so on—and then I remembered this strange little computer application someone had recommended to me a while before, VCV Rack. VCV Rack was supposed to be a musical synth application, but the first time I had looked at it, I couldn’t make heads or tails of it. I managed to use my computer keyboard to play little melodies on the default settings it loaded on startup, but producing anything other than this default sound seemed to involve arranging rectangular “modules” covered in confusing acronyms like VCO, VCF, LFO, and ADSR, and connecting various parts of them with virtual “patch cables.” It was supposed to simulate a synth system called Eurorack. So when I remembered VCV Rack in the context of learning the synth basics, I turned toward studying Eurorack, and started learning about the hardware modules that musicians use in Eurorack systems, from classic Doepfer modules that have been around since 1996 to exciting new hardware from small brands like Mutable Instruments, Instruō, and Make Noise. It was intensely interesting to me, but also quite expensive. So eventually I decided to try VCV Rack again,

Now I have a whole album, representing my first earnest attempt to record music in my adult life, which I created entirely in VCV Rack. Here are a few things I learned about modular synthesis and music in general while doing it:

Modes are a great shortcut into understanding western harmonies.

My real breakthrough with working western tonal harmonies into my music came from using a software clone of the harmonàig Eurorack chord quantizer from Instruō. From a single root note, which can be manually played by the musician or derived from random voltages delivered through a patch cable, the harmonàig can produce the notes of a chord according to some modal scale. There are better sources to explain the seven modes of the major scale, including the manual for the harmonàig, but let it suffice to say that each mode is essentially a set of seven out of the twelve notes in every octave of the chromatic scale (the scale represented by the keys of a piano). By constraining passages of music to the notes in one of these modes, you can create or just stumble upon interesting melodies and chord progressions that incorporate popular western tonal harmonies. People tried to teach me the modes back when I was taking piano lessons, but it didn’t “click” for me at all until I was configuring the harmonàig to to produce chords in these different modes and allowing it to do most of the math of that for me. Once I’d started to get the hang of them, I found them to be a great tool for writing chord progressions and improvising melodies on the keyboard.

Controlled randomness is a powerful muse.

In improvisational jazz, controlled randomness or spontaneity produces engaging musical performances that never sound the same as the last. Various forms of controlled randomness are also important components of modular synthesis techniques.

Firstly, what we perceive as acoustic noise is just especially random, sustained vibration. A noise source has many potential uses in a modular synthesizer system. When the amplitude and other factors of the noise are limited by a short envelope, noise becomes the basis of percussion. When noise is processed through a band-pass filter that reduces the amplitude of all but a narrow band of frequencies, within that noise, it takes on the sound of whistling at a particular tone. Noise is also a common input for “sample-and-hold” hardware, which, whenever some trigger signal occurs, takes a snapshot of the voltage coming from the noise generator and continues to send that same exact voltage to other pieces of hardware until the next trigger signal occurs. In a modular synthesizer, these stepped, random voltages can control anything that is voltage-controlled, from the frequency of an oscillator (the rate at which it’s vibrating, which you may perceive as the note it’s playing, or as the rate of an up-and-down modulation on some other aspect of the sounds the synth is making) to the decay time of an envelope generator (how long a sound takes to fade from its initial “attack,” whether in volume or in some timbral characteristic). A musician using this technique to produce random tones can even constrain them to a particular musical scale by passing the random voltages through a quantizer configured to convert the input voltage to an output voltage corresponding to the nearest note on that scale. A modular synthesizer can therefore produce even highly melodic, harmonious, and structured sounds out of noise.

There are other ways to introduce randomness, pseudorandomness, or at least unpredictability into a modular synthesizer. One VCV Rack plugin contains a wonderful “mechanical chaos source” module called Caudal, which produces smoothly-but-unpredictably rising and falling voltages based on physics simulations like double-pendulums. I even made some tracks where all the notes ultimately derived from feeding the voltages from Caudal into the harmonàig, producing very harmonious but completely spontaneous melodies and chord progressions. Similarly smooth-but-unpredictable changes can result from using otherwise predictable low-frequency oscillators to control their own or each other’s frequencies, or by layering unsynchronized, low-frequency oscillators operating at different frequencies.

Modular synthesizers reward experimentation.

In a modular synthesizer system like Eurorack, where connections and control voltages all conform to a common standard, a musician can be fearless about making cable connections between modules that may not have been envisioned by the manufacturers, because they know it’s safe to do so without ruining the hardware. All information in the system is just voltage within a certain, agreed-upon range, so one can run a cable from a sound source like an oscillator or a noise generator into a “control voltage” input or “clock” or “trigger” input that controls some parameter of another module, something one would typically control using something more steady like a low-frequency oscillator, and—who knows?—something interesting might happen.

Mutable Instruments produces a module called Branches that implements a “Bernoulli gate,” primarily intended to operate on trigger or clock pulses and randomly route them to one of two outputs with a probability factor the musician can control. But passing acoustic noise into Branches results in a sound that becomes more sparse and crackly depending on the probability factor, similar to an effect one might observe while turning a continuous tuning knob on a radio from a frequency with no discernable signal to one where the station signal is strong. Mutable Instruments advertises this effect as a feature of the hardware.

Filters are made primarily for softening some of the harsher, “buzzier” overtones of a sound, but when their parameters are pushed to extreme levels, many filters will “self-oscillate,” meaning they produce an audio signal of their own, typically a harmonically pure sine wave. Many synthesis techniques rely on combining this sound of the filter itself with the sounds it is filtering.

Experimenting with modules in this way and exploiting even their seemingly unintended behaviors is a major component of modular synthesis subculture. Using one’s instrument the “wrong” way is what it’s all about.

There are additive and subtractive approaches to creating music.

When people talk about different approaches to music synthesis, they often speak of “additive” and “subtractive” patches. In this context, additive usually denotes techniques that introduce more harmonic overtones and timbral complexity to a sound, perhaps using hardware like a wavefolder, where as subtractive usually denotes techniques that involve selectively limiting or suppressing some of the harsh, buzzy overtones in a sound source like a saw wave or square wave from an oscillator. But we could also use these terms more broadly to describe additive and subtractive approaches to creativity in music.

The additive approach to creativity, as I have experienced it, works like this: You make a sound you like, and then you realize it would sound even better with a certain other sound, and once you’ve added that to the mix you realize there’s still another sound you could introduce that would tie this all together a little more nicely, and so on. The subtractive approach starts with just making a lot of sounds, throwing everything at the metaphorical wall to see what sticks, and then moves on to refining the resulting cacophony down to just the combinations of sounds you really like. But to describe these as disparate approaches is misleading; I find that typically I need to do a little of both every time I make music.

Music is just making sounds you like.

If, like me, you grew up with “western” tonal music, you might be aware that it has a set of rules constituting what is termed its music theory. But these are more like guidelines than actual rules. Moreover, there are various non-western tonal systems in the music of the world, and various atonal kinds of music, and really, when you get down to it, music is just sounds you like. What people tend to like in tonal music is partly a mix of patterns and surprising changes in the frequency intervals between various sounds. Most people can instinctively hear when one tone has a vibrational frequency that is precisely double that of another tone; we perceive that as being one octave higher in western scales, and people also pick up on smaller mathematical ratios between tones on the same instinctual level. But there are so many other elements of sound that people observe and enjoy in a similar way: timbre, rhythm, and dynamic amplitude, to name a few.

Music is math, but it’s a spooky, alchemical kind of math you have to feel instead of doing it out with paper and pencil, one where the concept of correctness does not apply.

So if you can make sounds that play with these elements in a way that moves and excites you, and that you are motivated to share with other people, you can make music—whether that takes the form of harsh noise, rap, or violin concertos. I spent a long time thinking I couldn’t really make music because I didn’t understand music theory. But an instinctual appreciation for the qualities of sound comes first, and when you’re developing that, you can start to build a theoretical framework around it.

So get out there and make some sounds!


There are no comments on this article.

Leave a comment