Music software developer Sam Tarakajian on the trials and tribulations of the ultimate in DIY
As part of the Holly Herndon-guest edited Music Takeover for our Hack Your Future week, we got Sam Tarakajian, a developer for Cycling '74, the company behind the real-time performance software Max/MSP to present a How To guide of building your own electronic instrument. Besides working in instrument design and multimedia performance, he has spoken at the Code Control Festival on the subject of falling in love with soulless machines, and recently unveiled the new Mira controller for Max at NIME in Daejeon. For people interested in learning more about Max, Sam has compiled an enthusiastic if not entirely accurate tutorial series on YouTube under the pseudonym dude837.
Step one: give up. Building an instrument is hard. Really hard.
Think about all the things that we want out of a musical instrument. What we really crave is a living, breathing musical being. We want something we can fall in love with, something that we can spend hours playing and never get bored, something that we can continue to probe for new sounds, even as tastes change and civilizations rise and fall. In short, we want the impossible.
What we want will continue to be impossible, so long as we look at the instrument as something that can be completed, as an artifact unto itself. It's an easy mistake to make; after all, we talk about the violin as if it were an atomic, self-contained something. But it's exactly this kind of thinking that makes our task impossible. If we think of an instrument not as a physical entity but as an idea, a phenomenon spread across a performer, an audience, a thing being played and even the cultural context in which they all appear, then it's actually much easier to see where we need to focus our efforts.
No matter how we think of building an instrument, we have to start by deciding how we're going to interact with it. Are we going to strike it or bow it? Will it be a sensor array that detects visible light, or a ping-pong ball that houses an accelerometer? We also need to think about the kinds of sounds our instrument is going to make. Will it use sampled or synthesized sound? Should it model an acoustic phenomenon, or will it bring us impossible sounds from the digital world? Eventually, we will arrive at the most important and challenging step: to define a mapping between the inputs to the instrument and the sounds it produces. This is where the soul of the instrument really starts to take shape. This is also the part that's impossible.
No one will ever get this mapping exactly right, which is why so many instrument builders get stuck here. Standing at the vast, inscrutable gulf that is The Mapping Problem, many people turn around to focus on easier challenges, like incorporating more sensors or squashing bugs. We can overcome this anxiety if we understand that an instrument is not a step on the path to performance; rather, performance is a step on the path to an instrument. If there is one secret to building an instrument it is this: practice does not make perfect. In fact, you should try to perform a piece right from the moment you're able to make sound, long before you feel you or your instrument is ready. After all, the only way to evaluate an instrument made for performance is to see how it holds up in the context of performance. And you yourself are part of that instrument. The way you hold it, the way your body listens to feedback from it and the gestures you deploy to pull sound out of it are all part of the instrument. If it needs more sensors or more sounds, then performance will show you what changes you need to make.
There is no simple recipe for building a musical instrument, any more than there is an easy guide to falling in love. To build a musical instrument, begin by trying to build a musical instrument and engaging with an audience. Take careful note of the experience, then modify your instrument accordingly. Repeat. The task can seem overwhelming, so long as you insist on trying to get everything right the first time around. But you can think of building an instrument like a two day vacation in a foreign country. You aren't going to be able to see everything, so just do the best you can. Wear comfortable shoes, bring a notebook and, most importantly, start planning your return trip.