Eli Fieldsteel is a composer currently working and teaching at The University of Texas at Austin. I became aware of Eli's work through his fantastic series of YouTube tutorials on SuperCollider. Fieldsteel's accolades include first prize in the 2012 ASCAP/SEAMUS student commission, the 2014 ACCBDA Grant for Young and Emerging composers, as well as awards and recognition from other organizations, including the Bandmasters’ Academic Society of Japan and the Frank Ticheli Competition. Fieldsteel’s music and research reflects an ongoing interest in the intersection between music technology and contemporary instrumental practice, covering topics such as human-computer improvisation, interactivity, and generative music.
Carl Testa: When did you first become aware of SuperCollider? How long have you been using the software?
Eli Fieldsteel: My brother introduced me to SC sometime around 2007. At that time he was an undergraduate mathematics major at Wesleyan University, and I was an undergraduate music major at Brown University. The only thing we knew how to do was run the short code examples in the help files. I remember the phase modulation example in the SinOsc documentation completely blowing our minds. In 2009 I was a graduate student at the University of North Texas, where my friend and colleague L. Scott Price had a leg up on SC. He got me started with a few things and I was able to get a reasonable grasp on it. I've been using SC ever since.
CT: Did you have a background in music synthesis and/or programming before using SuperCollider or did you start from scratch?
EF: I come from a family of mathematicians (I was at one point a math/music double major), and I like to think that I've always had a mind for programming, but I had almost no experience in either synthesis or programming when I started taking SC seriously. In college, I took a primitive C programming course as well as a course on the fundamentals of electronic music, but that's about it. I was very fortunate to have taken an SC summer course at Wesleyan with Ron Kuivila and James Lipton in 2010, and I studied Csound with Russell Pinkston during the following academic year. Those two courses really got me off the ground, and I've been more or less on my own with SC ever since.
CT: Are you primarily a developer or a user? Or both?
EF: I'm primarily a user. I have developed a few classes and standalone applications, some on commissions and some for my own use, but nothing which has been included in any of the public SC releases.
CT: What do you feel is the greatest strength of the software?
EF: Flexibility. SC doesn't pin the user down to one particular workflow or programming style. On the contrary, using SC is always an incredibly fluid and dynamic experience. SC readily adapts to users' ideas, not the other way around.
CT: What is the most frustrating thing about the software?
EF: What's frustrating for me is seeing how much difficulty newcomers have with finding a foothold. I've heard complaints about the "unhelpful" help documentation and general lack of learning materials. But of course, this is the curse of such a flexible and powerful piece of software. As there is no singular approach to SC, it's difficult to write a truly instructive and comprehensive manual that doesn't railroad the reader into a particular style. It's for this reason that I started an SC video tutorial series on YouTube. I wouldn't claim that my approach is right for everyone, but I hope it's perfect for some.
CT: What could be changed about the experience of opening SC for the first time that might make it more friendly for absolute beginners?
EF: Hard to say. Certainly it might be worth providing links and citations for various SC learning resources on the main "Help" documentation page, and making sure these resources stay current with each release. But as I mentioned, SC is a deep program with many different entry points, so more introductory tutorials might in fact make for a more overwhelming first impression. All things considered, there's no substitute for a new user's sincere patience, diligence, and dedication (which of course is true for mastering any new skill).
CT: I think you have become quite well-known for your superb series of YouTube tutorials. What has the feedback from that series been like?
EF: Overwhelmingly positive! I've received lots of comments, messages, and emails carrying praise and appreciation.
CT: How do you use SuperCollider in your music?
EF: I use SC almost exclusively for electronic music. In earlier pieces, I'd craft sounds in SC and transfer them to a DAW for fine, non-realtime adjustments, but these days I've almost completely shed my reliance on track-based software. Although, to be fair, the DAW plays an important and convenient role in contemporary electronic music for which there's practically no substitute. More recently I've started designing visual interfaces with SC's GUI classes, so that SC serves as both the sound engine and performance conduit. Now if only I could get it to brew my morning coffee.
CT: Has using SuperCollider changed your approach to music?
EF: Definitely. In particular I've embraced the idea of constrained randomness in music. Some composers spend tremendous time and effort fine-tuning a single gesture until it is exactly what they want. I respect this approach very much, but SC has encouraged me to do quite the opposite and relinquish control. I compose and adjust the algorithms, and I let the sound do whatever it does. Sometimes the result is wonderfully surprising, and then I never hear the same result ever again. It's very refreshing.
CT: Do you come to SC with a particular idea that you want to execute? Or do you often come up with ideas in the process of working?
EF: Both, really. I usually start coding with partially formed ideas, but when code becomes sound, the result rarely sounds exactly as I'd imagined. But I don't treat these discrepancies as miniature failures, on the contrary, I treat them as points of departure. I've made some of my most interesting sounds by trying something that looks "wrong" (or by pure accident).
CT: I find Supercollider to be one of the most powerful sequencers ever designed, what with the included Pattern Library of classes. Is there another aspect of SC that you find very powerful?
EF: I don't find any aspect of SC to be quite as powerful as Patterns. In combination with creating custom Event types, patterns can sequence any type of data, not just sound. The paradigm is open-ended enough to be wonderfully flexible, but not so open-ended that the paradigm becomes meaningless. I'd say multichannel expansion is a very close second, as far as powerful features go. Syntactically, it's an elegant way to produce rich, complex sound with minimal effort. As far as I know, there's no comparable feature in Csound, and using poly~ in Max/MSP is considerably more time-consuming.
CT: Do you use SuperCollider in combination with other software or hardware?
EF: Sometimes, but not very often. As far as software goes, my recent interactive pieces use SuperCollider exclusively, although several years ago I would bounce audio back and forth between SuperCollider and Logic Pro during the compositional process until I was satisfied with the results, then I'd load the audio back into SC as sound files for playback. For hardware, I usually rely on a simple MIDI footswitch for triggering. For live pieces which I perform myself, I'll usually use some other simple MIDI controller with a few knobs and buttons. I bought a Korg NanoKontrol a few years ago and I've been very happy with it. For fixed media pieces, I still use a DAW for the final production stages, since that's what DAWs do best. I do have one interactive piece that included four motion sensors affixed to dancers' bodies, but that's not a regular thing for me, at least not yet. I would love to experiment with new hardware controllers, if I ever have the opportunity.
CT: Would you like to discuss current musical projects you are working on? Or upcoming performances/releases?
EF: My main project right now is a composition for wind ensemble and live electronics, which is just about finished. I'm working under a grant from the Atlantic Coast Conference Band Directors' Association, which is made up of twelve schools along the east coast. A laptop performer controls the electronic part, whom I treat like any other member of the ensemble; he or she reads from a notated part and follows the conductor. The computer keys trigger and sustain sound events.