Interlude Consortium’s competition-winning MO makes everyday objects interfaces and does some surprisingly-sophisticated analysis of gestures.

Nearly as long as we’ve had electronics, musical inventors have tried to imagine new electronic instruments. In the crowded world of new instrument design, the Margaret Guthman Musical Instrument Competition has emerged as a key prize for the best work, with creations battling fiercely for attention.

But in the oddball world of sound and music, how do you judge a winner? As a starting point, organizers this year asked the judges what they personally found important. With an expert panel including synth pioneer Tom Oberheim and reacTable creator Sergi Jorda, those answers are themselves revealing.

As for the competitors themselves, even with eclectic entrants, one theme stands out. Human gesture and performance presence is a strong dimension of the winners. And in perhaps the most promising first-prize winner yet, research begins to crack the code of how to make real gestural analysis work, even allowing everyday objects to become musical instruments.

To help us learn more, Competition founder and Georgia Tech Music Technology director Gil Weinberg grants CDM a window into the philosophy of some of these leading technologists, and introduces us to this year’s winners.

The Winners

First Prize: MO, Interlude Consortium. Everyday objects become novel gestural interfaces.

From the project site:

The MO tangible interfaces are a series modules to capture various gestures, from motion to touch. The central module MO contains motion sensors (3D accelerometers and 3axis gyroscopes) and transmits the data wirelessly. Moreover, two accesorries, i.e. other sensors can be added to both side of MO.

Second Prize: MindBox Media Slot Machine, Humatic Berlin. A vintage slot machine is transformed into a compositional interface.

Personnel:

Christian Graupner , Humatic
….media artist, director, composer

Roberto Zappalà
… performer, choreographer

Norbert Schnell, IRCAM — Centre Pompidou
… interactive music & sound design

Nils Peters, Humatic
…system developer and software artist.

Third Prize: Samchillian Tip Tip Tip Cheeepeeeee, Leon Gruenbaum. It began as an ergonomic computer keyboard, but years of layered work on relative pitch makes it an instrument – a bit like a macro keyboard for composition.

Honorable Mention: Hexenkessel, Jacob Sello. A conventional acoustic timpani is both projection surface and multi-touch input.

From the creator’s description on the video:

The Hexenkessel is a modded 22″ timpani using LLP multitouch technology for control of live-electronics & dmx-light. the realisation of the instrument involves a modified led-projector, webcam and IR-Lasers. the programming is done entirely using max/MSP/Jitter + CCV. The instrument-hack is non-destructive and costs less than 300$.The instrument is intended for the use in multimedial stage performances and innovative concepts of new music.

Pioneering Judges Offer Their Philosophies

A musical instrument design may seem like subjectivity atop more subjectivity, a meeting of the aesthetic of the object with personal musical expression. Judges were asked, therefore, to describe the philosophy they brought to the contest. The reason, explains organizer Weinberg: “To steer it away from general statements – this is the better instrument than this – to make it more personal, about the judge’s opinion and artistic manifesto and instrumental manifesto.”

Tom Oberheim, the man who created the first polyphonic synth product, responded:

The first thing that I look for in a new musical instrument is its musicality. This means where appropriate: does is sound good, is it playable, does it add to the music making language. Then I consider if the device has some sort of universality; in other words, can it be used by a variety of musicians from different backgrounds. Finally, I consider the ease with which the device can be learned.

Sergi Jorda, creator of the reacTable tangible interface:

The ultimate goal for any new instrument could arguably be the potential to create a new kind of music. In that sense, baroque music cannot be imagined without the advances of sixteenth- and seventeenth-century luthiers, rock could not exist without the electric guitar, and jazz or hip-hop, without the redefinitions of the saxophone and the turntable. Yet, this extremely ambitious objective is often beyond the reach of its creator (eighty years separate Adolphe Sax from Coleman Hawkins, and no less than thirty go by between Les Paul and Jimi Hendrix). Being a bit more pragmatic, as a performer, my goal when constructing the instruments I will play is clear. I need instruments that are enjoyable to play and that mutually enhance the experience when playing with other musicians. Thereby allowing me to create or co-create music that will surprise me as much as possible, that will keep revealing little hidden secrets at every new performance. Music not necessarily better, nor worse, than a piece that I could compose in a studio, but music, in essence, that could not have been created in any other possible way. As a ‘professional’ luthier, I need to take some additional considerations into account, but the overall goals do not change: my aim is to create instruments which people can enjoy playing; instruments that will be able to enrich and mature the performers’ experiences in any imaginable way; instruments that allow scope for the performer (particularly in the case of a non-expert user) to be proud of the music created. In order to survive in the extremely demanding instrumental ecosystem, any new instrument should clearly excel in something. It should either be able to do one thing that no other instrument could or, at least it should do it better (whatever this can be and whatever “better” may mean). My last advice would be that when envisaging new instruments one should not only concentrate on the instruments’ sonic capabilities, on their algorithmic power or on the amount of sensors used. One should also be especially careful about the instruments’ conceptual capabilities, and consider how new instruments impose or suggest new ways of thinking to the player, as well as new ways of establishing relationships, new ways of interacting, new ways of organizing time and textures; new ways of playing, in short.

Jason Freeman, a composer, technologist, and Assistant Professor at Georgia Tech:

For me, new musical instruments are significant for their potential to transform our experiences with music. They may enable us to create new acoustic or electronic sounds not previously possible. They may encourage us to think about musical content, structure, and hierarchy in unusual ways. They may suggest new methods of musical collaboration, performance, or education. And they may make musical creativity more accessible to everyone. I am interested in instrument makers who have thought deeply about their work from technical, musical, and design perspectives to create musical instruments that transcend novelty to suggest new paradigms for musical creativity.

A Chat with the Organizer

Now in its third year, the Guthman competition has become a coveted award. As a result, says organizer Weinberg, who is director of the hosting Georgia Tech Center for Music Technology, quality and quantity were up in entrants. And, he says, he feels that entrants have transcended some of the typical designs in the field.

“New interfaces for many [means], let’s think about an object that we didn’t use before, and some kind of gesture, stick on some sensors, make some music … But I think the winners of our competition were outside of this realm, really innovative, completely new approaches for playing music,” he says.

On the prize-winning MO tangible interface:

In a section of the performance, they took a ball – a soccer ball – and did some [musical] gestures with it, threw it … moved it … on the hands, on the floor. Each one of these gestures was recorded with the gesture recognition. And then they actually threw the ball to the audience. The audience members started to throw the ball back and forth. If you threw it in a particular way, it made a particular sound — and everything’s wireless, completely — if you threw it back and forth in a different way, it made a different sound. It was really fun; people threw the ball at each other, threw the ball back at the stage. And all made music that was pretty cool to listen to.

Basically, the instrument becomes an intelligent entity. It can sense similar but different gestures and create something smart and relevant musically.

On the slot machine:

The gesture is mostly visual — the intelligence here is of the human performer. He makes his own gestures, accompanied by sounds. And it allows you to manipulate and change [the sound] — and get some surprises, because it is a slot machine, after all.

You can play, explore it. He was able to very expressively pet and touch and click and manipulate the slot machine to create some very nice — not only musical outcomes, but visual outcomes. In some cases, this guy is lying in the sea and making gestures in the sea. Sometimes he’s hanging stuff on the walls, and making sounds with his mouth. Sometimes it’s basic stuff that you can manipulate in real time, with a pretty unique interface — it’s not a monome, it’s a slot machine. It surprises you.

On the Samchillian:

Some instruments – controllers – have this short or sometimes long learning curve, but once you get to a certain point, you know it, and that’s what it can do. And you cannot get better at it. I think the Samchillian is really an instrument with a learning curve that’s very long, and just like other acoustic instruments, violin, piano, there’s a wide range of [technique]. And this guy was really a virtuoso with this instrument. He was able to play chords, all kinds of arpeggiators.

What I liked about it is it’s an instrument more than a controller. There’s always more to learn about how to become better with it. And I think that’s valuable.

Notably, Weinberg has no illusions about the challenge of making new instruments. It’s no accident that the winners were typically the result of years of development and evolution. “I don’t think any of the great instruments were invented in months,” says Weinberg. “It’s a lot of iteration, a lot of building… only a few are good enough to stick.”

And perhaps the great electronic instrument, while getting nearer, hasn’t yet been created. Weinberg says one example of a new instrument design that doesn’t work particularly well is the legendary Theremin – it’s beautiful in the hands of only a couple of artists, but generally a design that stumps musicians and is hard to play.

Looking at the winners this year, though, there are ideas on which new work can be built, not just impressive one-off instruments but real research into handling pitch and gesture. That, at least, should present a bright future. But with the competition heating up, aspiring engineers may want to get started on those designs now.

Thoughts? Questions about the work? Let us know.

More on the MO tangible interfaces from the IRCAM-based Interlude:
MO Interfaces

That work isn’t yet available for download, but an “augmented score viewer” is.

  • http://ardour.org/ Paul Davis

    I like Sergi Jorda's philosopy, I like what Weinberg has to say quite a lot, and I think that the Samchillian is a great idea. But I'm still not seeing much evidence (maybe a bit) of Eno's dictum that "a good instrument has qualities that the body can learn, and the mind cannot". The Samchillian comes closest, perhaps, because of its reliance on muscle memory, but the disconnect between the sound that is played and anything other than the key that was hit seems a little weak from the video. Maybe if it has velocity/pressure sensing, it would satisfy the Eno rule (and quite wonderfully, I think). And indeed, maybe it does already.

  • Peter Kirn

    Wait a sec, Paul, are we accepting everything Eno says as law?

    I can easily accept the first half of the Eno idea, but I think the second half is up for debate.

  • http://marcodonnarumma.com Marco Donnarumma

    A very good insight on new instruments making, thanks Peter.

    It was good to learn about the criteria by which a winner is chosen, however I still believe that in our age it will be very difficult to encounter or create a new "electric guitar" (to quote Sergi).

    The present modalities of sound generation and synthesis, analysis of motion and gesture and related tools (hardware and software) are so heterogeneous and interrelated in today's technology that it's difficult to invent _the_ ultimate novel instrument.

    Drawing a purposely overstated paradox, if electric guitar would have come after the computer it would have possibly gained much less interest and it would have far less affected musical creation.

    IMHO today's variety in musical performance paradigms and the capability of multi-layered cross-modal interaction is exactly what can enrich and strongly characterise our present sonic culture. 

    Perhaps, in the perspective of a critical analysis of the quest for "new", authentic sounds, my latest ongoing research might be of interest:
    http://res.marcodonnarumma.com/projects/xth-sense

    http://marcodonnarumma.com/works/music-for-flesh-

    cheers,

    Marco

  • Jim Aikin

    The technology (sensors, OSC, and software synthesizers) to create new instruments is well established. The factors in instrument design that I would find persuasive are not mentioned by the judges. In fact, Tom Oberheim goes, implicitly, in the opposite direction when he mentions the ease with which an instrument can be learned. One important factor is, will the instrument respond to small, precise movements in reliable ways? Small, precise movements are NOT easily learned, and an instrument that does not lend itself to this degree of control is not worth a moment's thought. The second factor is, is the instrument being widely enough accepted within the musical culture that a literature and a pedagogy of performance techniques is being developed for it? That can't be judged for many years after the instrument's invention, but we can guess that a collection of kitchen utensils probably won't achieve that level of cultural penetration.

  • http://robotcowboy.com Dan Wilcox

    It can be argued that the instruments presented may not be the next violin or guitar and, as of yet, there is still no definitive digital instrument (nor will there/should there be?) but what I think is important is we are trying, everyone is constantly trying. NIME and this competition are showcases for new, novel, and awesome ideas, some working and some not, but I love to see how different people approach the same problem in their own way. I can see someone, at some point, combining the best bits of the successful ideas into THE instrument. We shall see. It's one thing to make something that works for you and quite another to make something for others.

    What I do know is open source software, hardware, and protocols (OSC) are making it happen!

  • http://robotcowboy.com Dan Wilcox

    Call me performative oriented, but after watching the Samchilliian clip again I have to say the next new instrument has to look badass and make the player look badass (or at least vice versa). The Samchillian is a bit lacking in the "make the player look badass" department as it seems like he's playing on a green pair of boxer shorts …

    … same goes for the eigenharp :P

  • Peter Kirn

    I don't know that even something being a competition means necessarily you can have ONE definitive instrument. Heck, the history of instruments – with the exception of the comparatively recent ubiquity of the piano and guitar – is wildly diverse. Even the keyboard and fretboard, the essential elements of those two instruments, have led to loads of variety.

  • http://braduro.com James

    …second half of that Eno dictum ", and the mind cannot." I don't really have a problem with the complete statement, if I substitute "mind" for cognition or intellect. Like riding a bike, deliberate 100% consciousness of what your doing fed back into your brain while in the act, would make you crash. But I'm of the position that the mind is basically highly refined body, while the body is simply less refined mind.

    The penultimate Not-An-Instrument for me would be anything that implements Mackie control. And I have a similar complaint with most of the grid based controllerism that you see now, or automapping: essentially anything where the engineering involves cognitive decisions, like choosing submenus, suspending one's intuition to understand that same fader to be a panner AND a level, or a pad to be a sequence tick AND a graphic equalizer.

    On the other hand, kitchen tools that trigger sampled sounds are a programming achievement (we see that a lot around here on CDM so it makes me wonder why it got 1st place), but deserve more to be called a sound collage, installation, game or toy, and to me do not imbue the vocabulary, expression, range, styling, versatility, and craft of an instrument. 

    Does an instrument need to be difficult to learn in proportion to its potential for music? The relation does hold up, even with less complicated technology. If I called myself an MC, my overhead might at most be a mic. But having a voice, which is arguably the most intuitive instrument, takes plenty of talent and hard work to cultivate. So far, I haven't found any shortcuts. Although if I'm wrong, I'd hate to be in the elitist's camp, that of difficult instruments simply following a trend of high art and whining relevance. With music being so readily consumable, perhaps we need to reevaluate why we're stuck to rules on how to enjoy it and how we're involved in it. 

    And yes, for a soloist, its got to look sexy. Before the electric guitar, there was the horn section. At least for glory, it's always been about proving oneself talented with the hands or tongue. Noodling around on a pair of boxer shorts will not help Samchillian's case, but I guess there needs to be room for the future Tubas, accordions, and oboe players in the world.

  • http://braduro.com James

    I just want to clarify that something not being an instrument in my view in no way undermines the level of accomplishment of the inventor, whether we call that person a composer, instrumentalist, installation artist, or engineer. Undermines my notion of it being an instrument? Sure. There is obviously a great deal of construction and invention here whether its called a piece of composition, a realization of some programming rules, or a chime in the wind.

  • http://music-interface.com mat

    First of all, deep respect to all creators of music tools mentioned here. As I know how hard it is to create something playable at all.

    But I wonder.

    Isn´t the first place just "I put a Wiimote into an object"? On first sight I thought they get all control information directly from the video, which I would find more impressive. Like: Here is a new object; that represents a bassdrum; the higher it is, the louder it becomes and so on… (maybe something like this also already exists?)

    However, what I wonder even more about is to call them "instruments". Aren´t they controller? To me an instrument makes sound by itself (drum, guitar). A controller routes information and might lead to new interactions which is an important part of making music (especially when these new interactions force/enable you to make other kind of music). But an controller always need that routing; that interpretation. Plus at least a sound to make. So for MO, the most part of their creative work isn´t shown in the video: it is choosing the sounds and routing the controls. The performance itself is bounded and restricted to that. (I am not sure if I can make my point clear here….difficult…)

    Anyway – it is great work.

    Maybe next year we got something based on gestures only. Without any hardwarecontroller. Like: I paint a geometric figure in the air and the system know what instrument I want to add. Next I wave rhythmic around to set triggers. And then I show the pitch with the highth of my hands. Then next instrument. …lol….just some spontanious thoughts. And yes: this would be "only" a controller too.

  • http://ardour.org/ Paul Davis

    @mat: i think its a bit more complicated the "controller" vs. "instrument" division you mention. consider playing the cello. what is the bow? what is the string? what is the resonant chamber formed by the body of the cello? what are you fingers doing when the change the effective length of the strings? what are they doing when they pluck the strings? in an acoustic instrument, more often than not, there's no clear division between the role of "controller" and "instrument". some might say that this is (still) a fundamental weakness of most (all?) electronic instruments.

  • http://ardour.org/ Paul Davis

    @peter: its not such that eno's always right, just he's right about this particular point, and quite a few others :)

  • http://www.jonathanleonard.com Jonathan Adams Leona

    Ppppppbbbbbbbbbbbb!!!!

  • http://robotcowboy.com Dan Wilcox

    bzzzzeeeppppp!

  • http://dinisnoise.org jag

    The more outlandish the approach taken to provide "musical input", more attention a project gets.

    There is hardly any innovation in the "sound synthesis" department. Same oscillators. Same filters. Same substractive synthesis or samples.

    My free software musical instrument called din (download for gnu-linux at http://dinisnoise.org/download) (covered at CDM last month thanks ;) uses the humble computer mouse as a bow and can produce surprisingly good results because the mouse is a high sampling input device with very low latency and extraordinary precision. Just ask your FPS gamers. But you probably are getting ready to boo me off now :D

    If you like to hear what the mouse can do let me hazard a sound sample:

    http://din.googlecode.com/svn/audio/ushrinivas.mp… (mixing mandolin player u-shrinivas with live din playing)

    cheers

    –^–jag

  • http://marcodonnarumma.com Marco Donnarumma

    +1 for the instrument vs controller.

    An instrument produces, generates sound.

    A controller modify parameters of a sound which is generated elsewhere.

    Here the concept is subtle, but a device which playback loops and allow one to modify parameters according to sensor data is _not_ an instrument to me, but definitely a controller.

    The birth of a similar paradigm can be dated back to the origin of music reproduction. Is actual music what we can playback on a cd player (does anyone have one by the way?) or only what we can listen to live?

    I find nearly impossible to answer such question, but IMHO a truly novel and excellent "instrument" has to actually resound and not playback. Otherwise we have incredibly good controllers.

    Personally, I keep myself as closer as possible to the physicality of sound: kinetic energy, vibrations of material, or to quote Francisco Lopez, sound matter.

  • http://www.bloomingtonelectronic.com Mark Kunoff

    @ Dan Wilcox – that was the prototype. here's the what the finished version would look like: http://www.youtube.com/watch?v=DbOIBIwg_E4

    Hope he brings it to market!

  • http://robotcowboy.com Dan Wilcox

    @ Mark Kunoff – That video is from 2005 … where is the product?

    Personally, I'd rather see a wireless device more like Michael Waisvisz's The Hands:&nbsp ;http://vimeo.com/1203648. And that project is form the mid 80s! I had the privilege to see one of his performances before his untimely passing and it was far more expressive then the noodling in the Samchillian video.

    Moving your two hands in space as the instruments themselves seems far more useful to me then bending an ab roller with keyboard keys up and down.

    Too bad McRorie hasn't applied to the Georgia Tech contest:&nbsp ;http://www.youtube.com/watch?v=_oTgI_rzj2A

  • http://robotcowboy.com Dan Wilcox

    Oh, and to open another can of worms: is the Reactable an instrument? I have heard it called so, but I consider it a composition tool / controller and not an instrument.

  • http://braduro.com James

    I really didn't factor in the controller vs instrument argument when I made my earlier comments. To me, this misses the mark.

    Lots of controllers are modeled after either a musical keyboard, or bank of bank of channel, both of which would have carried an acoustic or analog signal at some point in its development. There're both interfaces to manipulating sound.

    In both cases, there is a tactile component to playing the interface (which puts all this touch pad commotion into question). But ultimately I don't care where the sound comes from only in that there's a direct relation to my physical decisions and the sound.

    You could just as easily argue that an instrument is an invention that has enough natural engineering principles in its design such that it lends itself to being a controller.

  • http://music-interface.com/ mat

    @ James

    my main focus wasn´t at the point that an instrument makes music by itself while a controller doesn´t (although this is the fact where you clearly can discriminate them).

    It is that a controller needs further creative work to get a creative result (routing, sound choice). And that kind of uncoupled work changes your workflow compared to an instrument. You first have to set these before you "play". This gives you much freedom in what your controller do but it also block somehow your direct experience.

    But don´t get me wrong: Controllers are good. And it is a complete new experience within the last decade. Before the digital revolution in electronic music there were only Instruments (Synths are instruments cause they are hard coupled) but now – for the first time in history – the control and the sound are decoupled which is the base for many new experience. However, the way changed, it is more indirect…

    I once tried to write an essay about this topic, maybe that could make my point of view a bit clearer – http://www.tonvibration.de/extra/Essay_cdm.pdf (As you see I am really fascinated about this change ;)

  • http://braduro.com James

    Thanks mat, I'll take a look at the paper.