Currently in Pittsburgh, PA for NIME 2009, a packed 3-day conference featuring the cutting edge of musical interface design (pdf program). This afternoon featured some neat robotics with Shimon, the mechanical marimba player, and writing instruments for sound control from the folk at MIT's Responsive Environments group.

En route here, we stopped off for a few days in New York. This gave us the opportunity to visit the great people at Eyebeam, whose hospitality was unrivalled (though it did help to inadvertently wander in during their weekly open-studio hours). We also caught the superb Tangled Alphabet exhibitions at MOMA, followed by Sophie Calle's stunning "Take Care Of Yourself" at the Paula Cooper Gallery. One nine-hour drive later, the sheer verticality of New York has given way to Pennsylvania's expansive forests and rolling hills. Driving through Squirrel Hill Tunnel on Interstate 376 was a highlight; $25 tolls, less so.

Now on CMU's wireless so will again be checking email sporadically.

Recently demoed at the 2009 Consumer Electronics Show in Las Vegas was a control device that could rival the Wiimote for gestural input: Sixense's TrueMotion, which uses a wireless handset coupled with a (presumably USB) base station to determine motion and position information. The big leap here is the latter: the Wiimote's accelerometer can only give relative motion data, and absolute 3D positioning is nigh on impossible to derive from this without some inaccuracy and drift. Some positional info could be gleaned using the Wiimote's IR LEDs, but this was still less than ideal and had the inherent limitations of a line-of-sight system.

The TrueMotion device functions by locating the handset within a magnetic field, which appears to give snappy and precise location data. It reminds me of the Ascension Flock of Birds sensor, which I briefly played with a couple of years back, albeit without the clunky serial I/O and hefty pricetag: the TrueMotion is estimated at $100 for a base station and handset. There's an interesting interview with one of the Sixense chaps over at Engineering TV.

I'm looking forward to get more info on this as it sounds like a potentially paradigm-changing controller for audio and video. Transmission range? Multiple devices/base stations? Proprietary drivers? We shall see, as it is due to hit the market later this year.

In support of the p5_sc library, which provides the ability to communicate with SuperCollider's synthesis server from Processing, I've developed and amassed a series of basic helper functions ported from SC's client library. These provide frequently-used operations when interfacing between control and audio-generation objects, including:

• converting MIDI note values to cps
• random number generation with nonuniform distribution
• value clipping
• mapping between value ranges (linear and exponential)
• support for Spec objects, including a global named Spec register

It's undocumented, besides code comments, but hopefully useful.
Download sc_utils here.