This is a really nice passive multitouch input by Randall Jones. It was designed to be an inexpensive, and extremely expressive, musical interface. I really like the simplicity, and it only cost about $50 to build. [Thanks Dan]
Physical modeling synthesis has proven to be a successful method of synthesizing realistic sounds, but providing expressive controls for performance remains a major challenge. This thesis presents a new approach to playing physical models, based on multidimensional signals. Its focus is on the long-term research question, “How can we make a computer-mediated instrument with control intimacy equal to the most expressive acoustic instruments?”
More about the Multitouch Prototype 2
12 thoughts on “Intimate control for physical modeling synthesis”
I’m glad you like the work! I hope to hear from some other folks who try this.
One thing I want to mention though, lest anyone be disappointed: an 8×8 audio interface is needed to make the thing run, and those run a lot more than $50.
Also, I have only tried the RME Fireface so far. If I were trying another interface, I would test a 1×1 setup first (one taxel) to see if it worked, or add a buffer amplifier to match the antenna impedance to the line input.
Hey, this is amazing.
I’m looking to try something like this. I have a 4×4 interface, would that still work if I reduced inputs and bars? If so, what would be the effect on the tonal range of the instrument? I’d love to hear more about this project from you. my email is firstname.lastname@example.org.
You are right Randy, thats some expensive equipment there. I got my hopes up. Great job anyhow mate!
I’m thinking it should be possible to power the thing via a microcontroller (e.g. arduino). But the sampling speed might suffer. :S
I don’t think the sampling rate would be so limited by the microcontroller as the communication interface. If you do 64-point 7-bits per point at 115200 baud, you get about 120 Hz sample rate. If you throw on an ethernet shield or something, or maybe do something crazy like serial over audio, you might be able to get a fourth or eighth the response time he’s getting.
So if i have a 8×8 audio interface that i can use for this would that mean that it would have to be dedicated and i would need another one for sound from the computer?
for example: i have a motu 828 which im assuming would work, so then would i have to use another one (i also have a NI audio kontrol) for my actual sound?
Comments are closed.