nime: new interfaces for musical expression
a course at my master's program, itp. visit the course website.

aya karpinska
e-mail: ank212 at nyu dot edu

other journals: one || two || three(you're here)|| four || five || six || seven || eight || nine || ten

|| journal three|| 03. III. 03 || tentative communication.

last week i decided that only the two side flaps of the wooden box will move. so this week i decided to get data from two flex sensors into Max/MSP via the BX-24. i had note data coming out of one flex sensor last week, but changed my mind this past weekend AGAIN about what type of data i should be sending Max. i have now settled on control data, because it simply makes more sense to me: no worrying about constantly having to send a "noteoff" for every "noteon", and it just makes more sense to me conceptually to send a continuous value rather than discrete note values. check out my BX-24 code: MIDIout.txt i have the flex sensors set up to send continuous control values of 30-90. in reality, i only get a muddy range of about 36-53. below you will see a screengrab of a simple patch i used to check the values coming from my flex sensors.

i plugged this MIDItest patch into the final patch of a piece i did last year, one_two. the original piece used live voice, for this week's test i switched the input to come from a bunch of sound samples i created using csound. one of the flex sensors controls which sound sample is being played, the other controls which effects patch is being run on the sound. right now the data coming from the two flex sensors is unreliable, which makes "performing" pretty boring. tweaking two flex sensors and occaisonally getting "cool" sounds is not satisfying. however, i am able to hear what sounds good and what sounds bad - i may not be able to control it, but with careful tuning i can find the data that leads to a musically interesting experience. i guess i remembered a lot more of Max/MSP than i thought, good!

so, what to do next? back to the three basic component of my project:

• creating the banks of sounds using CSound: what sounds will make for good performance material? what frequencies? how rich should the sounds be? should they stand on their own or come to life only when put through effects processing?
• building the physical interface: how the heck do i build the box? i don't want the hinges to show. maybe discuss with gideon about the placement of hte flex sensors.
• programming a good Max/MSP patch for manipulating the sounds: explore what is out there in terms of effects patches. do much much experimentation, find a set of effects i can work well with