Mills College Signal Flow 2015: Days 2 and 3

Friday and Saturday went very well at Signal Flow. I spend the day on Friday going through my code to see if I could get things running more consistently and managed to figure out a workaround. The Timer library for Arduino ended up being the answer to my woes, and I was able to set up a signal to be sent to Processing at set intervals in order to keep the videos active when the board wasn’t touched awhile.

I also made some fliers to get people in the door, since the installation is so far removed from the music building and other Signal Flow festivities. L and I put them up all over campus and they definitely did the trick.




Lots of people came through and stayed awhile to stare at the weather balloon. I switched up the videos a lot yesterday to make sure it was always pretty fresh. My friend came to document everything and filmed for a good hour or so, so I’ll have plenty of good material to work with. People were really interested in the balloon and the projections, and I had SuperCollider throwing out my binary sequences as well as a semi-randomized pattern for my retrigger signal from Arduino. I pointed the speakers at the walls on either side of the room and it spatialized nicely. Most of the frequencies I used were harmonically related so they resonated nicely with each other in the room and it became very immersive. It was interesting talking with people about the history of the Ouija board and Spiritualism, and about early Electronic music and the Occult (some interesting connections there). A few groups used the Ouija board to ask some questions and had fun with the answers. One lady asked it when her granddaughter would be born, which I thought was sweet.

The new code worked so well that I was even able to leave everything unattended for a couple of hours to go watch everyone’s performances! That was an exciting change, as I had mostly been babysitting the installation in case things froze up. The tricky thing is that the capacitance threshold seems to be well calibrated to my body but not all bodies, so some people had a harder time triggering the letters. The fact that each letter has a different threshold is also not entirely helpful. I am going to spend some time after this weekend testing each individual point and calibrating everything individually, as well as setting an integer range as an output so I can use the board as a MIDI controller with Max MSP and Ableton.