Second day of prototyping

Additions
Because the light censor had worked surprisingly well we continued with that approach and with the hope of incorporate to different sounds depending on how far the hand would be from the censor (see illustration). We then talked about how we should make the prototype, materials and so on.

We all agreed that it should be a simple/sleek design with as little as possible of the electronics showing. We decided that coding the prototype was a one (wo)man job. David signed up for the job and worked with the code while the rest of us worked with the prototype. Fortunately we found a box matching just the way we wanted the prototype and the girls pretty soon made a functional prototype (still lacking the code though). Christoffer and I thought of the idea of changing instruments. With RFID tags we could realize that and we started decorate the tags so the tags reflected the different instrument (more or less, see picture below). We also started implementing this with our own InstrumentSelector class in java.

Sound Problems
David dropped working with sound (wave) in Java due to problems mentioned from the first day and some other unknown problems. He started using midi instead.

When we started testing the prototype it’s was hard to distinguish “high” and “low” movements from each other because of two reasons. First of all when removing the hand after a “low” movement the light censor value would increase until it reached a normal state and by doing so passing the “high” value thereby triggering both the “low” sound and the “high” sound. Second of all, due the different light sources, if you held the hand slightly to one sides of the prototype the amount of light being removed would correspond to the amount of light being removed if you held the hand in the “high” position thus triggering the “high” sound. We tried solving this in different ways but nothing helped (I later realised that the best way of solving this problem was to have a proximity censor so that the sound would only be played when this censor also was triggered). We ended up skipping the idea of two different sounds and just went for one sound.

The RFID tags worked as planned, but it wasn’t as easy to change instrument on run-time as predicted. David worked for this for quite some time but time was running late and the group had courses to follow or assignments to make.

Redesign
Christoffer and I had no plans so we decided to look at the code to see if there was some solution to the problem. I took a second look on sound in Java and Christoffer looked at the censor part. This ended in a complete redesign of the code. Christoffer managed to sort the events so a sound wouldn’t be played every time there was the slightest movement, but only when big changes in light occurred. I had sound working on my laptop and we seemed confident that the hours given Monday was enough to make it all play together.