In an attempt to appease my time management and your sanity, I’ll endeavour to make this fast…
This week I began work proper on what will possibly become the ‘core’ of the interactive devices I create this year. After getting some really helpful information about safe powering methods, I’m now testing out power over ethernet (PoE) as a promising option. I picked up the Freetronics ethernet shield and midspan injector for this purpose, as well as the nifty Wiichuck adapter, to quickly pull accelerometer information out of the Nintendo Wii nunchuck.
Aside from some missing jumper shunts, which I had to make myself, this was all very straightforward. There is plenty of Arduino and Processing code to be found at both the Freetronics and Wiichuck adapter sites, so I was up and running quickly. What has got me stumped is how to get meaningful data into Processing via the ethernet/network feed.
Unlike sending data via USB, when Processing reads network data, there seems to be regular loss of information. My knowledge of networking is limited (non-existant), so exactly why this happens is beyond me. What that means is I have to come up with a way of Processing not being concerned if the occasional integer doesn’t arrive (at the moment, I’m getting ‘out of range’ errors in my arrays). One option I’m yet to explore is using the UDP library, instead of the network one.
To test this, I needed a Gentleforce video game (of course). The plan is to run this on the television, simply as a proof of concept that I can capture data untethered to a computer…
…it may be silly, but things were getting a little serious around here for my liking. Based on Andy Best’s OpenCV tutorials, it’s a basic bubble-popping game. Instead of mouse movement, the Wii nunchuck will take care of the crab crusading around the screen.
Finally, the first of my two ethics applications was sent off this week. Fingers crossed that it goes smoothly and in a few weeks time, I can start looking for participants for this project.