"When you make a thing, a thing that is new, it is so complicated making it that it is bound to be ugly. But those that make it after you, they don’t have to worry about making it. And they can make it pretty, and so everybody can like it when others make it after you"
Skelo’s movies take you to another world where gangsters hack old video game controllers with MIDIsense kits, to make music. Or at least, that’s what I could understand from this video he just posted in the ladyada.net forums
I fixed some icky problems (in wxMIDI!) and now the MIDIsense software is crosscompiling nicely between Mac and Windows, therefore there is now fast and updated software for both platforms. Rad! Download it now!
Now that i’ve finished porting icky C++ code I can return back to thoughts of sweet sweet hardware design. Next up is a second MIDIsense board.
The next board is pretty simple: An analog and digital IO board. This is what most MIDI interface boards are used for. The atmega8 chip has 6 10-bit analog inputs so there will definately be 6 analog inputs. Then I decided to put in 5 digital I/O too. Not only will you be able to get digital inputs (switches) but also control the pins as outputs including PWM outputs. OK not very interesting, I admit.
So then I thought about improving that a little bit:
For the 5 digital I/Os, instead of just having the pin go to the microcontroller and thats it, I put in spot for installing a capacitor to +5V and a series resistor. (To install the resistor a trace must be cut).
Of course, the RC pair can be used as an output filter: so that PWM output can be smoothed out nicely (the output will be PWM’d at around 16Khz)
The resistor can also be used to choke the output, for driving an LED for example.
But of course the real trick is that now the board has 5 extra analog inputs for measuring resistive sensors: by setting the output of the pin to +5V (discharging the capacitor, its tied to 5V) and then measuring how long it takes for the capacitor to discharge through the resistor! Its a poor method of measurement and takes a while to get a stable result (as compared to the onboard A/D). However, it certainly works and for many situations, and its good enough for most MIDI sensor applications where the 10-bit A/D may be overkill.
(You can read more about the RC timing A/D trick from this old-school app note from phillips although its rather outdated, you can get an idea of how to implement it!)
So now the board has: 11 Analog inputs, or 11 digital/PWM outputs (5 with RC filters).
If you wanted to try out the MIDIsense hardware but didn’t have a Mac, well I finally finished porting the wxpython code to C++ and its all much faster and more reliable. I also improved the interface and robustness. Try it out and let me know how it goes, available for download from sourceforge
I finally got off my butt and finished documentation for MIDIsense, a simple and inexpensive MIDI/sensor system for artists, musicians and experimenters. I did a workshop with these in March and they worked great so I’m happy that they’re finally available.
The only board I’ve released so far is for log resistive sensors. These are pretty common: photocells, bend/flex sensor, force sensors. My example right now is a laser harp using $3 laser pointers and $0.50 photocells. I’ll probably do an example with a bend-sensor glove or tapping a force sensor next. I’m also, of course, hoping people decide to buy the kits and come up with neat new interfaces.
I’ll release the Analog/Digital I/O board next, which will be much simpler, in a sense…but will allow 5 buttons/switches and 6 analog inputs, such as distance sensors and linear potentiometers.
More importantly, I need to hack on the windows python code because the windows MIDI subsystem seems quite slow in comparison to even a 3 year old iMac!
If you have the opportunity to see their current show on tour, [R]Evolutions, I highly suggest it: sort of like an abstract ‘qatsi but less Glass and more minimal/industrial.