This page contains a journal documenting discoveries, observations, and challenges faced while doing Lab Assignments for the class of Physical computing.
Lab 1 - Fun with LEDs
06/30/05
What I've learned today is that soldering looks much easier than it is. Especially soldering two tiny wires to the tiny pins on the tiny header. On my first try, I accidentally soldered the pins on the header together. On the second try, I melted the plastic on the header and it got completely deformed. I had to force it back to shape with a pair of pliers. That was probably the hardest part of LAB 1. The rest was pretty basic, the wiring was illustrated well in the pictures, and the "Physical Computing" book helps a great deal. Making the little LED's light up made my day.
The Setup.
Switch off.
Switch on.
I just got a 10k ohm POT at the store and wired that into the board (again, hopefully I'll have pictures). I set up 2 groups of LEDs, which would alternate in brightness on the turn of the knob. For some reason the POT seemed to favor one group, is it would get really bright when the knob is all the way to the left, but the other group would be barely bright when the knob is all the way to the right. I will have to figure out why.
Pot -- dimmer switch.
Lab 2 - Fun with PICs
07/07/05
I got to program the chip for the first time today, and finally the breadboard is doing something fun. So far it's not interactive, but in the future it will be. All it does now, is light 7 LED's in sequence. The most difficult part of this Lab was the actual physical positioning of the LEDs and resistors. Since it's such a small space, I had a lot of trouble getting the lights to line up without having overlapping resistors and LEDs and causing possible short circuits. The solution was in that I used many short wires to spread things out.
The setup.
Action shot.
Lab 3 - Fun with Servos
07/13/05
I hate soldering! With a passion! The more I do it, and worse I get at it.
Besides that, the servo was a lot of fun. I did a pretty simple setup, where the Servo's arm is controlled by a potentiometer and it carries a wire connected to the power. There are 4 LEDs spread out on the breadboard with the metal plates attached to them, and as the Servo arm touches them, they're supposed to line up. It would work, if I would've soldered it correctly. Apparently only 1 out of 4 metal plates is soldered to the LEDs, so only that one goes on. I will improve on it by soldering it better and perhaps incorporating some sturdy materials, such as wood, to have everything properly lined up.
The setup.
Close-up of poorly soldered contact plates.
Midterm Idea
07/18/15
I've been trying to come up with something interesting to do for my midterm, but due to time and ability contstraints, I could only come up with a simple game, where the breadboard and the chip setup will be the controller. There will be targets appearing on the screne in random places and you have to chase after them. The tricky part is that you could only move vertically or horizontally at a time and you could switch the direction with a switch. It is sort of like those amusement park games, where you have a claw to grab prizes out of a bin, but your claw could only either move up or down or left to right, but not anywhere in between, say, diagonally. Hopefully I'll implement some sort of inertia into it, to make it a little bit challenging, but as of now, I can't even get it to show up on the screen.
Midterm Progress
07/19/15
Ok, so the project didn't work correctly for 2 reasons:
1) My chip was not popped all the way in
2) I was sending values to processing with my chip using DEC conversion, as opposed to raw data.
After I worked these two issues out, it was just a matter of setting it up. So I got my programming done and all I need is to make the interface for appealing. I want to enclose my circuit board in the box and get switches that are easier to turn and flip. Anywho, the game is that you're a mouse and you need to snatch as many pieces of cheese in 3 minutes as you can. Thanks to June for the mouse vs. cheese idea. Here's a screen shot
07/21/05
Ok, submitted my midterm. It went well. The box took me forever to make, because I got the measurements all wrong. With a lot of sanding and cutting it finally started looking like a box. For the final, I will improve more on the game, making it more challenging and maybe putting in a couple of unexpected surprises. Should be fun.
Final progress
08/02/05
Final progress is not going so hot. I needed a multiturn potentiometer to make my project work properly, so for my first attempt I took a regular pot, shave off the part that prevents it from turning all the way. That created a dead zone which was giving random values, not outside of the regular range of the pot and thus causing my mouse to skip all over the screen. After spending 4 days on trying to figure out how to fix the issue (in code, and with capacitors) I gave up and bought a 10-turn pot. I don't know if that pot is defective, or if they're all like that, but it wobbles, causing my mouse to shake. I spent another 3 days trying to get rid of a wobble and whatever I attempted only made it worse. As I've tried averaging up the values, creating threshholes, etc. instead of shaking the mouse would start sliding on the screen, which is unacceptable. I will keep trying to get rid of the wobble, but something tells me I won't be able to.
On the positive side, I put a mousetrap on the screen, which can kill the mouse dead and a countdown timer. Here's a new screeshot:
08/04/05
Some last minute fixing up: The wobble is still there. For some odd reason it rarely occurs when I run the program at work. When I run it at home, it goes complete crazy, but then I am using a different adapter there (even thought it's still 12v AC to DC). Another problem I face is collision detection. I thought that an image's coordinates were defined by its center, but in reality they're defined by the upper left corner. Another difficulty is the fact that the images appear to be irregular shaped, but they're defined in a rectangle. I tried my best to make the collision between the images more or less believable. Also, I added a tiled floor as the background. Here's a screenshot:
