October Game: Gameplay video

I had to do weird things with the emulator to capture this video.

October game: wait which month is it again?

Yeah I’m still working on this, apparently. No actual November game in sight. Kinda a bit low on days left in the month.

I started out refactoring the whole cheats system so that it’s a lot less hardcoded int arrays and a lot more object oriented and enumerable at runtime.
And then I added the stuff you can see in the screenshot:

1. A health display for bosses (Well, boss, there’s only one boss fight so far).
This was pretty easy to hack in, but I ended up refactoring the health-display so it was a bit less hacky. The hardest part is the scaling, since the one boss I’ve implemented has
too much health to fit on screen. At least I figured out a shiny way to autoscale and I don’t have to manually hardcode scaling for various boss healths.

2. A pixely font and border for the in-game text (pause, you win, you lose bits).

And now I have to go find an older version to screenshot to show how it changed, since I now see that didn’t post a pic of it (probably because it was so ugly). It took forever too, since I didn’t really understand how Android does its magical border-images. I’m still not happy with it, but unless I write my own font engine I think I’m stuck with anti-aliased fonts. WHY DO YOU HATE MY PIXELS, GOOGS?

Project: Zelda Door

Someone on tumblr said that they love this theme and that it should play whenever they enter their house.

Since it’s Halloween and I had a microcontroller handy, I made it happen.

Technical type details:

The Arduino is running a minor variant of the Debounce tutorial, with serial status writing added. On the laptop there’s a simple python script that waits for the serial status to show that the door has been open for 1/5th of a second, and then it runs an external command. I set the external command to “playsound zelda_oot_house.mp3”, so it plays the House theme (and not the Massive Attack one).

The biggest design problem I had was that the frame of the door is metal, so there’s always connectivity between the two halves of the security lock, even when the door is open. I had to rewrite it halfway through to detect the change in resistance (since the door open is about 600ohms and closed is about 1). It would have been nice to know that ahead of time, I really should have checked with my multimeter before wiring it up.