Tile dodger is a simple game that can get infinitely complex. Collect as many blue cubes as possible while avoiding the red bars coming from the walls. 

This was done independently for a one week game jam in 2014.

 
The most difficult problem I faced with Tile Dodger was definitely getting the tiles to move along the walls. In attempt to be a good programmer I tried making simple calculations and algorithms to get around. The problem with all of them was that they all had some kind of outlier that would break them. I don’t know about you but I’ve never heard of a simple equation to move around the surface of a cube. At one point I decided to focus on the results over exactly how it was done and try out some ‘messy’ solutions.

The one I settled with managed to not only work amazingly but was arguably simpler than my previous attempts. What I did was record all possible motions in a 4×3 matrix. Each of the 3 main arrays represented paths around the each axis while each of the 4 arrays within them were directions for each wall that make up those paths. All I had to do was do a check to see if I was going pass a wall, if I was then I’d update the index of the current wall and I would be given the appropriate direction to move along it. From this I learned that I shouldn’t waste time trying to find a specific way of coding something if it risks the result itself.