A Game A Week: Dungeon Runner

The first ever game-of-the-week™ is a game about running. And jumping. You do that too. It's called Dungeon Runner.

Dungeon Runner - Total Development Time 12 days


What it is: A game about jumping over gaps. Run all the way to the end of the dungeon to win.


Where you can play it: Here.

What worked: The jumping physics feel good. I used a rigidbody (a simulation of how physics affects rigid objects) and added an upward force whenever the user hits the 'jump' button, then tweaked the values until it all felt right. I'm also quite proud of the art style (despite the repetitive textures) - I drew all the assets with a ballpoint pen then used a piece of software called Sprite Lamp to add dynamic lighting. The sketch-like shadows that I ended up with look really cool.

What didn't work: The wall textures and floor textures look too similar (which they should - they are the same texture)! My only play-tester (who is totally not my girlfriend) told me playing the game made her dizzy because of the repetitive textures flashing by. Also - and this isn't a gameplay problem but a development pipeline one - making all the art assets by hand took ages, especially because Sprite Lamp requires each asset (each frame of the animation!) to be drawn six times (diffuse map plus five copies of the image lit from different angles). Like I said above, I'm really pleased with the results, but for my subsequent weekly projects I'm going to be relying a lot more on third-party assets and/or asset generation software to keep development time within the one-week-ish time frame. Also, Unity's 2D physics are a little finicky, and although I ironed out most of the bugs there seem to still be times when the physics don't behave as expected.

What I learned: Loads! I learned that the scope of these projects has to be way smaller than I had anticipated. I had originally planned to include enemies that got in the runner's way, but creating the assets for that would have taken way too long. Basically, to keep development time on track I need to cut out everything except the key mechanic(s). From a technical standpoint, I learned how to export Unity Web Player builds with the settings that gave me the product I want, and how to draw a good tiling image. I learned how to reset a level to its initial state. I also learned how to use Unity's UI interface, and how to change UI elements in code.

Development Breakdown:

• Idea and prototype: 1 day
• Coding gameplay mechanics:  3 days
• Creating art assets: 4 days
• Debug and polishing: 2 days
• Build, publishing, and write-up: 2 days