This was my first Ludum Dare – I went into it feeling completely unprepared, and finished the compo with a game I’m pretty happy with. I’m going to attempt to describe how that happened.

What Went Right

  • The Toolchain. I made a rather risky move going into this LD. I decided to use a game engine I’d never tried (Flixel) in a language I haven’t used in years (ActionScript 3) in an IDE that doesn’t run on my OS (FlashDevelop) Crazy, right? Actually, I don’t think it could have gone better. Flixel’s handling of basic motion and collisions is far more intuitive than any physics engine I’ve tried, and the engine as a whole seems perfectly tailored to quickly prototyping games. Due to some of the awesome features in Flixel, I ended up using Photoshop as my map editor, which worked quite well, and is something I’ll be thinking about doing more of in the future.
  • The code. The one piece of advice I got before Ludum Dare was to not get caught up in the code. That’s a problem I’ve had in the past (I’ve got a number of game prototypes that are more engine than game, nearly all of the utilities I’ve released have about 3x the functionality that can be shown in their UIs) but with the looming 48 hour deadline, I was able to ignore best practices and just get sh*t done. There’s dead code, unused variables, methods copied from class to class, and not a comment in the entire program. But it works, and ultimately that’s all that matters.
  • Music & Sounds. The audio ended up being much less of a hassle than I’d initially thought it would be. For the music, I used Garage Band on iPad which has a feature called “Smart Instruments” – I gave it a couple chords, and it gave me back a groovy baseline. I spent about as much time making the music as I did trying to get the .mp3 to loop properly. For most of the sound effects I just decided to record them myself. I grabbed my iPhone, went in to the quietest room of the house and made zombie noises for a few minutes. It was fun, actually.

What Went Wrong

  • The Theme. At least initially, I had no idea what to do with this theme. I didn’t want to wind up with the same idea as everyone else, so the obvious ideas of a zeldaish game or something where you only win if you’re holding the MacGuffin were out. I spent probably 4 hours tearing my hair out and contemplating leaving the competition before scrawling down “you start in a portal like chamber where you don’t get the macguffin, then you fight zombies or something”
  • The Graphics. Unlike the code, where I only did what I needed, as I needed it, when it came to the graphics – I didn’t really have a plan. I started the graphics before starting the code, which was a mistake – but the bigger mistake was starting the graphics before I’d nailed down what the game was going to be. I wasted a lot of time on graphics I didn’t end up needing, or which just didn’t look right (all the walk cycles) I’m relatively happy with where the graphics ended up, but it took too long to get there.
  • Planning. I didn’t really know what my game was going to be until after I was 3/4 of the way finished, so I didn’t even have a todo list until the last 12 hours. I’m pretty good at flying by the seat of my pants, but I think a little pre-planning would have helped. You know, trying to set a couple milestones, researching the engine I was using a little more, that kind of thing.

And if you read all that, here’s your reward! I’ve uploaded a Timelapse of the development of Super Shotgun Deathrace which you probably shouldn’t watch until after you’ve played the game


Leave a Reply

You must be logged in to post a comment.

[cache: storing page]