I came into this LD knowing I had to travel most of Sunday so the initial intention was to spend three hours on LD only. Ha, fat chance. I spent about 29 hours of the compo on my game, with maybe 16 hours of that being actual work. That’s still a lot more time than I budgeted.
Unfortunately I don’t think a lot of it really shows. I started out with an idea for the most basic of shooter and was not really committed to making it that unique or innovative and instead spent most of my time making assets. From a game-building perspective this was educational because of the time management aspect (I’ve done one other Ludum Dare in the past and only did one piece of pixel sprite and some music) but games are about gameplay and so it probably suggests that I should be more careful about where I put my attention in the future. I think I might have spent around four or five hours on the engine, tops (and there’s nothing really groundbreaking or innovative in it or the core gameplay). I also realize when doing this that there’s a lot of boilerplate and that I should start putting together good sample code and snippets together for reuse, since good organization will save me a lot of time.
Now that this is my second Ludum Dare, and I think one success is that I managed to build a game pretty similar to my first entry (though it’s much, much simpler) in approximately half the time it took me the first time, which is pretty impressive in itself. My programming environment was better, my experience was better, my coping with the pygame API was better, and my overall knowledge was just better. (Unfortunately my geometry and trigonometry is still terrible, but that’s a fixable problem – just not fixable in a single weekend). But what Fred Brooks said in “The Mythical Man-Month” applies a lot to what I did. “The second system is the most dangerous” because it’s the one where you think you know what you’re doing but you haven’t quite figured it out yet, so you’re prone to making a lot of mistakes. In my case, it was building much too simple of a “core game”.
I think the summary is: if you start by telling yourself that you’re going to create “something small”, then you’re going to end up with something small and not too interesting, but you’ll still somehow managed to spend all your time on it anyway.
While it sounds like I’m panning my game, I’m glad I submitted something – something is better than nothing and three hours (or sixteen) is something. I just think that the next LD, I need to pre-plan a little more and get a better sense on how to make more creative and interesting games, or, at the least, not tell myself, “I don’t have time for this” – though I do need to remember that awesome ideas take time. I bet, given an additional extra twelve or so hours of work I’d be able to add a lot of interesting aspects to what I’ve done so far (but you can also see from the blog that many people get innovation down right at the start). I guess it also sounds like I’m not proud of what I did. I am, actually (I’m really happy I did artwork this time and I did some killer music). I think there’s a little bit of disappointment that, well, it wasn’t pushing the envelope more technically.
That reminds me, I did have an idea that I really wanted to do, but couldn’t do easily within the limitations of pygame’s mixer: I wanted to do a game with synchronized audio and gameplay, but the mixer wasn’t accurate enough for that (too much latency) and trying to work around that would have probably risked the entire project. I think it’s something to look at after my deadlines are over, though – I have a basic game engine here from which I can work from and build a rhythm-based game on, after all. And then I’ll rock LD28. Right? Right??
P.S. Play my game anyway. It’s been submitted!