Ludum Dare ended five days ago and now it’s in the voting phase, where everyone plays everyone else’s games and votes on them. There’s 1406 of them, though, and I’m being incredibly remiss and have only rated 10 or so randomly selected games. I’ve been lazy, and I’d make excuses for that laziness but I’m too lazy to bother. It’s a certain very special kind of efficiency I’ve perfected.
Anyway, the consensus on my project, as far as I’ve been able to gather from comments, mirrors my own analysis of it fairly closely: It’s an impressive bit of programming for 48 hours and an intriguing screen saver, but there’s not a lot of game there.
It was an interesting project to work on for several reasons, but it struck me as unusual in this way: If I were to pursue it completely and wholeheartedly I have no idea where it would take me. That is, usually when I have an idea for a game I have a very defined end point, either a story I want to tell or an experience I want to convey. In this case, I want to simulate the impetuses of nature which cause a species to evolve over time in an interesting way: Then, build a game on top of it. This approach has two– I want to say limitations, but that’s inaccurate. The opposite of limitations– two breaches in the hull of the ship that defines the design space, where the design can go off to infinity.
The first such breach is the simulation. Hypothetically I could keep on designing and designing the statistical model until it perfectly mirrors everything we know about evolution. I don’t think that’s necessarily super interesting, because then I may simply have made a system which echoes back the facts I fed into it, but it also might be interesting in that it would produce conceptually viable species which didn’t happen to evolve in our world. This simulation would, of course, be an insane project for one person to undertake, and is inconceivable in depth and scope. Probably not a great idea for a goal, but a worthwhile concept to keep in mind just to compare other design possibilities with.
Alternately, and this would, I think, be the approach I’d take, I could use physical reality as a reference point for certain forms of evolution I’d like to see (such as camouflage and poison) and then build a system to enable those while, at the same time, including things into the stat system that wouldn’t exist or be included in a realistic model of biology. It would be interesting to work this into an evolutionary model that includes statistics such as faith and luck, which independently affect each other. Such a simulation could certainly provide a surface with sufficient tooth to sketch a compelling fantasy world onto.
In any case, there’s no end to the design, at least no end one could practically expect to encounter. One could keep on adding stats and relationships between stats ad infinitum. This is actually a problem I ran into during Ludum Dare, since the original design called for a kind of recursive stat system, where the creature’s age, food level, water level, etc would modify its stats and those multipliers would themselves be mutated with each generation. Of course, that would then open the door for the multipliers multiplying each other and so on, and it would be a nightmarishly difficult system to implement and debug in the 48 hours allowed, so I trashed that idea.
Other ideas which were cut or discarded because of time constraints include: carnivores, scavengers, food decay system and food poisoning, regrowth of vegetation, perception of and preference for different terrain types, day/night cycles, seasonal cycles, weather, fall damage, terrain modification with water re-leveling, mountain streams, sight-lines, different senses (hearing, smell), different types of vegetation… and so forth; and so forth..
The second breach in the design was that one could build nearly any conceivable gameplay on top of this stuff. The fact is, I didn’t come up with an idea for a game, I came up with an idea for a simulation– which, while it is cool and compelling and a fun project to work on, kind of leaves me up a shit creek on a shit island in a shit ocean when it comes to constructing a relatively complete game project. Which is kind of… crappy.
I have less to say about this breach since I mostly just stopped thinking about it to concentrate on the project, but I did have a few idea for how to game-ify it. Really, the question boils down to what the player’s role is in the world.
Does he control one of the species? It could be an interesting strategy game to try to maintain survivability for your species by carefully protecting the breeding pool while at the same time trying to ensure that the traits that are most desirable get bred for.
Perhaps the player should be a third party? This suggests the ‘hunter/caretaker’ design which I proposed, where the player controls a hunter who must consume the animals to survive and grow strong and needs to keep a stable population of diverse and healthy creatures towards this end.
Maybe he should control an abstract and distant observer, which suggests a god game where he can warp terrain and deploy natural disasters and observe their effect on the population.
Or one could do something completely different, and simply use it as a system to generate behaviors and stats for the creatures in an RPG or something.
The possibilities are truly endless.
Anyway. It’s a powerful and interesting design, even if perhaps it’s not ready to be a game yet. And, honestly, I’m not sure if I’m ever going to work on it again, since if there’s one thing I think I have communicated here it’s that this design would take a lot of work to finish to my standards. Maybe I’ll just sit on it for a while and brood, and we shall see how it hatches.
We shall see.
Next time: What I learned