August 24th-27th 2012 :: Theme: Evolution

Back to Browse Entries


by Problem Machine - 48 Hour Compo Entry

World with simple evolution modeling. Watch randomly generated species, observe which seem to be the most successful

* Click on/near a creature to view its stats
* Generate new species with the new species button, see which survive to breed
* Generate new terrain with the new terrain button if you get tired of the landscape.
Note: Terrain generation takes a few seconds, be patient

*TypeA, TypeB, TypeC: The creature's type represented by 3 numbers. These are converted to colors for display

*Health/Max: When health reaches 0, creature dies. Initially based on size.

*Food/Max: Goes down over time based on hunger, recovers based on digest when creature encounters food. If food goes below 0 health starts decreasing.

*Water/Max: Goes down over time, recovers when creature encounters water.

*Age/Max: Begins at 0, increases until it reaches max and creature dies.

*Drown/Max: Increases when creature encounters deep water, if it reaches max creature dies

*Decay/Max: Increases after death, when it reaches max the creature disappears completely

*Size: Base stat used to determine many other stats

*Speed: Maximum movement speed

*Aware: How often per second, on average, the creature checks its surroundings

*Sight: The maximum distance at which the creature is aware of other creatures

*Wander: How far the creature will wander away from its algorithmically determined destination

*Heal: Rate of health regeneration when well fed and watered

*Digest: Ability to digest grass

*Hunger: Rate at which food is consumed

*Social: Maximum distance creature tries to maintain with allies (lower means closer packed groups)

*Wary: Minimum distance creature tries to maintain with enemies

*Fear: Threshold of type difference to identify creature as enemy

*FearRnd: Randomization of fear on each check

*Swim: Ability to traverse water

*Climb: Ability to move up slopes

*Breed: Chance to reproduce each second

*Mutate: Drift of stats with each generation (higher mutate values decrease chance of successful birth)


My original concept was to make an entire ecology of randomly generated species with recursive stats and random mutations which eventually would cause species to differentiate. Basically, faithfully reproducing the mechanisms of evolution in a game. On top of this, I planned to either cast the player as a curator/hunter, who would hunt the animals for food and also try to cull them or lead them in order to produce a healthy balance of animal populations.

Obviously, I didn't completely achieve those goals.

I did, however, make a pretty good start, particularly considering the complexity of the idea (NEVER AGAIN) and the constraints of Ludum Dare.

So here's what IS in the game:

Random terrain generation: Creates height-map, populates it with water and grass

Random species generation: Creates stats within boundaries defined in code

Eating/drinking: I didn't have time to program meat consumption, so they all consume grass. If they don't encounter water soon enough they will start to die.

Species breeding: When two animals of the same species meet, there is a chance they will breed. This produces a third animal with a slight mutation

Species AI: Very simple, animals perceive each other and react. They will run towards allies and away from enemies based on their stat parameters. Species don't differentiate terrain at all, so be prepared to watch them drown themselves a lot and wander far away from food for no good reason.

Downloads and Links




Geti says ...
Aug 27, 2012 @ 2:54am

Interesting visually, but not much of a "game"

Colapsydo says ...
Aug 27, 2012 @ 8:37am

The concept is interesting, but there is not enough visibility of interactions between species and how the situation evolves.

Aug 27, 2012 @ 11:04am

There unfortunately isn't much inter-species interaction since they're all herbivores. They will run away from each other when they see each other and stick closely to others of the same species, but that's it. I've updated the game's description with a run-down of what all the stats do and a few notes on the AI (in the post-mortem), so it should be clearer from now on.

agersant says ...
Aug 27, 2012 @ 1:20pm

Not much of a game but I liked seeing the packs of dots herding =)

ChronoDK says ...
Aug 28, 2012 @ 12:34pm

I find it quite impressive that such a simulation can be build within 48 hours. Not a game no, but still good work for such a complex idea.

rwos says ...
Aug 28, 2012 @ 3:03pm

It's not a game, as there's no game-play to speak of - but as a simulation it's quite nice. It could use some speed-up button or so - interessting stuff happens only fairly rarely and it does take quite a long time to get to observe any long-term trends.

Still, well done - I would absolutely play a completed post-compo version of this. The planned features sound really interessting!

JH says ...
Aug 28, 2012 @ 3:13pm

Interesting simulation. It's funny to see how they flock together after a while when you add a few new species. Sadly, interactivity lacks a bit. But nice work for such a short timeframe.
You should check out berkano's and my own game, which are both simulations too. There are probably more but I have not found them yet.

lurkingshadow says ...
Sep 3, 2012 @ 9:33am

impressive simulator for 48 hours :D next time try working on interactivity.

infernet89 says ...
Sep 3, 2012 @ 10:21am

Nice simulation, too bad that the only interation is "generate new one".

You must sign in to comment.

[cache: storing page]