The synthetic ecosystem has now been running for over a day. Sor far, the most successful tiger has managed to propogate its "genes" for seven generations. The most successful cow has only propogated its genes for three generations. Nothing goes extinct because I guarantee a minimum population, but so far, these creatures cannot survive on their own. Watching cows charge tigers makes this clear.
I'm also thinking that I'll have to convert the app to Tk. I've been trying to make it a console only app that can run on any operating system, but it's proven more difficult than I thought. I think Tk will solve these problems. Too bad I don't know Tk (yet)
Re:Balance each other
htoug on 2004-04-09T11:24:38
In a predator/prey system with only two species you usually get an oscillating system, where the number of individuals in each species varies in a sine curve, with the predators lagging behind the prey.Adding more species, eg grass that the prey 'preys on', gives a more interesting picture...
Re:Balance each other
Ovid on 2004-04-09T15:33:12
Adding more species
... Long term, that's my plan. I want to add bodies of water, larger maps (it's only 30x30 right now) and omnivores. First, though, I have to get it working
:) Re:Enough cows?
Ovid on 2004-04-09T15:25:20
I keep the number of predators and prey from falling below a certain preset value for each; they'd go extinct before they'd evolve. Currently, they each hover at at 9 to 10 animals.
I could add new traits to the tigers, but for now, I want to see what's wrong (if anything) with my model. The cows have bumped up to a maximum bloodstream of four generations and when this happened, my "world data" file was rewritten. In checking that, I saw some very curious results which I think are indicative of a bug in the neural network. If so, and if I can fix it, I might have better results. I'm starting on that tonight.
