I have searched around quite a bit trying to find out more about A-life how best to implement it, without much luck. Is it as simple as using arrays and comparing based on the rules that I have in mind? (I have a bad habit of making or thinking things should be more complex.)

It depends on what you want out of your A-Life project. Conway's Game of Life is quite easy to implement and gives some very interesting results. You can further extend Cellular Automata for more complex behaviors. On the other end of the scale, you have things like Creatures which is quite complex. I'd recommend reading Steve Grand's book for an interesting look into the project.

-Kirk

Quote: Original post by kirkd
It depends on what you want out of your A-Life project. Conway's Game of Life is quite easy to implement and gives some very interesting results. You can further extend Cellular Automata for more complex behaviors. On the other end of the scale, you have things like Creatures which is quite complex. I'd recommend reading Steve Grand's book for an interesting look into the project.

-Kirk

"Creatures" what is that?

Creatures....as in Norns, Albia....ahhh...okay....now I remember the game!

Yes, Creatures the game with Norns, etc. I'm a bit old school so my games of choice are a bit dated. 8^)

Quote: Original post by kirkd
Yes, Creatures the game with Norns, etc. I'm a bit old school so my games of choice are a bit dated. 8^)

But it was such a great game! Innovative... entertaining... and I hated it when my Norns died 8(

Thanks for the suggestion on the book Kirk.
The system I want to end up with would be fairly complex. For the time being I think that I'll start out with a simple system and add in layers as I feel the prior one(s) are in place & worry about finding most opt. means once the base is in place.

Timkin: I agree completely. Black & White was another good one, but it just didn't have the zest that Creatures (I, II, or III) had.

Lost: You're welcome. The book is very interesting, and has a reasonable amount of technical information while remaining approachable by a general reader.

A few words of warning, though - do everything in stages. I see many, many posts in which people put together some very well thought out systems and have a high quality architecture, and then plunk their agents into a fully developed environment complete with enemies where they have to learn to eat, drink, breathe, defend, move, etc. all at the same time. Start with the simplest survival tactics first and add more complex behaviors as you go. It sounds like you have a grasp of this already, but it's worth mentioning.

And, keep us updated on your progress. Its always nice to see final versions. Ooops....I suppose I should finish some of mine....PacMan comes to mind.... 8^)


-Kirk

Are you talking about free roaming agents, or cellular-automata like things sitting on the grid ? I had sytem built for latter, self-replicating cellular-automata like system (not exactly conventional cellular automata, each node kept complete "programm"), together with genetic algorithm for evolution (turnament-based). Genetic algorithm was extremly slow - took couple of day running for small improvment in efficiency. It's still sitting on my hard drive, may be I'll return to it later...

Quote: Original post by serg3d
Are you talking about free roaming agents, or cellular-automata like things sitting on the grid ?

I'm shooting for a simplistic (or maybe not so simple) ecology type of simulator.

Beings have land that has the potential to produce X amount of food, they are able to grow in population until the food resource is stress and then some will migrate to nearest area that has highest food producing potential. I also hope/plan down the road to add other resources (wood, stone, metal, hunting value, etc), with an end of adding more creature groups and interaction between them.