posted
I'll be interested to see what kind of fruit this bears in a few years.
Actually, this reminds me of an article I've read about the unexpected dynamics of NPC interaction in the upcoming Elder Scrolls: Oblivion.
Posts: 16059 | Registered: Aug 2000
| IP: Logged |
posted
That does sound interesting. For some reason it conjures thoughts of Phillip K. Dick. He would write the story from the point of view of one of the virtuals, perhaps seeing holes in their reality.
Posts: 2506 | Registered: Jul 2003
| IP: Logged |
posted
Oh, I know, eslaine...that's just the cultural lens through which I see it (i.e., the Matrix is a much more familiar cultural reference to me).
Posts: 4077 | Registered: Jun 2003
| IP: Logged |
quote:I don't see how someone could program a computer with all the parameters of reality.
quote:Yeah. That's the flaw in their experiment on a serious level.
Guys, the goal of the experiment is NOT to create a "virtual" version of earth. Its to make a fairly complex world, with a large number of "agents" which iteract with eachother. These "agents" have their own behavior programs and learning abilities. From the interactions from these agents often comes interesting results.
Heck, I took a couple weeks in my Artificial Intelligence class to build a simple world with "predator" and "prey" organsisms and I got some interesting results. I ended up with herding behavior in the prey, and group hunting behavior in the predators. Neither of these behaviors were hard-coded into my "agents".
That's all they are trying to do here, except they are trying to do it on a much larger scale/complexity, and with the added capability for their "agents" to communicate.
I would bet that they will get some very interesting results.
Posts: 5656 | Registered: Oct 1999
| IP: Logged |
posted
I want to know if they have an eccentric poet character who spouts erratic metaphors about sunshine and love. How do New and Emerging Worlds deal with him?
Posts: 8504 | Registered: Aug 1999
| IP: Logged |
posted
My project was on the extremely simple end of the spectrum eslaine.
The behavior emerged because 1) It made sense to stay near a food source 2) You don't have to look all over for a mate 3) When a predator sees you, there's a good chance you'll be able to get away.
Since the predator's and prey had the same base speed, one predator was unlikely to catch one prey in an even race.
So the predators who did catch the prey were usually the ones who both happended to run after the same prey at one time were the ones who got to eat dinner. Or when a whole group of predators attacked a herd at the same time, they could often snag one or two.
A prey agent who wondered away from the group usually ended up getting chased by a wondering predator. A chase would ensue, and when the prey ran out in front of a different predator, they would both start chasing him, and often catch him. Both predators would then "eat".
No, that was beyond the scope of what I did. I had a blast doing this project, but it was right before finals that I was doing it, and then I didn't work on it at all during summer like I thought I would.
If you look at the behavior algorithms in the programming part, you'll see how simple they were. Didn't have any time to develop more interesting ones .
But that's sort of the "theme" of Alife. Complex behavior emerging from simple rules. My favorite example of that principle are Boids.
Posts: 5656 | Registered: Oct 1999
| IP: Logged |