Yes, that will do it. But then they will figure out that running away is a good idea ;3
So give them more points of they kill them faster, and take away points as time goes on too. :V
Yes, that will do it. But then they will figure out that running away is a good idea ;3
Exactly this. And if you subtract too few points for hitting the edges, they'll start sliding along the edges as a tactic. If you add too many, they'll start carefully avoiding the edges or going much more slowly. In extreme cases they might even start (and have started) sitting still and waiting for the enemy to come towards them, leisurely dodging missiles all the way. They are crafty little buggers and they exist solely to exploit anything you do. Balancing for AI is hard. I'm considering making an AI to balance the AI balancing.Silverware wrote: ↑Tue Aug 08, 2017 3:16 pmYes, that will do it. But then they will figure out that running away is a good idea ;3
So give them more points of they kill them faster, and take away points as time goes on too. :V
Or just have an infinite field, and penalize for distance from enemy
Most programming is stuff you just "go ahead and do". If you don't, you never get anywhere. Anyway, to explain it simpler (as I tried on IRC): I basically reward things I want it to do, and punish things I don't want it to do. then it learns, through trial and error, over time, which things I want it to do and does those more often.Fawkes in IRC wrote:<Fawkes> it sounds horribly complicated
<Fawkes> like stuff you don't just go ahead and do
<Fawkes> it's an "awful(ly complicated) piece of programming weirdness"
<Fawkes> at least to me
<Fawkes> that sounded better?
Flatfingers wrote: 23.01.2017: "Show me the smoldering corpse of Perfectionist Josh"
Users browsing this forum: No registered users and 15 guests