▲ 16 points
back
10 comments
I really wish it was true for serious games with a lot of metagame and strategy elements (Think Dota 2 or Company of Heroes), not just simple arcade toys...
Usually the AI in those games simply cheats horrendously and still is utterly stupid and having a real challenge in single player would be lovely
Well, as with all things game-related, the developers are trying to give you the best experience for the amount of time they were given to develop the game. I don't think spending more time and money on AI is a winning proposition for them.
On the theoretical side, I have a tiny bit of AI research under my belt, so I can make a few comments. The problem is simply that current AI techniques are geared towards solving isolated, "relatively simple" problems. These individual problems can be surprisingly complex, such as creating a very good algorithm for how to "micro" certain combinations Starcraft units. Even though there are ridiculous numbers of combinations, our current AI techniques are good enough to solve these.
The problem comes in when you try to combine multiple elements together. Combine your "micro"-algorithm with a general scouting and build order strategy, as well as having some kind of high-level decision making strategy which ties it all together... essentially this results in the complexity of each part being multiplied together. The number of combinations goes from "barely computable" to "won't happen, ever". What we actually need is something which is able to detect boundaries within the search space, decide where to start up new sub-algorithms, and how to combine them. To my knowledge, this field isn't very mature yet.
I believe we'll be getting interesting clues from how biological intelligences solve this problem within our lifetimes (not to say it will be completed solved by then). It will be very interesting, to say the least!
That's because writing a bot for most of those games is ridiculously difficult (because it "enables cheating" - maybe true - maybe not).
One "exception" to this rule is StarCraft [1]. But only because somebody hacked the core of starcraft to enable this. If you want to play a starcraft bot - be warned. They're not programmed to be fun, they're programmed to win.
There is an AI competition for StarCraft 1, and while having some impressive tricks, it's not even close to beating a competitive player.
That's quite a broad statement. I assume this is only for some very simple games.
I would love to see how _that kind of AI_ plays against person in DotA 2 or Starcraft 2.
I think an AI could be made to be pretty good at laning, at least against a static set of 1 or 2 opponents without others coming in.
Like the built-in AI in DOTA2, you mean?
The built-in DotA 2 AI is no near being good.
NoooOHHHooohhh! My high scores!!!!!!!!!!!!!!!!!!!!!