Derek is currently putting the first FE build of the AI (first build I've checked in) through its paces.
It's a 3 person game, tiny map.
We'll see how it goes.
As the 0.26 version makes clear, the AI is about 25% done.
I'd put the odds of him winning at about 50/50.
The AI is still pretty piss poor compared to what it'll eventually be.
If you'd like someone that is significantly worse than Derek to make your AI look better give me a yell.
Derek - I'd like to point out Brad gives you a 50/50 chance of beating his piss poor 25% complete AI. Assuming that ratio stays true, you must have a 1/16th chance or so of beating the complete AI? Kael =
[Edit] I'm horrible at math. Brad must mean Derek has a piss poor chance of beating a 100% complete AI.
Is that 26% measurement of the AI including the Tactical AI? I hope the love is being equally distributed between the tactical side and the strategic side.
I think Tac AI is the 1% we are hoping towards.
It's great that you have a game designer with a great background give feedback on the AI, imho.
It's been forever - surely Derek must've been beaten by the AI already. Let us know what happened!
The first A.I. Build is passed. The system goes on-line November 2nd, 2011. Human decisions are removed from strategic defense. FE begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, November 3rd. In a panic, they try to pull the plug.
Three billion human lives ended on November 4, 2011. The survivors of the nuclear fire called the war Judgment Day. They lived only to face a new nightmare – the war against Sovereigns. The computer which controlled the Sovereigns, FE, sent two Champions back through time. Their mission: to destroy the leader of the human Resistance. Derek Paxton.
That is hilarious!
Looks like it was so traumatic for Keal that he won't dare to talk about it.This looks promising, crabs anyone?
I'll take that bet. Where do I send my money?
"It's been forever - surely Derek must've been beaten by the AI already. Let us know what happened!"
Well I think they mentioned something about there being 'epic' battles, perhaps he's in a stalemate wizard battle
C´mon, how was the game Derek?
I´m pretty sure Derek rushed the AI with no probs. and Brad stood behind him and couldn´t believe what he sees.
Maybe he got cancer from the AI and is now in the hospital. [e digicons]:'([/e]
heh
Definetly! It'll be interesting to see what he has to say.
It reminds me of RTS betas where competetive gamers found abusable holes in both the A.I and the balance but mostly how they discovered some new tactic/strategy that completely wrecked the balance cause the designers did not anticipate anyone playing like that.
50% chance of winning against an A.I that's 25% done?? Either Brad takes extreme pride in his work OR he doesn't think too highly of Dereks TBS skills
In all fairness, part of Frogboy's AI could be different flavors. He could be testing the "base" flavor (balanced game approach) versus an aggressive Ai or turtler etc.
I think Brad had his own TBS Skills in mind. I mean, we saw him playing WoM and he lost, because he moved his champs in a tactical battle towards the enemy. (Sry, Frogboy )
I suck at tactical battles.
well then give us an update on how derek has fared instead of leaving us in the suspense..
That comment could mean that Derek humiliated the AI in tactical battles.
Brad, you're good at the TBS aspect of AI right? Then why not apply the same logic to the tactical battles? Instead of thinking of the tactical battle as something different, approach it the same way you'd normally approach the TBS portion of a strategy game.
I look forward to the next round.
Brad versus Derek.
Just kidding.
Go Brad! Crush and humiliate him!
TheProgress, they're two different beasts, strategic and tactical. Tactical is much more about anticipating and countering your opponent's every single move, whereas strategic is about playing the optimal game and countering the opponent's grand strategy. While both use a TBS system, you can't really apply the exact same way of thinking into creating the AI for them.
Strategic != Tactical. Very different animals.
As for the Derek vs. AI, we quickly ran into some "world" settings issues. Basically, in 0.26 the seeding of monsters and such made expansion a bit of a luck of the draw. It's a balance thing.
If I get a decent starting point, I can absolutely crush the AI without batting an eye. If I don't, I get crushed instead.
When you guys get the beta, the first thing you'll notice is that nothing "spawns" throughout the game. It's all there on turn 1. The quests, the monsters, etc. They don't just "show up" anymore.
So for example, before you go and try for the ability to go on master quests, you should probably make sure you can actually get to master level quests because researching the tech doesn't just magically make said quest appear.
Now, the balancing part comes by deciding where to place these things when the game starts.
In the game I lost to the AI, what happened is that a large troll village was relatively near my starting point and when I walked too closely, they wiped out my key army.
After that, the AI had a big advantage over me.
In another game, the AI started out relatively close to the Wild Lands which are really dangerous to go too close to.
Getting that kind of balance nailed will be crucial over the next few months.
I don't see how tactical AI and strategic AI differ conceptually if you simplify the two to their basics. Like you said, you must counter the moves of your opponents. You must position yourself to take advantage of mistakes / openings the opponent makes. Divide and conquer. Make use of the terrain. Etc etc. Both AIs operate under different conditions to be sure, but they both need the same hard logic and reasoning. While the tactical AI is doing this with soldiers the strategic AI is simply doing the same with armies.
There are many great features available to you once you register, including:
Sign in or Create Account
