Mods for a AI challenge

Matrix Games Forums

Forums Register Login Photo Gallery Member List Search Calendars FAQ

My Profile Inbox Address Book My Subscription My Forums Log Out

Mods for a AI challenge

View related threads: (in this forum | in all forums)

Logged in as: Guest

Users viewing this topic: none

Printable Version

All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Mods for a AI challenge

Page: [1]

Message

<< Older Topic Newer Topic >>

Mods for a AI challenge - 8/4/2007 6:28:59 PM

Barre

Posts: 27
Joined: 1/21/2005
From: Slighty left of center.
Status: offline

Played stock through late 43 as Allies until the Japanese began to fold like a wet paper bag. PBEM is difficult for me b/c of time constraints. I recently restarted as Allies on "very hard" and see the AI heading down exactly the same path to mediocrity. I'm interested in the mods, but don;t know which represents the best challenge to an Allied player. Love your opinions.
B

Post #: 1

RE: Mods for a AI challenge - 8/4/2007 9:56:19 PM

DrewBlack

Posts: 828
Joined: 7/3/2004
From: North Wales, UK
Status: offline

There is NO Ai its scripted.... every game you play agianst Jap Ai the same targets/bases will be hit and no counter attacks.

I have said it before, if and when Witp2/Gold ever comes around I would pay Real Good money for a half decent Ai

. The developers need to look at something like the Athena Ai in AGEOD latest American civil war game or the MAtrix DB series, they are not perfect but im sure would help Witp 100%.

Drew

(in reply to Barre)

Post #: 2

RE: Mods for a AI challenge - 8/6/2007 9:15:26 AM

dpazuk

Posts: 119
Joined: 2/25/2002
From: Ottawa, Canada
Status: offline

quote:

ORIGINAL: DrewBlack

There is NO Ai its scripted.... every game you play agianst Jap Ai the same targets/bases will be hit and no counter attacks.

I have said it before, if and when Witp2/Gold ever comes around I would pay Real Good money for a half decent Ai. The developers need to look at something like the Athena Ai in AGEOD latest American civil war game or the MAtrix DB series, they are not perfect but im sure would help Witp 100%.

Drew

I am not sure if this has been added to the wish list, or if it could even be accomplished within the existing game engine, but it would be great if the devs could look into a complete overhaul of the AI.

It would appear that the designers originally had in mind an AI that would allow you to learn the basic game mechanics (nothing more, nothing less), at which point the game would become a strictly PBEM affair. The AI was never intended to give a human opponent a run for his/her money.

I would LOVE to have an AI that would actively probe you for weaknesses and attempt to exploit them. Punishing you, for example, in stripping India and shipping everything to the front to create land death stars for Burma's recapture, only to land behind your lines and spoil the party. If you want this type of game, you have to play a human.

Quite frankly, given the sheer scope of the game, I really doubt if a 'smart' AI could even be developed.

< Message edited by dpazuk -- 8/6/2007 3:58:33 PM >

_____________________________

Blah Blah Blah

(in reply to DrewBlack)

Post #: 3

Smart AI - 8/6/2007 5:18:50 PM

herwin

Posts: 6059
Joined: 5/28/2004
From: Sunderland, UK
Status: offline

Smart AI is very hard. I'm currently the house neuroscientist for an intelligent systems research group, and our interest is in how to use models of biological intelligence to solve that problem.

The purpose of brains is to control behaviour. To do that, the brain has to be able to judge outcomes. One approach is caching previous outcomes--this is how habit works and how actor-critic models work. The problem is that the state space that contains your cached values is enormous. Another approach is predicting future outcomes. That requires an internal model of the external process--yet again something that tends to be enormous. On the other hand, there's good evidence that goal-directed behaviour works in that way, since modifying the goal value changes the behaviour. I strongly suspect the internal model is solved in parallel, since the solution is found very quickly in mammalian brains, and the response to goal value changes is also very fast. We don't know how to model this yet in silicon. My own suspicion is that the goal values used in behaviour are found using both methods, and an averaged value is actually used, with the two sides being weighted based on the reliability of their predictions.

In game terms, the current approach of scripting is probably a good starting point, but the scripts should be supplemented by tactical look-ahead, and the value of actions should be a weighted average of the expected values of the scripts and the predicted values of the look-ahead. Maintain variability measures and current values associated with scripts and look-aheads, so that as the game continues, the AI converges to a better estimate of the current value of actions.

_____________________________

Harry Erwin
"For a number to make sense in the game, someone has to calibrate it and program code. There are too many significant numbers that behave non-linearly to expect that. It's just a game. Enjoy it." herwin@btinternet.com

(in reply to dpazuk)

Post #: 4

RE: Smart AI - 8/7/2007 12:42:27 AM

Naskra

Posts: 325
Joined: 3/12/2005
Status: offline

Barre, you might try playing as Japan. The Allied AI is no better and can still be crushed, but the Allied infinite supply of materiel will require more planning on your part to overcome. Fun things to do:
Kill all the Brits.
Kill all the Chinese.
Kill all the Russkies.
Kill all the Aussies and the pesky little Kiwis too
Get your economy over 20K HI
Sink 4000 ships
Liberate San Francisco.

(in reply to herwin)

Post #: 5

RE: Mods for a AI challenge - 8/7/2007 1:56:43 AM

faraonej

Posts: 18
Joined: 3/13/2007
Status: offline

I'm in the midst of a stock game -- My first. It's January '43. I have managed to liberate the Solomons, the tail of New Guinea, The Gilberts, and almost all of Burma. The AVG has racked up over 1000 (count 'em) kills. In the Rabaul campaign, I managed to sink 9 enemy battleships (including Yamato) several CA's & DD's with repeated raids by B-17's and B-24's after they had supressed the airfield. BTW, none flew below 10,000 feet....

There has only been one carrier sunk on either side (Hermes).

Well, I'll rescind that. The AI is either very lame, or I was born 50 years too late. (OK, it's lame) Why the ships remained in port while they were pasted, pulped and pooped on is beyond me.

Regardless, I'm enjoying this. Maybe I can end this in 1944 and not use the Big One.

Cheers,

Vlad the Impaler

(in reply to Barre)

Post #: 6

Page: [1]

All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> Mods for a AI challenge

Page: [1]