Matrix Games Forums

Forums  Register  Login  Photo Gallery  Member List  Search  Calendars  FAQ 

My Profile  Inbox  Address Book  My Subscription  My Forums  Log Out

The challenge of creating a competent AI

 
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 >> The challenge of creating a competent AI Page: [1]
Login
Message << Older Topic   Newer Topic >>
The challenge of creating a competent AI - 2/1/2002 6:03:00 PM   
USSMaine

 

Posts: 213
Joined: 12/23/2001
From: Maine (USA)
Status: offline
Matrix and 2by3 have a great challenge ahead of them creating a compentent AI opponent for a game of this scope. As a former programmer with nore than a few years under the beer line, I don't envy them at all. Where I think this game will really shine is in PBEM play and I can't wait to play either side :-) [ February 01, 2002: Message edited by: USSMaine ]



_____________________________

Post #: 1
- 2/1/2002 10:36:00 PM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
I think it is much more than a challenge. I can see the possibility of developing a reasonable A/I for UV. The goals are relatively obvious for each side, depending on the relative strengths of the two sides. The Japanese must secure New Guinea and Guadalcanal; if successful, then isolate Australia; all the while striving for decisive Naval battle while they have the advantage. For the Allies, preventing the Japanese from obtaining their objectives, then isolating and/or taking Rabaul. However, the entire idea seems hopeless to me for the entire Pacific. I do not believe that current technology will allow a meaningful A/I to be developed for so vast an area with virtually infinite possibilities for both sides. And it would open the door for endless argument. I think that once 2by3 and Matrix got a good idea of just how formidible this undertaking was, they made an excellent decision to insert UV, which could be brought to market more quickly and from which much could be learned. The expansion of UV to the full WitP would not be terribly time consuming except for the A/I. I would hate to see this game delayed for months trying to develop a meaningful A/I, which is doomed to failure, anyway. Forget the A/I. Develop a good LAN and PBEM version. Why beat up the A/I? I'm almost as dumb. Paul

_____________________________


(in reply to USSMaine)
Post #: 2
- 2/2/2002 2:10:00 AM   
DougAngle

 

Posts: 37
Joined: 4/4/2000
From: Arvada, Co USA
Status: offline
You guys are right. What a job! Here's hoping for a good lan/pbem option. Good luck Matrix!

_____________________________


(in reply to USSMaine)
Post #: 3
- 2/3/2002 2:07:00 AM   
USSMaine

 

Posts: 213
Joined: 12/23/2001
From: Maine (USA)
Status: offline
In my opinion a reasonable AI for the first stages of the war can be accomplished for either side. At least it will be a surprise the first few times through what the enemy is up to. The problem I fear is the post-Midway, post-Guadalcanal period as the Allies build up their strength and the Japanese attempt to recover (or some facsimile thereof). Of course battle results will greatly effect the course of when/where this occurs but I think the AI will surprise people the first time through at least. The late war period will be problematic at best ;-) I still look forward to this game and await the further testing of UV with great anticipation :-)

_____________________________


(in reply to USSMaine)
Post #: 4
- 2/6/2002 10:48:00 AM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
I think that you must envision a scripted AI, if you think the first months could be done. But then you refer to post Midway and post Guadalcanal. Why would they occur? If you have a scripted AI (say the Japanese), then Wake Island can be successfully defended, at least until Nagumo returns from the Indian Ocean, maybe longer. That, in itself, changes the script. What then? Ignore Wake, mount the Port Moresby offensive? Postpone the Coral Sea operation, mount a major operation against Wake. How can the AI figure that out? Do you assign values to targets and try for the most valuable? It would be totally beyond my capability to come up with even a block diagram decision tree. Wow, think about it. What a convoluted horror. Come to think of it, while we are waiting... How do you post block diagram IF statements to the forum? Paul

_____________________________


(in reply to USSMaine)
Post #: 5
- 2/6/2002 12:11:00 PM   
Chiteng

 

Posts: 7666
Joined: 2/20/2001
From: Raleigh,nc,usa
Status: offline
Actually I think a Heuristic AI would be much easier and far smaller code. Examples: IF fast CV == 6 or greater then attempt expansion
of defense perimeter.
Fast CV being the critical currency. Those many
crap Jap CV dont cut it. Another example: If total planes < 2000 then priority == produce planes. or If merchant ship losses last turn > 1 then increase ASW assets. Heuristic programming would at least mimic
human thinking.

_____________________________

“It is clear that the individual who persecutes a man, his brother, because he is not of the same opinion, is a monster.”

Voltaire

'For those with faith, no proof is needed. For those without faith, no proof is enough'

French Priest

"Statistic

(in reply to USSMaine)
Post #: 6
- 2/6/2002 6:47:00 PM   
USSMaine

 

Posts: 213
Joined: 12/23/2001
From: Maine (USA)
Status: offline
When I mentioned Midway/Guadalcanal I meant battles similiar to those and not necessarily those exact battles. But then I am stuck in a rut in Pacwar and always start my Allied offensive in the Solomons ;-) Fortunately I am not developing the AI :-)

_____________________________


(in reply to USSMaine)
Post #: 7
- 2/6/2002 9:52:00 PM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
Chiteng: Your location suggests an NC State affiliation. Is that correct? I hear a lot about heuristic approach for A/I. I'm not sure I understand this. In fact, I'm sure I don't understand it. Your examples are nothing but IF statements, common practice from FORTRAN to 'C'. What is the heuristic element? Your seemingly simple example: "If I have more than 6 fast carriers, expand the defensive perimeter" gets rather more complicated if someone asks specifically what expanding the defensive perimiter consists of. Taking San Francisco would expand the defensive perimeter, would it not? Can I accomplish that? Is this what the heuristic mechanism is for? Another simple example you give regards increasing ASW capability if you have merchant marine losses. At whose expense? We have unlimited demands on limited resources. Can a heuristic decision model help with this? Paul

_____________________________


(in reply to USSMaine)
Post #: 8
- 2/7/2002 3:15:00 AM   
Chiteng

 

Posts: 7666
Joined: 2/20/2001
From: Raleigh,nc,usa
Status: offline
I am not sure there is an easy reply. Heuristic programming is quite dated now,
I first was exposed to it in 1976 and yes
it was Fortran and on a Wat4!
(Waterloo Fortran Compiler 4) However it is useless to see it in a specific sence it must be looked at in a global sence. In the macroverse it merely means 'goal seeking'
You give the machine a hierarchy of 'goals'
(usually specific hexes) to secure.
They are weighted cardinal values, thus easy to
compare to each other.
Thus once the hierarchy is established it is easy to shift by simply replacing the list. I may be
explaining this badly. In combination with this you give the machine
a baseline of 'equitable' forces.
That means what 'you' /the programmer feel
is adequate to accomplish a task. For example: Phillipines == 3 divisions + air support more than enough. Celebes == two regiments/brigades more than enough. When programming chess this is even more easy
since you can quantify the ordinal values.
Certain squares are simply worth more. That is obvious to everyone. Where Heuristic programming fails is when
ALL the lists are compromised by either position
or losses. IE the USA lands in Hokkaido and
you assumed no one sane would do that.
OR You dont have any 'Fast CV' left so all
solutions with that assumption in them are void. In PacWar the AI totally collapes with ALL
airgroups placed in Tokyo and none getting
anything like needed replacements. A simple
subroutine would have prevented that. It is obvious that with Gary's programs he
utilizes a 'Group force directive' decision making
This is nowhere more obvious that BTR.
In fact he allows you the luxury of tinkering
with the AI by assigning your own priorities
should you wish to. However the AI fails in the CRUCIAL task of MIXING priorities. Sometimes
it simply isnt practical to try to hit the u-boat
pens, and yet that AI will still try. THAT is foolish. The model is fragile/brittle.
The german forces in that game are simply a list of regional directives that give an illusion of overall control. That is all it is, illusion.
There is no overiding command AI. The production system is useless because the it doesnt matter
what you produce because it will be caught and destroyed on the ground by the allied strafing
cheat. so why spend alot of time trying to fine-tune it. This is just opinion of course I have no accsess
to the de-compiled code. It is just my educated hunch. I dont know what restrictions Talonsoft
placed on the designer but they bear the ultimate
responsibility, not Gary. It was THEIR game.

_____________________________

“It is clear that the individual who persecutes a man, his brother, because he is not of the same opinion, is a monster.”

Voltaire

'For those with faith, no proof is needed. For those without faith, no proof is enough'

French Priest

"Statistic

(in reply to USSMaine)
Post #: 9
- 2/18/2002 11:49:00 PM   
Ron Saueracker


Posts: 12121
Joined: 1/28/2002
From: Ottawa, Canada OR Zakynthos Island, Greece
Status: offline
I'm clueless regarding programming. Therefore, what I'm contributing may be of little value. But, then again... Perhaps begin by dividing the areas in question into theatres much like was done in reality, perhaps further divide these theatres into sub areas etc. Assign each a priority value and a chain of command with varying levels of aggressiveness, competence etc., and have an overall doctrine (historical or random)which can react to the game results). I like the concept of heueristic approach...maybe this can be incorporated in a chain of command encompassing logistical, political, and military responsibilities. Just rambling on here, but man, I can only begin to see the vastness of the problem. It would be interesting to do the research (primarily with regard to how historical leaders reacted to situations of all varieties over the period in question) if supplied with a concept of how AI engine works.

_____________________________





Yammas from The Apo-Tiki Lounge. Future site of WITP AE benders! And then the s--t hit the fan

(in reply to USSMaine)
Post #: 10
- 2/19/2002 1:28:00 AM   
Jason629

 

Posts: 83
Joined: 1/21/2002
From: Charlotte NC
Status: offline
I think the easiest and most common AI solution is the one approach that all gamers are sure exists in any given time in any given game vs. AI. "Let the AI cheat!"

_____________________________


(in reply to USSMaine)
Post #: 11
- 2/20/2002 4:35:00 AM   
ratster

 

Posts: 166
Joined: 1/21/2002
From: PA
Status: offline
Making an AI that can respond to an ever chinging situation is a real bugger. Years ago, I wrote an AI program for a board game (VitP/AH). Even with years of experience with the game and knowing what a "winning" strategy was for both sides, plus plenty of input from other experienced players, it quickly grew to enormous proportions. The first turn(it was at most a 9 turn game) was easy, but variations increased exponentially every turn, especially when you need an AI to be on the offensive. Defensive AI is generally far easier. I never expect the AI, in any game, to be very good offensively, and I'm never disappointed. As long as it can hold its own defensively I'm content. The nice thing about doing a game that involves carrier action with limited intelligence is the computer can still catch a player by chance and spank him good.

_____________________________

" If it be now, tis not to come: if it be not to come, it will be now; if it be not now, yet it will come: the readiness is all"

Clan [GOAT]

(in reply to USSMaine)
Post #: 12
- 2/23/2002 3:34:00 AM   
Charles2222


Posts: 3993
Joined: 3/12/2001
Status: offline
I'm strictly an AI guy. I can say I won't even consider purchasing a game without AI, crummy or otherwise. Of course this is something Matrix or 2X3 would have to weigh, but as I've said before on these Matrix boards, those who play computer games, 70% of them play "nothing but the AI". Perhaps in the wargaming community it's a far smaller percentage, but look at the product here. So many people just automatically have a mantra and don't apply it to the situation. If you told me that BTR (Bombing The Reich) was something that would be better without the AI and should only be PBEM out of rote habit of saying that, and you actually got the game and tried to play it as PBEM only, you'd be kicking yourself and demanding a refund. Why? Because the game is MASSIVE assuming you don't want to delegate everything to your side's AI (which of course wouldn't be possible since AI is so bad), and therefore unplayable via PBEM (assuming you like to finish what you start). Is it possible that such a massive game as this could (and it's coming from Gary you know, which might just have a lot of BTR influence in it) possibly be too large for playing in such a way? IF, IF, this game incorporates very much of the detailed influence of BTR into it, you can forget this game being anything as "easy" to play as Pacwar.

_____________________________


(in reply to USSMaine)
Post #: 13
- 2/24/2002 12:05:00 AM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
Charles_22: I certainly agree with your points and I don't think anyone is suggesting that there be NO a/i with the game. I, at least, am suggesting that a competent a/i is impossible and Matrix should not waste months trying to get something that works well. It won't! Perhaps what would work best would be multiple scripted formats triggered by IF statements. Perhaps that approach would allow manageable limited actions to be followed. For example, a reasonable scripted action could be developed if the 1st Marine Division landed on Guadalcanal, depending on available resources (more ifs buried in ifs, buried in ifs....); but what if the 1st Marine Division assaults Wake Island, instead. That triggers a different scripted series of actions. And, of course, if there is only one script for a given IF, then gamers will quickly figure that out. IF (frog has wings), fly, hop. OK, I think I have it now. Paul

_____________________________


(in reply to USSMaine)
Post #: 14
- 2/24/2002 3:14:00 AM   
Charles2222


Posts: 3993
Joined: 3/12/2001
Status: offline
Paul Goodman: OTOH, you know there are people who are so eager to get it the way they want it, they don't want something they don't want 'delaying' things, but it's quite likely that anything but shorter campaigns will be unplayable PBEM anyway. Also, don't assume that once the AI figures what area you're concentrating on, that it should automatically 'if' in your direction. 'If' the Japanese have it in mind to take Australia at a crucial time, but send some small force to Midway first, the US AI would be in bad shape to throw a lot at Midway and therefore leave Australia neglected. Probably one of the large 'if' reactions should be based on how many carriers are spotted in an attempt to take some land, therefore it could conclude with quite a lot of certainty just where great offensives were intended. Would three main carriers, be enough to where the US AI should consider that a major thrust, should those carriers attack that land, or would it require more for a reaction (assuming the US's own plan wasn't more important if it had an offensive in mind elsewhere)?

_____________________________


(in reply to USSMaine)
Post #: 15
- 2/25/2002 1:32:00 AM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
Which really points out the impossibility of a really effective a/i. And look at the carrier count as a trigger. Mogami, for example, would figure that out right away and send the whole Imperial Army to Hawaii with NO carriers. The big IF would conclude no big deal and authorize liberal shore leave, where the crews would engage is some form of hueristic ritual. Paul

_____________________________


(in reply to USSMaine)
Post #: 16
- 2/25/2002 10:24:00 PM   
Ron Saueracker


Posts: 12121
Joined: 1/28/2002
From: Ottawa, Canada OR Zakynthos Island, Greece
Status: offline
Reference scripted AI. There were two basic aspects (influences) to Japanese strategy in the pacific. One was over-expansion, ie victory disease (an army flaw primarily); the other defensive consolidation (mainly a navy concept) based on the realization that the allies were too strong industrially and would have to (hopefully) suffer unacceptable losses trying to crack the shell to get at the nut. The allies suing for peace was a reality in their strategy. Is it possible to base the AI on something as simple as this? Maybe have the Japanese keep pushing until the've spread themselves too thin (parameters within AI) or have been made so by decisive strategic setbacks (similar parameters), forcing them into a defensive posture? Given the limitations to Japanese military and industrial output, this should be relatively simple Due to the scope of options facing the Allies, limiting play to human vs Japanese AI might be preferable. Restricting the Japanese initial opening of hostilities strategy to roughly resemble historical aims I believe is crucial to an AI game (shouldn't affect play too much given the resources available to the allies early on)and it sets the tone and feel for the conflict as realistic, and possibly keeps the game from becoming silly. We all know that the Japanese are going down, the variable is the price the AI charges the human player for achieving this goal (ultimate cost is allies sue for peace if human player thoroughly stinks); thereby setting up the basis for assigning victory points. If anyone was to lose to AI...give up. [ February 25, 2002: Message edited by: SGT_Saueracker ] [ February 25, 2002: Message edited by: SGT_Saueracker ] [ February 25, 2002: Message edited by: SGT_Saueracker ]



_____________________________





Yammas from The Apo-Tiki Lounge. Future site of WITP AE benders! And then the s--t hit the fan

(in reply to USSMaine)
Post #: 17
- 2/26/2002 1:21:00 AM   
Paul Goodman

 

Posts: 198
Joined: 7/5/2000
From: Portsmouth, VA, USA
Status: offline
I don't find Japanese defeat that cut and dried. I don't think you can overestimate the fickleness of the American public. I'm sure you would agree that North Vietnam was not as militarily or industrically strong as Japan. Yet.... While realizing what might or might not happen in a game situation is totally different from the historical event, I see numerous situations in 1942 which might have driven the U.S. out of the war. Two obvious examples right off the top. "Miracle at Midway" was exactly that. Even with the knowledge that we had of Japanese intentions, it is amazing that we won. Under game conditions, statistics take over. The Japanese would have won. Then what? If the Japanese had destroyed the invasion fleet as part of the Battle of Savo Island, could the American public handle the destruction of the 1st Marine Division? We were hiding actual losses from the public for fear of adverse reaction, as it was. I think one of two things would have happened, depending on where the press led the sheep. Either we would have negotiated some form of peace with the Japanese or the Pacific Theater would have gotten a much, much higher percentage of forces. As a balance to my thoughts above, the Japanese "Victory Disease" would have probably have led them to insist on conditions which we would not find acceptable, regardless of the level of disaster. The result of that, then, would be a complete restructuring of the war effort to a Pacific strategy. However, I do not think the game should allow for that. I think that the capture of Midway, Port Moresby, the Solomans and the Fiji chain, along with the destruction of all fleet carriers not in CONUS waters, should result in a Japanese victory. Finding a way to script all this is a different matter, entirely. Paul

_____________________________


(in reply to USSMaine)
Post #: 18
- 2/26/2002 11:44:00 PM   
Charles2222


Posts: 3993
Joined: 3/12/2001
Status: offline
quote:

Due to the scope of options facing the Allies, limiting play to human vs Japanese AI might be preferable.
I think quite the opposite. If for some reason the game was restricted to the AI always playing one side, it should always be the US. If the AI is really so lousy, then surely it should take up the side who won overwhelmingly, yes? I think there's a number of us who don't feel they accomplished anything if they've won as the side that won convincingly. I suppose a lot of people like the challenge of being inferior militarily early on in a game like this, but inevitably that same sense of inferiority will pass on to the Japanese. The only things the US is worth playing for is to see if you can take certain areas before the historical counterparts and the aforementioned early disadvantages. Considering this is a strategic game, if someone can find the data, it would be nice to see some sort of in-game figure and compared to the historic amount. Perhaps it could be something as simple as total war tonnage sunk and see if you can best it (maybe total warship tonnage sunk instead). Such a thing might be the only victory a lot of people may find of interest, since even if the program could allow for an Allied suing for peace, the human Japanese player might find it all but impossible to achieve, and if not, a hollow victory at best.

_____________________________


(in reply to USSMaine)
Post #: 19
- 3/1/2002 1:00:00 AM   
Ranger-75


Posts: 610
Joined: 6/29/2001
From: Giant sand box
Status: offline
Regarding the "miracle at midway":
In a previous thread I outlined how the Japanese violated 8 out of the 9 Principles of War, while at the same time the US paid special attention to several of these same principles, which is why The Japanese got their a$$ handed to them by the USN. Savo Is:
The Japanese admiral got cold feet, just like Admiral Pye in early Dec 1941 with Wake Is. Here, it was the USN who violated too many Principles of War. On a "Japanese Victory":
Admrial Yammomoto said (and this is often mis quoted and misunderstood) that the only way the Japanese would be able to win a war that she started with the US was to dictate peace terms in the White House. A lot of folks seem to take this as Yammamoto's thinking that he could actually defeat the US, but what he was really saying was that if you want to take the US on, you had better be prepared to fight it out right to their doorstep, because that's the only way you are going to defeat them. About US resolve (WWII vs. Vietnam):
These 2 eras could not be further apart. on the one hand, there was a nation technically at peace, savagely attqcked without a declaration of war, causing massive damage and loss of life. Even if Midway (or a similar battle) had gone badly, the US would not have given up, "germany first" would have changed to "germany as soon as we beat the Japanese Devils back into the stone age" The US and British CCS had several global contingency plans that they constantly updated in the event of a serious and unexpected turn of events (on either side). They would have reacted appropriately, but there was no question of throwing in the towel. The only way to acheive a "victory" as the Japanese, should be by racking up enough "victory points" to make it clear who the better player is.
(I personally like grinding the Japanese units into dust.) On the other hand, there was a nation tired from Korea, with an ambigious "incident" used as a pretext to support a government that wasn't even a real democracy. We weren't attacked, it was as if the government was trying to pull the country into the conflict. On AIs:
I used to program quite some time ago (I also programmed Fortran IV on a CDC Cyber 70 or something) It's tough programming an AI especially for something as complex as PACWAR or WITP. Its even harder to account for "stupid human tricks" like assualting Los Angeles (to destroy the factories there, like LA really had that many factories and like the west coast really had that many shipyards, the east coast had about 10 times the capacity of the west coast), or sending a TF with NO AIR SUPPORT to invade Hawaii, because the AI has all the carriers in Australia, or invading Ceylon, or invading India by the coast, or even invading Australia. The Japanese Army as, ineptly led as it was would not even consider any of these items EVER. They had no troops to spare and didn't want to get caught in another land campaign. Thier primary objective was to "settle the 'China indident'", They had 6 years to do that without the additional burden of being at war with the US/UK. After they started the war, the Army would be even more strained. We have the advantage of hindsight, which they didn't have, but one must also remember that the high command of the Japanese Army was bent on wwar and was living in a fantasy world for some time.

_____________________________

Still playing PacWar (but no so much anymore)...

(in reply to USSMaine)
Post #: 20
- 3/14/2002 1:47:09 AM   
berto


Posts: 20708
Joined: 3/13/2002
From: metro Chicago, Illinois, USA
Status: offline
I think we all agree that creating a competent AI to play hundreds or even many dozens of "turns" is extremely difficult to impossible.

I have a solution to this problem: Just bundle with the game dozens of shorter, AI-capable scenarios, both historical and hypothetical, scenarios lasting just a half dozen, or dozen, or a couple dozen "turns".

Although ideally I'd love to face an AI capable of fighting the entire war in an intelligent fashion--and the designers should try their best to achieve it in a reasonable time frame (say, within the next 3-6 months?)--I have to admit that I will simply never have the time to play a campaign game beyond a few dozen "turns" anyway. Wargaming is a hobby, not my life. Besides, I have many other games that will compete with WITP for my attention (Uncommon Valor, SPWAW, Europa Universalis, TOAW and other Talonsoft games, etc.), making it even less likely I will ever play a WITP campaign game anywhere near to completion. So, for me, a very good campaign-game AI is moot.

So, I suggest:

--For the campaign game, have a really good PBEM option, and a half-way decent AI.
--For the (hopefully) dozens of short-term scenarios, try for a darned good AI (and a good PBEM option).

BTW, I will never play PBEM. I would drive my opponent crazy with impatience waiting for my next move. I'll go weeks at a time when I simply don't have the time to wargame. No, AI is critical for me. If the game has no AI, I won't buy it. (In 40 years of wargaming, easily 98% or more of my games, both computer and tabletop, have been solitaire.) If the game's AI is weak, I won't play it for very long.

(in reply to USSMaine)
Post #: 21
- 3/14/2002 2:04:58 AM   
Von Rom


Posts: 1705
Joined: 5/12/2000
Status: offline
[QUOTE]BTW, I will never play PBEM. I would drive my opponent crazy with impatience waiting for my next move. I'll go weeks at a time when I simply don't have the time to wargame. No, AI is critical for me. If the game has no AI, I won't buy it. (In 40 years of wargaming, easily 98% or more of my games, both computer and tabletop, have been solitaire.) If the game's AI is weak, I won't play it for very long.[/QUOTE]


Agreed.

Is there anyway to have BOTH scripted tactics/strategies and a program to allow the AI to "learn" as it plays?

The new game "Superpower" is using this type of AI. It's not out yet, so it's hard to say how successful this approach might be . . .

Also, it might be helpful to get the experienced players' input of PacWar as to the strategies they use for both the US and Japan for each year of the war, and program in some of their devious strategies for the AI ;)

_____________________________


(in reply to USSMaine)
Post #: 22
- 4/7/2002 6:50:15 PM   
Ron Saueracker


Posts: 12121
Joined: 1/28/2002
From: Ottawa, Canada OR Zakynthos Island, Greece
Status: offline
Any further thoughts or decisions made regarding AI in WITP? I've read in UV forum that UV AI will be the model and UV will act as basic test bed.

_____________________________





Yammas from The Apo-Tiki Lounge. Future site of WITP AE benders! And then the s--t hit the fan

(in reply to USSMaine)
Post #: 23
Page:   [1]
All Forums >> [Current Games From Matrix.] >> [World War II] >> War In The Pacific - Struggle Against Japan 1941 - 1945 >> The challenge of creating a competent AI Page: [1]
Jump to:





New Messages No New Messages
Hot Topic w/ New Messages Hot Topic w/o New Messages
Locked w/ New Messages Locked w/o New Messages
 Post New Thread
 Reply to Message
 Post New Poll
 Submit Vote
 Delete My Own Post
 Delete My Own Thread
 Rate Posts


Forum Software © ASPPlayground.NET Advanced Edition 2.4.5 ANSI

0.766