Detailed battles AI sucks

[solostaran]

I have started several games as the CSA and I have used detailed battles in all of them. The problem is that the AI will attack with 1 or 2 bde and then call in reinforcements but the AI never attacks me usually I will attack them and destroy their units one by one and win the battles loosing only 200-300 and the AI 5000-6000.
even if I choose just to stay back and take defensive positions then the AI will not attack. usually their will to fighe is 8-9 and mine 16-18. I even had a battle with 2k CSA vs 46K USA and all the AI did was attack a garrison.
Is there any way to improve the IA's performance ????? this is really annoying since i enjoy the detailed battles..

[ericbabe]

Which difficulty level are you playing on?  (On other threads, players are complaining that it's too difficult to win battles.)

[Alan_Bernardo]

ORIGINAL: solostaran

I have started several games as the CSA and I have used detailed battles in all of them. The problem is that the AI will attack with 1 or 2 bde and then call in reinforcements but the AI never attacks me usually I will attack them and destroy their units one by one and win the battles loosing only 200-300 and the AI 5000-6000.
even if I choose just to stay back and take defensive positions then the AI will not attack. usually their will to fighe is 8-9 and mine 16-18. I even had a battle with 2k CSA vs 46K USA and all the AI did was attack a garrison.
Is there any way to improve the IA's performance ????? this is really annoying since i enjoy the detailed battles..

It is very hard to please everyone, if not impossible. I'll bet that of the posts on the internet, since the medium's beginning, more than half of the posters wished they could take their posts back.

This looks like one of them. Rashly posted, wild in content, with multiple question marks to end one sentence, this is a prime candidate.


Alanb

[Adam Parker]

ORIGINAL: Alan_Bernardo
Rashly posted, wild in content, with multiple question marks to end one sentence, this is a prime candidate.
Alanb

Ah. The question marks were mid sentence. I note this only because I watch a lot of Law and Order.

[Hard Sarge]

Well, hassle is there are a lot of things that could effect this

depends on the level of the game, the time, how good the troops are, what there morale is, how much supply they have, are they waiting on reinforcements, are they stuck in a swamp ?

plus there are different Tactics the AI can pick, and it may also not be too happy with the defence it is running up against

to be honest, with out more info, it can be really HARD to tell what you are seeing


???? I like to put ???? greeesh

next you guys going to be complaining about how people spell

[Alan_Bernardo]

Ah. The question marks were mid sentence

 
Actually they were not.  There is a sentence (i.e., independent clause) on each side. 
 

???? I like to put ???? greeesh

next you guys going to be complaining about how people spell

 
 
Seriously, it's just that folks who use multiple question marks at the end of sentences (as if one were not enough) or multiple exclamation marks, are typically over-hyped.  Rational thought is gone.  It's best not to post under such circumstances, until one settles down.  Personally I find it hard to take anyone seriously or to agree with anyone's perception of things when they use a crazy amount of question marks at the ends of sentences, when one is as good as twenty.
 
As such-- again, not being afflilated in any way with FoF-- I can say that the original poster's perception of things is pure nonsense.
 
 
Alanb

[Hard Sarge]

Well, we need to find out why, he is seeing what he is seeing, it may be something simple, or not

[Malagant]

This sounds like the problem several of us brought up in the Support forum...where the AI gets in to a fight that it decides it can't win, so it just sits there doing nothing unless "prodded".

I believe Eric had floated the idea of making the AI withdraw if it weren't actually going to take any offensive action.

Another recommendation would be to ensure the AI does not make a strategic move to put itself in a situation where there's a fight it can't win, which it seems to do quite often.

[freeboy]

well, in detailed combat it may take all day to get reinforcements, and if not on close setup all day justto find the enemy if you dont want your troops to be all over the place, I am a seasoned wargamer, sorry, 38 years enough? and as I see it this is one of the best ia... try playing the union and see how good the ia can be in detailed combat

[Artmiser]

Yeah do try that, confed ai has good troops and commanders, usually aggressive as all get out at least early on.

[Hard Sarge]

ORIGINAL: Malagant

This sounds like the problem several of us brought up in the Support forum...where the AI gets in to a fight that it decides it can't win, so it just sits there doing nothing unless "prodded".

I believe Eric had floated the idea of making the AI withdraw if it weren't actually going to take any offensive action.

Another recommendation would be to ensure the AI does not make a strategic move to put itself in a situation where there's a fight it can't win, which it seems to do quite often.

roger and agree, which is why I am trying to get all the details we can to be sure what and why it is happening

we got something in the works, so the AI will give up instead of just sit, and yes, the AI does make checks as to what it can or thinks it can win or lose before it moves, but it needs some more work
(one hassle is what is a Stonewall in HW, may be a easy win in QC, so the AI is not always wrong, when it makes it's judgement)

also as has been said, terrian and reinforcements can come in also, if the AI is calling for reinforcements, it will tend to hold in place until they come in and reach them, so it does not come in one at a time, plus depending on what tactics the AI decides it is going to try

Art, roger that, or if you get into the high levels of the game, the AI gets more morale, if the AI gets enough morale, it will get very agressive, you play the game on Col or one of the Gen levels and it will not give you any peace, you get the AI will morale of 6 or 8, and look out

[ravinhood]

First AI disappointment thread I've seen, but, only 1 person unlike the many I usually see. When an AI truely suks is when you get the game you start out on the hardest difficulty and win easily, that's a suky AI (RTW). I always start on the hardest difficulty and work my way down not on easy or normal. I see easy and normal for people who've never played a wargame or strategy game before or those that just want a casual experience and nothing complicated or hard or those that just can't stand to lose to a machine haha. I relish the joy when the AI beats me, that means it's hard enough and challeging enough I have something to accomplish. It's why I keep pushing for difficulty sliders with even more options to improve die rolls, morale bonuses, troup counts etc. etc. like Madden Football gives. I love all the difficulty sliders. The more options and sliders the better I think.

[Hard Sarge]

LOL
I pretty much agree (only I don't like games to be HARD because they cheat (like in Maddon, you have a guy with 90 speed and he is slower then the AI guy with 85 speed, because each level of Diff adds 5 to each stat)

but over all, one the key stats in this game is morale, the higher the morale, the more the AI is going to do

but it can have it bad points too

wait till you see a trench line with double rows of troops and each one is a 8 or 10 morale, talk about murder, and for both sides

[Jakerson]

I dont care about AI in most of games I use it mostly as training opponent but I try mostly in all games find human opponent its a lot more fun that way.

[siRkid]

I wish game companies could make AIs truly smarter at higher difficulty levels and not just better stats.  In FoF for example, at an easy setting the AI makes head on attacks while at the hard setting it tries more flanking maneuvers. You know things like that.

[Gil R.]

Kid,
In the "Command/Control madness!" thread in the Support sub-forum Eric posted the following message (the latest in a series of such posts on both this forum and the "Crown of Glory" one). This will explain why asking for superior AI often falls into the category of "easier said than done."

The AI is never going to be able perfectly to simulate the minds of commanders on the field of a Civil War battlefield -- there will always be little things wrong with the AI. I'm happy to work at identifying and correcting some instances of poor behavior, however, as I've mentioned, I have found in the past that often in doing this, the "fixes" that rule-out poor behavior end up crippling the AI in other ways and frequently create just as many problems as they solve... though not always! sometimes fixes do work, but it's a lot of effort to separate the wheat from the chaff, so to speak.

Every six months or so I write a little post on the theoretical limitations of AI; I won't go through the whole thing now, but will summarize. The way that AI works is by searching through something called a "state space" for the game, to look for optimal changes to the game's current state. The larger the size of the game's state space, the longer it takes the AI to find maxima within it. For most games that people enjoy playing, the state spaces are so enormous that AI's never actually search the entire state space, even for a single turn. A game like tic-tac-toe (or, uhm, noughts-and-crosses) does have a solubly small state space; a game such as Chess does not. I've calculated the rough size of FOF's state space, and to search through all the permutations of the state space for a single turn at PC processor speeds would take longer than trillions of times the age of the Universe. So most games using AI have to use approximation techniques to search subsets of the state space -- they have to identify vast parts of the state space to exclude from their search in order to whittle down the size of the space to something that is a searchable size. For a game like FOF, this means excluding more than 99.99999999999999% of the possible moves from the AI's consideration.

Programmers working with AI spend a lot of time refining the approximation to the search of the game's state space, and there are many documented techniques for doing this. Even for relatively simple games, however, these techniques have not been very successful. I've studied the AI for the game Go quite a bit in the course of trying to teach myself how to program AI. Go has a state space that is considerably larger than Chess, but much, much smaller than a game like FOF's. There have been PhD computer scientists who have dedicated their careers -- one guy I read has been working on this for 30 years -- to developing good Go AI algorithms, and there's a whole community of computer scientists who are working on solving the problem of Go's AI. Given all the enormous human effort poured into writing a good Go algorithm, it's disappointing to learn that the best Go algorithm that's been developed to date only has the Go ranking of "weak amateur" when playing against human opponents, and even that algorithm requires more computing power than a standard PC has and takes a long time between turns to make its moves. Given that FOF's state space is exponentially larger than Go's, and also considering that I'm not about to devote my entire career to developing a FOF AI algorithm, and also considering that there is not a community of PhD computer scientists working on this problem, to expect the AI to play above amateur level or (much harder) to simulate what historical generals might do is, I humbly submit, an unreasonably high expectation.

[Knu]

This same thing about detailed battles has been discussed in various threads, for example in this one:

http://www.matrixgames.com/forums/tm.asp?m=1323640

I cannot understand this constant excusing, that "it might be that you are playing this or that level", or "perhaps this and that scenario", or something along those lines.

There is several complaints, that the detailed battle seems to be broken in a sense, that AI cannot launch at least somewhat coordinated attacks. I just hope the developers will take these comments seriously, because this could be an excellent game.

I at least lost all the intrest to the game, due to the detailed battle problem and I'm hoping future patches will fix the problem, so I can get back to the game.

Happy Christmas to all!

Kimmo

[siRkid]

Gil R.,
 
I get what you’re saying about the complexities of AI programming.  What I am looking for are some modifications to the decision making process. For example, on easy setting the AI might conduct a charge even if the odds are bad, but on expert the AI would calculate the chance for success and only charge if it knows its going to win. Another example, on easy the AI picks targets at random, but on expert it concentrates its fire on a few units. Things like that are what I am thinking.

[Gil R.]

Kid,
Not being a programmer, I can't tell you how easy it would be to do that and get that right. But you might want to suggest it on the "Wish List" thread. (Eric doesn't like it, and rightly so, when people just say "You should make the AI better" but does indeed welcome concrete suggestions such as yours.)

[General Quarters]

I found the message posted by Gil quoting Eric on the challenge of designing an AI for a complex field, much more complex than chess, very illuminating and interesting. It gives me a lot more sympathy for designers and the challenges they face -- and a lot more appreciation for their creations.