Honestly How Competant Will the AI be this time ?

[Northern Star]

Well said, I never played so far against the AI, only pbem and solo games (both sides).
The results were not satisfactory so far imo, but this is the best way to practice.

[Flaviusx]

Neal, I basically agree with all that, but this makes me all the more impatient to see more playtesting of the game between human beings. In the long run, I expect this game will create a flourishing PBEM community and that will really test the game model to destruction.
 
Already the very brief 1943 AAR has yielded excellent data in this regard. (The very weird air results.) And there'll be more surprises to come.
 

[Berkut]

I find all this talk about how we hope the AI will be great kind of bemusing.

The AI won't be great. Not to be a ass about it, but it is simply foolish to think that the team developing this game has anything like the expertise or resources to create a better AI than all the wargames that came before this one. Hell, programming a really good AI is probably twice as hard, ten times as hard? as programming the entire game to begin with.

And really, what is an AI anyway, when we talk about wargames? All it is is a set of heuristics - some rules - that govern what the computer will do. It is a combination, at the end of the day, of some rules that someone sits down and codes, and some scripts. And then it tries to execute those scripts and those rules by evaluating a situation.

So here is the thing - assuming that the people doing this are not some insanely intelligent computer science super-brains on the verge of a breakthrough in computer "intelligence", they are going to create the AI in the exact same manner that every AI for a game is made - by hand. And how "good" it is will depend mostly on how much time they can spend coming up with a basic plan. Then coming up with a sub-plan to deal with situation A that is slightly different. And then another sub-plan to deal with situation A1 that is slightly different than A, and A2 that is slightly different than A, and A3, and A4 and A5.

Then they can spend some more time coding up situation A1a, and A1b, and A1c. And of course A2a, and A2b, and A2c. And then some playtested will do something they didn't think of, so they will need a set of rules to handle A1a-i, and A1a-ii, and A1s-iii. And on. And on, And on and on.

So, the question is - how much time do they really have? And, by the way, a lot of this will end up wasted, since until you get a BUNCH of people playing the game, there are going to be decision trees the people doing the AI have never thought of - the playtesters will get some, but not nearly all of them by any means.

So - what am I saying?

I am saying it is not even possible to program a competent AI against decent human players. The only way it could be done, is if the rules and scripts were constantly tweaked in response to human play, and it would take a ridiculously large amount of work and would be an ongoing effort, not something they can code up and then call it a day. And even then...well, humans are tricksy beasts, I am still not putting my money on the computer against any decent human player.

You guys are asking for them to do something that they could not do even if they had a team of 10 people doing nothing but working on the AI, and they don't have that. The AI, at best, will be competent enough to be a stand in for a human opponent while the human player learns the mechanics of the game, and little more. If you cannot stomp a computer opponent once you learn the mechanics of the game, then either the AI is cheating, or, to be completely blunt, you aren't very bright.

I've been playing computer wargames for about as long as there have been computer wargames, and this has always been true, and will continue to be true.

[karonagames]

The issue of the AI really has to be looked at in relation to what players really want from the game overall. For me, I want to test out the other strategic options available to the Axis in 1941 and 1942 e.g. can Leningrad and Moscow be captured before the winter counter offensive. In my 1941 Campaign tests to date with the AI set on Normal I have captured Leningrad in October 1941, survived the winter counter offensive in pretty good shape, and gone on to capture Stalingrad in August 1942 and Moscow by November 1942. On challenging I have not captured Leningrad,had my army wrecked by the winter counter- offensive and therefore got nowhere near Moscow or Stalingrad, but I am not sure what this tells you about the strength of the AI or my lack of mastery of the game mechanics as I have only been playing for 31/2 months.

What I do know is that playing H2H is a completely different experience to playing the AI, and so far has probably been the more rewarding, as I am finding myself making the kind of decisions that I think my historical counterparts were making. I hope that a strong PBEM community develops around WITE, and that I don't get as frustrated as I did with the TOAW community where I never completed a full game of FITE or DNO because people dropped out.

[SGHunt]

Seems the AI is kicking some arse just now in Jon's AAR.  

t does seem strange hat it has not taken on Hoth with a massed armour assault, but when you have mass like the Russians do, maybe multiple attacks will be effective and less prone to the German back hand blow?    Even if most of these attacks are sealed off, and they will be I expect, if just one or two make a significant, strategic break through, then even an experienced player like Jon will have to seriously regroup and shelve his ambitions to just 'holding on'.

S

[Zorch]

I feel that the benefit of a good AI is to be able to validate historical alternatives...in my mind the playing strength of the AI only needs to be competent, not near perfect. I am less interested in beating the AI and winning the game than I am in learning some historical truths. "A war game is a paper time machine" (Jim Dunnigan).

It would be nice if the human player could give some strategic directive to the AI, to prevent it from aimlessly shuffling units around and falling victim to a well coordinated strategy like PyleDriver's (as someone already said).

I think the excellence of the German infantry is often overlooked; a good wargame like GG's WitE brings this out.

[Flaviusx]

The AI in Jon's AAR lost the game many turns ago. (Which is why I lost interest in it.) Moscow is gone, and the Red Army is dwindling away by salami tactics. Its situation is utterly disastrous. There's no sugarcoating this.
 
Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.
 

[Berkut]

ORIGINAL: Flaviusx

Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.

Indeed - and it is because I *know* that no matter how much effort they put into it, unless it cheats (which will become obvious if it does, and hence still be a failure), the AI is going to stink against decent play, I would much rather they NOT put a lot of time and effort into it. The difference between an AI that they spend a lot of time on, and one that they spend a minimum of time on, is going to be one of degree, and not much degree at that.

So, IMO, I would MUCH rather see development resources go into creating a seamless and smooth PBEM system, for example, rather than the standard half-assed system that most wargames "enjoy".

[PyleDriver]

Guys I played against the AI also going south also... In 42 the Germans still have the skill and the power and must and I mean must go straight after the Soviets...

[Neal_MLC]

ORIGINAL: Berkut

ORIGINAL: Flaviusx

Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.

Indeed - and it is because I *know* that no matter how much effort they put into it, unless it cheats (which will become obvious if it does, and hence still be a failure), the AI is going to stink against decent play, I would much rather they NOT put a lot of time and effort into it. The difference between an AI that they spend a lot of time on, and one that they spend a minimum of time on, is going to be one of degree, and not much degree at that.

So, IMO, I would MUCH rather see development resources go into creating a seamless and smooth PBEM system, for example, rather than the standard half-assed system that most wargames "enjoy".

Too each his own I suppose but I rarely play PBEM as I do not have the will to play a game on a schedule. For me I want a relatively competent AI. Please do not discourage the developers from making the AI as competent as they can. Please keep in mind that Matrix has to make this game appeal to the broadest market possible and to do that they have to consider the people who WILL play this game and this includes AIers and well as PBEMers. So let let them do what they do best.

I am guilty of being being impatient too but the folks at Matrix are doing the best they can.

[Wild]

ORIGINAL: Neal_MLC

ORIGINAL: Berkut

ORIGINAL: Flaviusx

Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.

Indeed - and it is because I *know* that no matter how much effort they put into it, unless it cheats (which will become obvious if it does, and hence still be a failure), the AI is going to stink against decent play, I would much rather they NOT put a lot of time and effort into it. The difference between an AI that they spend a lot of time on, and one that they spend a minimum of time on, is going to be one of degree, and not much degree at that.

So, IMO, I would MUCH rather see development resources go into creating a seamless and smooth PBEM system, for example, rather than the standard half-assed system that most wargames "enjoy".

Too each his own I suppose but I rarely play PBEM as I do not have the will to play a game on a schedule. For me I want a relatively competent AI. Please do not discourage the developers from making the AI as competent as they can. Please keep in mind that Matrix has to make this game appeal to the broadest market possible and to do that they have to consider the people who WILL play this game and this includes AIers and well as PBEMers. So let let them do what they do best.

I am guilty of being being impatient too but the folks at Matrix are doing the best they can.

Agreed. Please make the AI as good as possible.

[Berkut]

"As good as possible" isn't going to be very good though.

My point is that they should not spend any inordinate amount of time trying to create something that will, in the end, be pretty poor regardless. They can, and should, spend enough time to make an AI that can handle the basics and give the player a slight challenge as they learn the mechanics of the game, and the tendencies of the AI.

Beyond that, you will spend a lot of effort for very, very minimal gains.

Please keep in mind that Matrix has to make this game appeal to the broadest market possible and to do that they have to consider the people who WILL play this game and this includes AIers and well as PBEMers. So let let them do what they do best.

You are acting as if this is something under their control - as if they had the option to make a decent AI if only they have the will to do so. They do not. Nobody does - and many a company with vastly greater resources than Matrix and Co. have not managed to create a competent AI.

[PyleDriver]

First of all its Gary, not Matrix on the AI. I can't begin to count the saves I have sent to him... Were not done, and I scramble for ideas now...I have played so many lame game AI's that I have that I don't play anymore...I press Joel and Gary for a better one, give us some time...Right now 90% of the players will be hard pressed with it...

[Wild]

ORIGINAL: Berkut

"As good as possible" isn't going to be very good though.

My point is that they should not spend any inordinate amount of time trying to create something that will, in the end, be pretty poor regardless. They can, and should, spend enough time to make an AI that can handle the basics and give the player a slight challenge as they learn the mechanics of the game, and the tendencies of the AI.

Beyond that, you will spend a lot of effort for very, very minimal gains.

Please keep in mind that Matrix has to make this game appeal to the broadest market possible and to do that they have to consider the people who WILL play this game and this includes AIers and well as PBEMers. So let let them do what they do best.

You are acting as if this is something under their control - as if they had the option to make a decent AI if only they have the will to do so. They do not. Nobody does - and many a company with vastly greater resources than Matrix and Co. have not managed to create a competent AI.

Any gains to the AI no matter how minimal are worth it to me as i'm an AI only player who plans to play this game for a very long time. As has been stated before most people who buy the game will
only play against the AI.
Besides if anyone can program a good east front AI it's Gary. Let's give him the opportunity to do so, thereby increasing the enjoyment of all us poor deluded fools out there that have not acknowledged the wisdom that is PBEM.
I mean i really don't see what the problem is.

[wodin]

I'm an AI only player and I wish those who only play PBEM would to be blunt shut up....listen you will have to wait so that we (AI only players) get the best AI we can...I play a certain game PBEM but I don't begrudge the time put into the AI for those who dont...I think it's just selfishness....if you think the AI will be lame and you play PBEM then why are you worring about it? It has to be because you want the game quicker...well wait...don't be so selfish....your not the only one who wants the game and experience it...

The only way around this is if they release two versions one with and one without...relase the one without early to keep them happy..

It really does annoy me to be honest....

[Karri]

ORIGINAL: Berkut

ORIGINAL: Flaviusx

Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.

Indeed - and it is because I *know* that no matter how much effort they put into it, unless it cheats (which will become obvious if it does, and hence still be a failure), the AI is going to stink against decent play, I would much rather they NOT put a lot of time and effort into it. The difference between an AI that they spend a lot of time on, and one that they spend a minimum of time on, is going to be one of degree, and not much degree at that.

So, IMO, I would MUCH rather see development resources go into creating a seamless and smooth PBEM system, for example, rather than the standard half-assed system that most wargames "enjoy".

Obivous solution is to make the AI moddable.

[Berkut]

ORIGINAL: wodin

I'm an AI only player and I wish those who only play PBEM would to be blunt shut up....listen you will have to wait so that we (AI only players) get the best AI we can...I play a certain game PBEM but I don't begrudge the time put into the AI for those who dont...I think it's just selfishness....if you think the AI will be lame and you play PBEM then why are you worring about it? It has to be because you want the game quicker...well wait...don't be so selfish....your not the only one who wants the game and experience it...

The only way around this is if they release two versions one with and one without...relase the one without early to keep them happy..

It really does annoy me to be honest....

Well, you really should get over your annoyance. It only harms you - I can assure that I do not care one bit whether the truth annoys you. And more improtantly, the truth doesn't care that you are annoyed by it either.

In any case, you guys are simply not listening. I am not saying that they should ignore the AI because it isn't important to *me*, I am saying that you guys need to adjust your expectations because the AI is not going to be any better than the AIs in any number of other similar wargames - in other words, it isn't going to be very good. And it doesn't really matter how much time they put into it - because the reason AIs in complex wargames are not very good has very little to do with a problem than is fixable by pouring more and more time into it.

This has nothing to do with anything under the control of Matrix, Gary, or anyone else. It is the simple reality of how incredibly hard it is to program a decent AI for any game more complex than checkers, or chess, or any other "simple" game with an extremely finite and measurable space.

[Berkut]

ORIGINAL: Karri

ORIGINAL: Berkut

ORIGINAL: Flaviusx

Compared to historical results for that time period, the only metric I really care about, Jon won and handed the AI its ass. He spanked it hard and good. Furthermore I am completely unconvinced that a human opponent playing at Jon's skill level would do anywhere near so badly.

Indeed - and it is because I *know* that no matter how much effort they put into it, unless it cheats (which will become obvious if it does, and hence still be a failure), the AI is going to stink against decent play, I would much rather they NOT put a lot of time and effort into it. The difference between an AI that they spend a lot of time on, and one that they spend a minimum of time on, is going to be one of degree, and not much degree at that.

So, IMO, I would MUCH rather see development resources go into creating a seamless and smooth PBEM system, for example, rather than the standard half-assed system that most wargames "enjoy".

Obivous solution is to make the AI moddable.

That is a rather good idea on its own merits, but it isn't a solution - just a neat idea that would make the game more interesting, as people could devise new AI strategies. It won't make the game more challenging against the AI in general though - it will still be operating under the same limited constraints, and a competent human player will still be able to predict how it will respond to a given circumstances, and hence be able to exploit it.

[Berkut]

ORIGINAL: PyleDriver

First of all its Gary, not Matrix on the AI. I can't begin to count the saves I have sent to him... Were not done, and I scramble for ideas now...I have played so many lame game AI's that I have that I don't play anymore...I press Joel and Gary for a better one, give us some time...Right now 90% of the players will be hard pressed with it...

Awesome - sounds like they are pretty close to having the AI about as good as it is going to get then - that is certainly very good news.

[Hard Sarge]

Awesome - sounds like they are pretty close to having the AI about as good as it is going to get then - that is certainly very good news.

Hmmm, not so sure we are that close, Jon is sending in his AI saves, and so is Andy and I, along with the rest of the gang

(LOL, I got told I was doing good with a Explotable Bug, so my results shouldn't count, but until someone tried to explote it, we didn't know there was a bug there !!)

we are working