Are Wargame AIs Fated to Suck?

[wodin]

me too.

ORIGINAL: sulla05


I hate how some game makers use the phrase " multiplayer " as an excuse to not work on the AI at all.

[aaatoysandmore]

ORIGINAL: wodin

me too.

ORIGINAL: sulla05


I hate how some game makers use the phrase " multiplayer " as an excuse to not work on the AI at all.

Or "emphasize" mulitplayer vs solo play.

[wings7]

Wargaming on! [:)]

[MrsWargamer]

I can't get this comment out of my head.

When I discovered war games there weren't any AIs.

Those of us who had ftf opponents played them. Those that didn't have ftf opponents played the simulation as both sides. We had no choice, someone had to move the pieces and if it wasn't another it was us.

And war games sold just fine (well as fine as war games sold).

The AI wasn't there to create the hobby.

[sIg3b]

ORIGINAL: MrsWargamer

I can't get this comment out of my head.

When I discovered war games there weren't any AIs.

Those of us who had ftf opponents played them. Those that didn't have ftf opponents played the simulation as both sides. We had no choice, someone had to move the pieces and if it wasn't another it was us.

And war games sold just fine (well as fine as war games sold).

The AI wasn't there to create the hobby.

Uh, well, as far as computer games go, it did.

Boardgames are of course another story.

But with pc games, one would like the AI to be at least a decent sparring partner.

[sIg3b]

ORIGINAL: aaatoysandmore
That generally wouldn't be so bad if the AI played to actually win.

This.

I mean, yes, it´s terribly difficult to write a good AI, given that in a Wargame, unlike Chess, you can usually move every unit every turn.

Add FOW to this, and the AI designer has a Heraclean job.

But: What I *do* object to is the cases where the AI clearly does not even try to play to win.

The least I want the designer to do is try to beat the game and make an honest effort to let the AI imitate his play.

I feel let down when I do not get the impression to play a Golem made in the game designer´s image.

And the worst case scenario is when one gets the hunch that the AI is intentionally weaker than need be so as not to frustrate the casual players.

[MrsWargamer]

ORIGINAL: Tesuji

ORIGINAL: MrsWargamer

I can't get this comment out of my head.

When I discovered war games there weren't any AIs.

Those of us who had ftf opponents played them. Those that didn't have ftf opponents played the simulation as both sides. We had no choice, someone had to move the pieces and if it wasn't another it was us.

And war games sold just fine (well as fine as war games sold).

The AI wasn't there to create the hobby.

Uh, well, as far as computer games go, it did.

Boardgames are of course another story.

But with pc games, one would like the AI to be at least a decent sparring partner.

I really stopped waiting. I tend to be ok with games if they are multi player and the AI can be cut out of the loop. Because then you get a board game with no dust on the board the cat can't knock over the counters and you can play 3 separate complete games and not need three copies of the game. Some games in board game form are so big you can't even play them without a warehouse grade floor space. I would like to play all of Fire in the East, but will have to settle for War in the East, because I can at least find space for it.

More fun against a human. But as long as the game permits accurate parameters, realistic potentials, credible simulations. The moment you add an AI the words accurate/realistic/credible are NOT words you should be using.

[Joram]

The only AI I've ever really found properly challenging is Tin Soldiers. Other games can be challenging but it's either because the AI is given advantages (resource, die roll modifiers, or even on occasion ignoring rules) or the game itself is so complex I don't have the patience to think through the permutations and play more instinctual.

[EwaldvonKleist]

ORIGINAL: Iain McNeil

Currently neural nets are nowhere near being able to play even a simple wargame so there is no way to use them or make reusable code. Maybe at some point in the future but there are simply too many possibilities for it to work it out currently. There are just vast - billions of irrelevant moves that can be made that a human dismisses instantaneously that make it very hard for a neural net to evaluate. There are also many options that seems good in the short term but have serious long term consequences or put you in trouble against someone with any experience which a neural net would find almost impossible to see coming, at least now.

Thx for the discussion McNeil, it is always interesting to hear the opinion of somebody who is an expert of the industry.
Nuronal nets: The question is how much was invested to adapt them for wargames? If i understood the Google Go enginge right, the neural net was only a part of the AI. The developement had for sure much more time and money and experts than wargame developers can dream off?
Neural networks will maybe only be a kind of support, but not replace the "classic" AIs technologies for wargames. However, highly speculative because i am not an expert in this topic.

[aaatoysandmore]

Then tell me why Dave Landrey and Chuck Kroegel games were so good? Why am I drawn to games they made like "War of the Lance" and "Battles of Napoleon"? For games to not have the ability or impossible to create an AI that can challenge a player/veteran these were two of the greatest. As much as I tried to exploit these two games I couldn't. It was indeed almost like playing another human player.

[sIg3b]

ORIGINAL: aaatoysandmore

Then tell me why Dave Landrey and Chuck Kroegel games were so good? Why am I drawn to games they made like "War of the Lance" and "Battles of Napoleon"? For games to not have the ability or impossible to create an AI that can challenge a player/veteran these were two of the greatest. As much as I tried to exploit these two games I couldn't. It was indeed almost like playing another human player.

Errr, in the case of WotL, I could. 2nd playthrough, I found a guaranteed and not too complicated way to win and lost interest, even though I liked the game itself.

I am now in a moral double bind, though, because I don´t want to spoil your fun by telling you the shortcut strategy to defeat Evil. [:)]

BN is an excellent example, though, it´s one of the (very) few complicated Wargames with a great AI.

The AI covers its flanks, keeps its units in line, goes aggressively for the objectives and uses formations well. (It also cheats like there is no tomorrow, but nobody is perfect. [:)] )

I managed to totally destroy the French at Quatre-Bras. (I was playing the French.) My appreciation of Marshal Ney has increased thanks to BN.

[Kuokkanen]

ORIGINAL: aaatoysandmore

Then tell me why Dave Landrey and Chuck Kroegel games were so good? Why am I drawn to games they made like "War of the Lance" and "Battles of Napoleon"? For games to not have the ability or impossible to create an AI that can challenge a player/veteran these were two of the greatest. As much as I tried to exploit these two games I couldn't. It was indeed almost like playing another human player.

I searched YouTube for vids about those games. Didn't find any.

[wings7]

ORIGINAL: Matti Kuokkanen


I searched YouTube for vids about those games. Didn't find any.

"War of the Lance" came out in 1989...
https://en.wikipedia.org/wiki/War_of_th ... ideo_game)
"Battles of Napoleon" came out 1988...
https://en.wikipedia.org/wiki/Battles_of_Napoleon

I'm not surprised there is nothing on YouTube.

[bairdlander2]

ORIGINAL: wodin

I'd love for you to play a scenario or two a CMx2 game and then post a screenshot of the end of game AAR o prove how easy you find the AI in that game.

ORIGINAL: aaatoysandmore

I think it's lack of programming time put into AI's. I feel the game should be built around the AI not the other way around.

Take for instant Battlefront's AI in the Combat Mission series. It's more or less just a charge in your face AI that doesn't take into consideration flanks and feints. Two simple tactics that they couldn't take the time to program a counter or defense for.

I would like to see that also and a screen shot of Brazil taking Berlin and Moscow in HOI3,which he claims he did in another thread[8|]

[sPzAbt653]

came out in 1989... came out 1988...

And yet, so little improvement since then. If any. In fact, some backwards progress.

Most games are played solitaire. Get with it developers - stop making lame-ass lazy excuses or crying about how difficult it is to program units to move around a grid. Stop favoring the elitist minority Human vs. Human players.

RISE UP SOLITAIRE PLAYERS AND BE HEARD !!

[aaatoysandmore]

ORIGINAL: Tesuji

ORIGINAL: aaatoysandmore

Then tell me why Dave Landrey and Chuck Kroegel games were so good? Why am I drawn to games they made like "War of the Lance" and "Battles of Napoleon"? For games to not have the ability or impossible to create an AI that can challenge a player/veteran these were two of the greatest. As much as I tried to exploit these two games I couldn't. It was indeed almost like playing another human player.

Errr, in the case of WotL, I could. 2nd playthrough, I found a guaranteed and not too complicated way to win and lost interest, even though I liked the game itself.

I am now in a moral double bind, though, because I don´t want to spoil your fun by telling you the shortcut strategy to defeat Evil. [:)]

BN is an excellent example, though, it´s one of the (very) few complicated Wargames with a great AI.

The AI covers its flanks, keeps its units in line, goes aggressively for the objectives and uses formations well. (It also cheats like there is no tomorrow, but nobody is perfect. [:)] )

I managed to totally destroy the French at Quatre-Bras. (I was playing the French.) My appreciation of Marshal Ney has increased thanks to BN.

You did this on hard mode? I've played many a game and never found or seen an exploit in it. I do not see a fast way to defeat it. There's just no getting to the Highlord keep that fast.

Plus, all the diplomacy involved.

Also, I never said I couldn't defeat it. More that it's a very challenging AI to begin with.

Plus, I don't really look for exploits like some do. Even when I played Brazil in HOI I wasn't looking to take Berlin or Moscow, but, I did. [:)]

[aaatoysandmore]

ORIGINAL: Matti Kuokkanen

ORIGINAL: aaatoysandmore

Then tell me why Dave Landrey and Chuck Kroegel games were so good? Why am I drawn to games they made like "War of the Lance" and "Battles of Napoleon"? For games to not have the ability or impossible to create an AI that can challenge a player/veteran these were two of the greatest. As much as I tried to exploit these two games I couldn't. It was indeed almost like playing another human player.

I searched YouTube for vids about those games. Didn't find any.

Yeah, you probably won't since they were made in the 80's. Very few vids of 80's great games.

[aaatoysandmore]

ORIGINAL: sPzAbt653

came out in 1989... came out 1988...

And yet, so little improvement since then. If any. In fact, some backwards progress.

Most games are played solitaire. Get with it developers - stop making lame-ass lazy excuses or crying about how difficult it is to program units to move around a grid. Stop favoring the elitist minority Human vs. Human players.

RISE UP SOLITAIRE PLAYERS AND BE HEARD !!

Amen to that. When others have stood up and said there are "fun" ai's and "challenging" ai's and most developers go for the "fun" ones because they are easier to program and don't take much effort or work. All they really have to do is function going from point a to point b and many of them aren't even good at that.

[wings7]

Alot of complaints, let's hear some AI programming solutions...

[MrsWargamer]

It presupposes there is either A. a solution, or B. sufficient desire to generate the solution.

I guess I am with the camp that says I never asked for AIs.