Are we looking at CiV's AI the wrong way?

[Arsenic Steel]

Random question -

If you have a powerful military, what reasons should be in place where you WOULDN'T want to conquer that neighbor, even if you aren't setting out to conquer the world?

Because it a generally accepted idea that peace > war.

 

[testdummy653]

Having a military does not effect a civs SP or cultural output. There is a limit to the number a cultural buildings and Wonders that can be built in any given era.

The flaw in your logic is that you infer having a larger military force means that a civ is going to conquer others, it could be present solely for defense. Sadly that flaw is true in civ V and part of what is wrong with the AI's victory conditions.

No actually i did not infer that, i'm inferring that the Civ could of spent the time producing something more useful like another worker, or spent it on building culture building, or a market, but instead they need to waste those hammers on a military unit.

Also you missed the point that i made is that each civ get assigned a victory condition at the beginning of the game, and that determines how they play.

Let me break this down for you again.

I would want to play as a Cultural victory with my nation of let say France. Ideally i build my cultural building first(giving them priority), then i build my libraries, and markets. I would build an occasional military unit.

If i lived in a box, with no one to fight with that would be fine.

But that never is the case, because when war is declared you have to switch off from building markets, and temples, to building military units. You can rush buy, but that still a could been put to a better use (bribing, rush buying buildings).

Now in our case, as humans we get to fight against a horrible AI in combat. Plus as human we can plan ahead logically in seeing hey Siam has a huge military and they will probably come at me next, so i need to bulk up when i can afford it. (even then your losing hammers)

But the Cultural AI, has to fight a Military AI (both are the same AI, just they have different goals), and now the Cultural AI has to produce military units, that will be eaten up by the military Civ's army. Those turn are once again could of been used to build a recently discover opera house, or build a harbor. They may still get built (if the cultural civ survives), but they will not be built right away, or when they become available.


If you want to feel what the Cultural AI feels, try a one city challenge going for cultural victory on standard Pangea, at like diety, you can't build your broadcast tower right away(or ever) because you need to build that mech infantry to defend yourself against the Persia horde. (if you make it that far).

 

[testdummy653]

Because it a generally accepted idea that peace > war.

Not according to Rambo... Just saw the new one last night.

 

[testdummy653]

Are there really people saying that the AI can't win another way because of the game rules ?

How come this 30+ city civ, ALONE on its continent, with double my income, land, tech and 5 times my army cannot win diplomatically or space ship and let me win culturally ? There's no excuse for this. I don't know how anyone could blame the game rules and not the AI developpers on this -_-.

One i should point out that i'm not blaming anyone.
I'm just staying what i think logically is going on.

Two, game rules have to come from somewhere, and yes i'm saying that the game rules, makes it nearly impossible for any other civ to win other than conquest.

Three, I think people have missed this, so i will reinstate it in every response i give. IMO, the game assigns them victory condition to go for and they will strive for that condition, and they won't change to another victory condition.

So Siam is going for a military victory, and has taken the entire continent, it will not attempt any other victory even if they are at researching the redundant future tech. They will strive to conquer everyone since that what they are told to do. (but ai flaws make it hard for them to understand water, so most likely they will lose).

Random question -

If you have a powerful military, what reasons should be in place where you WOULDN'T want to conquer that neighbor, even if you aren't setting out to conquer the world?

Think of this in term of real life. Why does the China, not declare war on other nations? What are the disincentives for them not to wage war? They do have the army to go toe to toe with anyone?

If you can implement the same type of reaction that would happen if China did go to war with another nation, that would be a game changing, and industry changing experience.

Please do not make this an issue... I don't want this thread go to China vs. (your country), who will win

 

[cccv]

If you can implement the same type of reaction that would happen if China did go to war with another nation, that would be a game changing, and industry changing experience.

Please do not make this an issue... I don't want this thread go to China vs. (your country), who will win

Well, I guess they kind of do that now. If you fight with other civs, all the other civs hate you for being a warmonger, right? I presume this would lead to people no longer cooperating with China and possibly DoWing on them, which is fairly similar to what might happen in actuality.

I guess there are two problems with the way they do it now. First of all, all civs will hate you anyway, so it's your better option to have them hating you because you have your sword to their throat so they can't do anything about it. And second, they don't simulate it very well. ie, the AI makes no distinction as to why you're fighting. If another nation DoWs on you and you take their city just to get them to stop fighting you, you're still the evil aggressor.

So maybe it would help to make it harder for the AI to hate you so that it means more when they actually do, and of course as other people keep saying, to make it so you won't look like the evil one for finishing a war you didn't start rather than letting the AI throw troops at you forever. But that's what everyone says.

 

[UWHabs]

China doesn't go to war because they can't afford the 2 unhappiness from each city they capture, obviously.

The AI really should be reconsidering their target VC every few turns. So every 10-20 turns, they scan through each one and say, "okay, given what I know now, how close am I to each VC." That would mean that a target who took their whole continent but thinks "whoa, I don't want to go overseas" would turtle up for space. As long as there's checks in there so your 3-city Gandhi empire with a 15-city Monty empire next door, who is massing cavalry at his borders, doesn't think "well, I guess I'm not going to win conquest. Let's build some culture buildings!"

For all we know, it could be like 3 lines of code change to make going for culture the dominant AI VC. I mean, it's a nice change to have an AI actually try to win, as opposed to civ4 (pre-BtS) where they would steadfastly refuse to invade their friendly neighbour, even if it would be enough to put them over the domination threshold.

 

[Arsenic Steel]

OCC? Didn't you already say in your first post that India would be building a few cities to get a lot of culture? If you want to make the criteria a Pangaea cultural win why not also start talking about an archipelago domination win since you are trying to use the map to tip the scale in favor of your opinion.

Unless you are teching amazingly high there will be times where there is no wonder to build and all accessible cultural buildings have been made. Also not every city will have enough to warrant trying to build a wonder. There are enough turns to build some(not all) defensive structures in cultural cities, military units(not enough for no1 but a lot better than last place), and some wealth infrastructure.

Military strength is not just for waging war, at least that was true until V. There is no logic about how the AI is going to attempt to win via victory conditions. The AI leaders are only out for land, high score, and will declare war on whoever is close and weaker(please no social commentary).

 

[testdummy653]

OCC? Didn't you already say in your first post that India would be building a few cities to get a lot of culture? If you want to make the criteria a Pangaea cultural win why not also start talking about an archipelago domination win since you are trying to use the map to tip the scale in favor of your opinion.

Unless you are teching amazingly high there will be times where there is no wonder to build and all accessible cultural buildings have been made. Also not every city will have enough to warrant trying to build a wonder. There are enough turns to build some(not all) defensive structures in cultural cities, military units(not enough for no1 but a lot better than last place), and some wealth infrastructure.

Military strength is not just for waging war, at least that was true until V. There is no logic about how the AI is going to attempt to win via victory conditions. The AI leaders are only out for land, high score, and will declare war on whoever is close and weaker(please no social commentary).

Okay once again, you didn't read/understand my post... there is always something better to build, always!!! I rather build gold, or research when I'm going cultural or diplo, even science. The AI wants to win, and they will attempt to win their cultural victory through producing whats best for their victory condition.... When at war they produce military units which is not as useful to cultural AI. This is an example opportunity cost (http://en.wikipedia.org/wiki/Opportunity_cost). Even in ICS, it better to have your gold cities producing gold, than units...


Also i wasn't using a map type to tip in my favor... I was saying if you want to play like how a cultural AI would play, play on pangea,one city challenge going for culture, because their you have to make choices(opportunity costs) such as trying for culture victory condition or build some military units....

I was using france as my example in that specific post, and i said let for an example let play as france.... (so once again read!!!)

And yes there appears to be logic to how they will win... Look at the original post.

Honestly, I don't know if you need me to break down for you again... Let me know if i have some apparent contradiction(quote it) and i'll either recant or try to explain why i'm right

If you want me to rebreak it down my theory, then i can do that for you too...

 

[testdummy653]

China doesn't go to war because they can't afford the 2 unhappiness from each city they capture, obviously.

The AI really should be reconsidering their target VC every few turns. So every 10-20 turns, they scan through each one and say, "okay, given what I know now, how close am I to each VC." That would mean that a target who took their whole continent but thinks "whoa, I don't want to go overseas" would turtle up for space. As long as there's checks in there so your 3-city Gandhi empire with a 15-city Monty empire next door, who is massing cavalry at his borders, doesn't think "well, I guess I'm not going to win conquest. Let's build some culture buildings!"

For all we know, it could be like 3 lines of code change to make going for culture the dominant AI VC. I mean, it's a nice change to have an AI actually try to win, as opposed to civ4 (pre-BtS) where they would steadfastly refuse to invade their friendly neighbour, even if it would be enough to put them over the domination threshold.

Well what could happen with that is the domination AI would stay domination, and conquer everyone they can along as they aren't across the sea, go ICS and then switch to Space and win at year 1457

 

[polypheus]

Random question -

If you have a powerful military, what reasons should be in place where you WOULDN'T want to conquer that neighbor, even if you aren't setting out to conquer the world?

Well IRL (and even in MP games with a lot of high level rational players), one of the main reasons is that such a move could be threatening to players other than the weak neighbor being attacked and a coalition might form against you. Look what happened when Iraq invaded and conquered weak Kuwait. IRL nations prioritize their threats and help each other (even temporarily) to counter the biggest threat ATM.

Civ5 though doesn't have this "balance of power", big picture idea in mind. All AIs just blindly and dumbly become quickly hostile to each other and especially the human and then go all out war with everyone against everyone. Civ AIs often times even do things that help the strongest faction in spite that it is against its long term strategic interests. The thing here to realize in a real world and even MP context is that cooperation and alliances and balance of power considerations play a huge role. But in Civ5, a player that is actually helping out the AI by weakening its most dangerous foe is not seen as a friend but just another equivalent "warmonger".

Its somewhat laughable when people say that Civ5 AI is good because of playing more like humans and going all out to win. The fact of the matter is that in a good MP all-human game, players do see the value of temporary alliances and friendships and balance of power considerations just like IRL. All I can say is that if I played a MP Civ game but behaved exactly like the irrational Civ5 AI, I'd pretty much alienate everyone and get beat down hard. And if my ally and I are strongly cooperating against everyone else but everyone else is only playing completely for themselves without likewise cooperating, we would simply defeat all others one by one with our combined might.

As for AIs in Civ5 being better because they are far more aggressive and "play to win", all I can say is that Civ4BTS with betterAI and aggressiveAI does this a lot better.

 

[Arsenic Steel]

Fine if you want quotes:

Okay once again, you didn't read/understand my post... there is always something better to build, always!!! I rather build gold, or research when I'm going cultural or diplo, even science.

The idea of what is "better" depends on the player. In a game where wars happen because of proximity and weak armies then building some defenses and military is a "better" use of hammers after getting most of the cultural buildings for that era done. Science, wealth, military, and culture are always building options. Once a victory goal is selected then those buildings related to the condition become more important to make and causes the others to become equally situational.

The AI wants to win, and they will attempt to win their cultural victory through producing whats best for their victory condition.... When at war they produce military units which is not as useful to cultural AI.

If the AI wants to win then when playing continents, archipalgo, or any map with islands then the AIs on secluded islands would try for victory via culture or space. Yet the AI does nothing but sit on the island contently researching military tech since there is no other civ to war with on that land.

Also i wasn't using a map type to tip in my favor... I was saying if you want to play like how a cultural AI would play, play on pangea,one city challenge going for culture, because their you have to make choices(opportunity costs) such as trying for culture victory condition or build some military units....

You don't need to use OCC to say cultural victories have opportunity costs. My point is that military and defense also are choices that are part of opportunity cost evaluation. Building some military is not mutually exclusive from trying to get a cultural victory. Building walls makes cities easier to defend, while building a market makes it easier to build wealth, and neither directly make it influence culture.

And yes there appears to be logic to how they will win... Look at the original post, and then read it..

There is no logic represented there. All your first post does is arbitrarily assign imaginary victory goals to a bunch of civs.


Your theory doesn't really hold up if you use a the Improved Demographics mod and you can see that all AI civs devote most of their research to military tech and aspects that allow them to gain the most land possible.

 

[testdummy653]

Fine if you want quotes:
I wanted quotes from my other threads.. not the one before, but i appreicate it

The idea of what is "better" depends on the player. In a game where wars happen because of proximity and weak armies then building some defenses and military is a "better" use of hammers after getting most of the cultural buildings for that era done. Science, wealth, military, and culture are always building options. Once a victory goal is selected then those buildings related to the condition become more important to make and causes the others to become equally situational.

It depends on the AI specific victory goals....., Not a HUMAN player, but i agree with what you saying.

But once again one of my points is still alluding you... The AI whose goal is domination victory will, build more military that of AI cultural, and since they are both crippled by the horrible combat AI, the most units usually win in wars bwt AI... so that why AI whose goal is domination either wins, or forces the cultural victory to lose....

If the AI wants to win then when playing continents, archipalgo, or any map with islands then the AIs on secluded islands would try for victory via culture or space. Yet the AI does nothing but sit on the island contently researching military tech since there is no other civ to war with on that land.

Okay i addressed this somewhat in the main post, but i can explain that better here. (this is actually a very valid point).

For some reason the AI sucks at waging wars overseas(this will explain archipelago and continents). This is a proven fact. (as it is repeatable, and observable, and maybe measurable)

WHY?
IMO, there are a few things that contribute to this.

The AI looks at it military, and sees numbers, not types. This means that AI could have an army size of 300,000, but only have 6 ships and the rest are land units.

The AI doesn't know who to protect its units at sea.

Another reason why the Conquest AI, doesn't seem to be able accomplish its goals, is that AI in general, build up it military force on the enemies borders first, before DoW. Note: I have seen wars declared every once in awhile, with this happening, but its not often.

Finally, I play Archipelago lot as England, and I do get DoWs call on me all the time.

That is why, conquest doesn't do so well on those maps.

Now to address your other issue....

The AI DOES NOT SWITCH VICTORY CONDITIONS.
But to your point, play a game on islands and watch how many times a Civ will declare a war on someone. (City State included) If a Cultural AI gets a DoW on itself, it will produce units, plain and simple. Even if the cultural AI doesn't even get attacked, the AI will perceive that the attack will come from the aggressor. If you can play a game on Archipelago, with no wars declared and no citystates (on a decent difficulty), and you not trying to win (because honestly the human will be better than the AI in this game), that you might find your self surprise by the results.

You don't need to use OCC to say cultural victories have opportunity costs. My point is that military and defense also are choices that are part of opportunity cost evaluation. Building some military is not mutually exclusive from trying to get a cultural victory. Building walls makes cities easier to defend, while building a market makes it easier to build wealth, and neither directly make it influence culture.

No, your stilling missing the point, with out war as a threat, what do you need walls for?
Nothing.(unless you have UA that gives you gold, or a UB)

Markets are more useful that walls, and that the point!!! The AI has to chose between useful option A or less useful, but key to survival B. Think of it as a fight between Long term goals, and short term goals. Long Term the AI wants to win Culturally, but short term it need to survive now. Ideally the Ai wants to win, but it knows if it does produce military units, instead of something more useful, they can attempt to survive and to fight the run-away conquest AI. This is where opportunity cost come back into play.

There is no logic represented there. All your first post does is arbitrarily assign imaginary victory goals to a bunch of civs.

Not to be nit picky, but these victory conditions are not imaginary. I won a diplomacy the other day, and i won a conquest the day before that. (JK) But really no logic is a stretch, can you give me bad logic?

Your theory doesn't really hold up if you use a the Improved Demographics mod and you can see that all AI civs devote most of their research to military tech and aspects that allow them to gain the most land possible.

One, I'm not sure how this mod works.. But lets assume that its right, and that i gonna ignore your last part about land(I gonna flat out and say it that this is not true), and focus on technology aspect.

First off, I do not factor tech research into the AI specific victory conditions(other than space but thats general). I think that may be giving 2k games, and Firaxis too much credit.
The AI of course will research military techs first, and it make sense too.
WHY?
What would happen in CiV when i don't keep up with my military tech? I get DoW by the conquest AI.
What would happen in CiV when i don't build an army? I get DoW by an AI.
Now don't get me wrong here, or say that i'm contradicting myself. India maybe cultural, but it will still build a small army to defend itself, even at the expense of building something better.

I also think there is another reason for this focus on military techs, is that of element of the human player. If i met a civilization, that was gonna beat me by getting a Cultural victory, i would DoW them in a second. But lets say that civ focus on the growth and research tract instead of military.
How much of a defense will they put up with their spearman against my riflemen?
How well would they survive in the game against the AI going for conquest?
The answer to these question is that they wouldn't do well at all. In fact, they would do so bad we would be complaining about this instead of the lack of challenge, and it would be joke. We would say: why we would never worry about the cultural Ai, because he is a push over.

The reason why i picked cultural is that they would have unique tech tract, than anyother in the game. And in my personal opinion, i think Cultural would be the hardest for the computer to win.

 

[TheMeInTeam]

Are there really people saying that the AI can't win another way because of the game rules ?

How come this 30+ city civ, ALONE on its continent, with double my income, land, tech and 5 times my army cannot win diplomatically or space ship and let me win culturally ? There's no excuse for this. I don't know how anyone could blame the game rules and not the AI developpers on this -_-.

There's plenty of blame for both aspects; non-military has AFAIK never been viable in most MP settings. The reason is that there has never been true balance between VC. Actually, shoddy AI is central to the series also...I'd argue V's is better than others, especially given the demands placed on it.

But man does it still have a long way to go.

 

[LegionSteve]

From what I've seen, I think testdummy653's theory is correct. It is also true that, while this probably happens to some extent on all maps and all levels, the best way to see it is on deity archipelago, playing peacefully.

I tried such a game just to see what would happen, and Iroquois won a space victory in 335 turns. Why? Probably because he chose space as his objective at the start and was protected from all the warmongers by ocean. His only close neighbour was me, and I had no military except my exploring caravel for most of the game.

There were other Civs bigger than him and just as advanced, but they never got near any kind of victory. It looks like once they conquered all their neighbours and were trapped by ocean, they remained locked into conquest mode and unable to do anything else.

 

[testdummy653]

From what I've seen, I think testdummy653's theory is correct. It is also true that, while this probably happens to some extent on all maps and all levels, the best way to see it is on deity archipelago, playing peacefully.

I tried such a game just to see what would happen, and Iroquois won a space victory in 335 turns. Why? Probably because he chose space as his objective at the start and was protected from all the warmongers by ocean. His only close neighbour was me, and I had no military except my exploring caravel for most of the game.

There were other Civs bigger than him and just as advanced, but they never got near any kind of victory. It looks like once they conquered all their neighbours and were trapped by ocean, they remained locked into conquest mode and unable to do anything else.

Did the Iroquois declare war on anyone? Did anyone declare war on them?

 

[JBConquests]

I agree with the OP.

How can you possibly play this game without war with everyone? If I create a bunch of units purely for defensive purposes, I get accused by all my neighbors of putting an army on their borders. Within a few turns they go from liking me to hating me. So they build up a bunch of units on their borders. Before long we are at war.

If I don't create defensive units, then a civ whose victory goal is domination will take a chunk out of my empire. Other civ's will start accusing me of having a weak and pathetic military and will start hating me. Either way, every civ will hate you and the game turns into death match.

If I found a city anywhere near another civ they will hate me though they will do that to me. If I complain about it they will hate me.

Regarding the ocean invasion issue, I did see Washington make an attempt at sending units across an ocean to invade me in my current game. He supposedly has enough units to "wipe me off the face of the planet" - which I believe scouting his empire. So he declares war and shows up with 3-5 missile cruisers and 5-8 Mechanized infantry and rocket artillery. (No tanks or modern armor? I should look and see why because he has more tech than I and I have modern armor. He has a big enough land mass he should have the resources)

He did take my city since my army was a long ways a way at the time fighting someone else. My army took the city back but 1/2 of those initial units of his sat out in the ocean and did nothing until my rocket artillery took them out. He never did reinforce his position on my continent but was content to let me eliminate his units in my territory - even though on his island he had 3x the units I had.

 

[LegionSteve]

Did the Iroquois declare war on anyone? Did anyone declare war on them?

I think he was involved in war declarations. I seem to remember him asking me to join in wars, which I politely refused. This was a huge 19 civ game though so I was getting war requests and PoS requests every other turn. No actual fighting happened on our islands though. He didn't want to cross the water to fight, and neither did the other AIs. All the other AIs that shared a landmass or close group of islands were fighting each other the whole time, or at least until there was only one left in that area.

 

[Clement]

From what I've seen, I think testdummy653's theory is correct. It is also true that, while this probably happens to some extent on all maps and all levels, the best way to see it is on deity archipelago, playing peacefully.

I tried such a game just to see what would happen, and Iroquois won a space victory in 335 turns. Why? Probably because he chose space as his objective at the start and was protected from all the warmongers by ocean. His only close neighbour was me, and I had no military except my exploring caravel for most of the game.

There were other Civs bigger than him and just as advanced, but they never got near any kind of victory. It looks like once they conquered all their neighbours and were trapped by ocean, they remained locked into conquest mode and unable to do anything else.

Funny you should mention that, a very similar scenario happened in one of my games, Hiawatha had an island to himself and won a fairly comfortable space victory, i can't remember what turn unfortunately.

 

[testdummy653]

I agree with the OP.

How can you possibly play this game without war with everyone? If I create a bunch of units purely for defensive purposes, I get accused by all my neighbors of putting an army on their borders. Within a few turns they go from liking me to hating me. So they build up a bunch of units on their borders. Before long we are at war.

If I don't create defensive units, then a civ whose victory goal is domination will take a chunk out of my empire. Other civ's will start accusing me of having a weak and pathetic military and will start hating me. Either way, every civ will hate you and the game turns into death match.

If I found a city anywhere near another civ they will hate me though they will do that to me. If I complain about it they will hate me.

Regarding the ocean invasion issue, I did see Washington make an attempt at sending units across an ocean to invade me in my current game. He supposedly has enough units to "wipe me off the face of the planet" - which I believe scouting his empire. So he declares war and shows up with 3-5 missile cruisers and 5-8 Mechanized infantry and rocket artillery. (No tanks or modern armor? I should look and see why because he has more tech than I and I have modern armor. He has a big enough land mass he should have the resources)

He did take my city since my army was a long ways a way at the time fighting someone else. My army took the city back but 1/2 of those initial units of his sat out in the ocean and did nothing until my rocket artillery took them out. He never did reinforce his position on my continent but was content to let me eliminate his units in my territory - even though on his island he had 3x the units I had.

From what I'm told the AI can and will build unit, such as modern armor, that require resources, but since it can't spam them(due to resource limitations) your more likely to see mech infantry...

 

[JanusTalaiini]

Random Personalities.
But yeah.. I just wish the AI's did NOT have omniscient future planning minds, didnt view the game as a GAME, but rather did its part in suspending our disbelief (which is the developers part of the deal, our part is to immerse ourselves into the world and pretend its real. Sid Meier himself spoke about this for an hour at the GDC keynote, yet it seems the civ5 dev team completely disregarded it all).
Instead of, predicting gunpowder, every optimal path to a specific victory, and declaring war on everyone around in order to make the next 1000 years fit their conquest objective.

I wish they just simulated a nation, played turn per turn, enjoying the journey rather than being obsessed about a score at the end of time. Like i play. Maybe thats why i suck and rarely win in Civ4, but at least i have a hell of alot of fun and immersion, playing to survive through the ages rather than predict every single tiny variable and strategically plan thousands of years into the future, gunpowder, nukes and beyond.

If i want to do that, if i want to play a game with a build order and tech order for optimal conquest, i'll play Starcraft 2.
I never played civ4 to play like that, nor did i buy civ5 to play like that.
But that is what civ5 appears to be, a game of less diplomacy, more RUSH RUSH WIN WIN SCORE SCORE, much inspiration from Civ Revolutions, it seems.
Which is a shame, i wish i'd known the game was like that before, i could just have kept playing SC2 instead and saved 60 bucks.

Yup, couldn't have put it better myself. I keep trying to like this game. I've probably played about 40 hours or so, by this point. But the lack of immersion kills it for me every game I try.

I booted up Civ 4 yesterday, and played for about five hours. Not sure I'll be back to 5 anytime soon. :-\