AI empire building cycle and the army cap

[Slaists]

In my campaigns, I tend to leave at least half the map to the AI's in the hope of some major opponent faction developing there. However, this is what I see (before and post patch IV beta): an AI faction expands, stabilizes, then gets destroyed by a minor AI. That formerly minor AI expands, stabilizes, gets destroyed by another minor AI, rinse, repeat. I suspect this is caused by the current army limit system which tends to favor a swarm of one-region AI factions over their larger AI neighbors. The issue is: each one region AI gets 3 armies (or so I think) while a faction needs to hold quite a few regions to be allowed to field 6 armies. Thus, as a successful AI expands [and, given the army cap, over-extends itself], it's armies tend to concentrate where the action is taking place. In the process, the heartlands of that very faction become wide open for any one-province minors marching in from another direction.

Anyone else sees this?

In the previous titles, larger, successful AI's would be able to muster absurd number of armies (for their size), but that would help these AI's to handle minor faction incursions.

[pajomife]

There ir a simple answer for it,they dont test any new feature.

[Monsieur Alphonse]

Apart from being overwhelmed by smaller factions who don't have tot defend all their borders the AI still tends to run out of money. In my latest campaign as Rome I have an alliance with Carthage. They still managed to lose some regions due to rebellions and some were captured by the Measles (that whatever overpowered African faction). The regions that they lost were not proper defended because the AI still wants to send all its armies on a sailing cruise and is not capable to maintain those armies. In the end I helped to destroy the Africans and Carthage was helped by their client states.

[Americanus Supremus]

The army cap limit, I believe put in to stop the early game steam roller effect, has only come back to bite them in the ass. Because just as you put it, if one faction become aggressive and grows (which is a great thing) they cant and are not programmed to leave behind garrison forces on top of the crap garrisons that the game has from buildings. Which in turns leaves there Heartland exposed to every other minor faction in the area that is sitting in its one region starving to death with three full stacks. Then after this happens, its just as you describe, rinse and repeat. I have NEVER seen another Super power in game outside of my spawned civil war... EVER

Rock, Paper, Scissors has more logic to it than this game.

[nameless]

The army cap limit, I believe put in to stop the early game steam roller effect, has only come back to bite them in the ass. Because just as you put it, if one faction become aggressive and grows (which is a great thing) they cant and are not programmed to leave behind garrison forces on top of the crap garrisons that the game has from buildings. Which in turns leaves there Heartland exposed to every other minor faction in the area that is sitting in its one region starving to death with three full stacks. Then after this happens, its just as you describe, rinse and repeat. I have NEVER seen another Super power in game outside of my spawned civil war... EVER

Rock, Paper, Scissors has more logic to it than this game.

The army cap limit is working fine.

As I posted in this forum in patch 4 I've seen Egypt, Seleucid and other factions becoming quite large and still maintaining their empires 80-100 turns in with little problem. Usually Egypt dies like after 10 turns in. In battles I've seen the AI throw full stacks at me wave after wave and in some case they combine two 14/20 armies into one.

The issue as others have pointed out is the squalor/food penalties which CA adjusted to allow this in patch 4.

The only thing left IMO is to boost the AI's income and continue adjusting the AI's reponse to food/squalor which should lead to more improvements.

[Slaists]

Since patch IV beta, I have not seen AI plagued by food shortages. AI food shortages still happen at times, but not everywhere and also, AI seems able to "solve" the problem once it starts.

Seleucids is a nice exception to the problem I mentioned in the original post. AI Seleucids tend to survive because they are surrounded by a swarm of minor vassals who kill any AI that is hostile to Seleucids. Egypt? Hmm, I have yet to see AI Egypt survive.

[nameless]

Since patch IV beta, I have not seen AI plagued by food shortages. AI food shortages still happen at times, but not everywhere and also, AI seems able to "solve" the problem once it starts.

Seleucids is a nice exception to the problem I mentioned in the original post. AI Seleucids tend to survive because they are surrounded by a swarm of minor vassals who kill any AI that is hostile to Seleucids. Egypt? Hmm, I have yet to see AI Egypt survive.

Again it's TW games. Not everyone's campaign will play out the same.

Egypt has control of Libya all the way to Jerusalem but some other minor faction just conquered Jeruasalem. I'll post a screenshot when I can but this is like 80-90 turns in so that's pretty good.

So if CA continues I would suggest an increase to income so when the AI reaches it's cap it can maintain large armies and maybe further adjustments to squalor/food.

[Americanus Supremus]

So Nameless your saying that the ai can be fixed by adding money to it? Runs to do so and comes back to tell you no it doesn't fix it. The AI still does not protect its homeland in most cases, it still will have food shortages depending on its starting region (mostly the capital cities as they have to build a certain temple or delicatessen to get food), and still doesn't allow for massive expansion of major factions. Just so you know adding money can be done many ways in game, through db files and editing startpos files.

In most of my games the Seleucids sit still in there starting regions due to being surrounded by client states and a very poor starting setup. One 4 unit army, no spies, no navy and no buildings (no barracks at all) in their cities other than ports. Not only that it will take it, with proper research and building, at least 10 turns to put out anything other than its base units, though uber, eastern spearmen

Umm the Ptolemies start with Jerusalem I believe so that is no conquest, and I have noticed since patch 2 I think that Cyrenaica is taken out quiet early by them.

Again an increase in money will not change things overall, you give to much money to a faction and it will build the high squalor and high food demand buildings that cause the food shortages and other problems.

[Emperor of The Great Unknown]

too be honest I quite like to see this cyclic empire creation and destruction cycle that the AI seems to replicate, it's realistic, (although over a larger time scale.) I want to see more challenges to the player maintaining larger empires, I think small factions should still be seen in the late game, as in real life it was hard to manage an empire.

[Slaists]

too be honest I quite like to see this cyclic empire creation and destruction cycle that the AI seems to replicate, it's realistic, (although over a larger time scale.) I want to see more challenges to the player maintaining larger empires, I think small factions should still be seen in the late game, as in real life it was hard to manage an empire.

I'd like to see more AI alliance-block building. It seems to work pretty well for Seleucids in the game but for them, it is hard coded from the game start. So, what we need is a way to achieve something similar dynamically. As in: once a nation becomes locally dominant, it forces several neighbors to become vassals, as a consequence securing a flank, which allows that faction to expand in another direction.

As to giving bonuses to the AI: the real cash in RTW 2 is food. The AI needs extra food more than it needs extra cash.