Feels Like a Major Problem With the Stationing System is the AI Never Constructs any Military Buildings

[InsidiousMage]

Admittedly, this is a random sampling not quite 200 years from the start of an 867 game but one thing that struck me is the fact that the AI rarely constructs military buildings in their holdings. In a sample of 14 holdings across nine rulers, I found six military buildings out of a total of 56 building slots, including duchy building slots. The AI always builds a fortification building and then mostly fills the rest of the slots with economic buildings. Of those counties with a military building, only one MAA regiment was stationed in a holding where it could benefit from the military building, two if you count the duchy building, which was a Royal Armory that gives a bonus to all MAAs. They AI will occasionally get some MAAs bonuses from their fortification building, but, again, its hit or miss. So, simply by building a single military building, the player gets a massive advantage over the AI without even getting to the fact the AI is very bad stationing its units correctly. I'd be curious to see what the AI's holdings look like is other people's games.

 

[Emperor 1997]

Admittedly, this is a random sampling not quite 200 years from the start of an 867 game but one thing that struck me is the fact that the AI rarely constructs military buildings in their holds. In a sample of 14 holdings across nine rulers, I found six military buildings out of a total of 56 building slots, including duchy building slots. The AI always builds a fortification building and then mostly fills the rest of the slots with economic buildings. Of those counties with a military building, only one MAA regiment was stationed in a holding where it could benefit from the military building, two if you county the duchy building, which was a Royal Armory that gives a bonus to all MAAs. They AI will occasionally get some MAAs bonuses from their fortification building, but, again, its hit or miss. So, simply by building a single military building, the player gets a massive advantage over the AI without even getting to the fact the AI is very bad stationing its units correctly. I'd be curious to see what the AI's holdings look like is other people's games.

The worst Part about this is, that the AI is capable of stationing their Men-at-Arms correctly at a Holding, which would benefit them, but guess what is the Issue here, that the Devs have locked Buildings Slots behind Innovations and that the AI builds at random, if an Player builds the Buildings before giving a Barony to an AI Vassal, the Vassal will mostly place their Men-at-Arms correctly, for an AI at least.

 

[Comrade_Boris]

But building fotifications + only economic building is the optimal way to play....

MAA are already so powerful the boosts from military buildings is pointless, and the extra levies just tank your economy. Building economic buildings helps you support more, expensive (better) MAA. The only boost you need on top of that is your marshal training commanders.

 

[ray243]

I would rather remove the bonus from stationing MAA. It's a mechanic that made no sense narratively speaking, even for the player. Why would stationing troops make them stronger? ideally, your MAA have to be stationed at bases/provinces anyway.

 

[Emperor 1997]

I would rather remove the bonus from stationing MAA. It's a mechanic that made no sense narratively speaking, even for the player. Why would stationing troops make them stronger? ideally, your MAA have to be stationed at bases/provinces anyway.

Yes, but I would really like, that the Devs fix their mess, they have created instead of removing it.

 

[InsidiousMage]

But building fotifications + only economic building is the optimal way to play....

MAA are already so powerful the boosts from military buildings is pointless, and the extra levies just tank your economy. Building economic buildings helps you support more, expensive (better) MAA. The only boost you need on top of that is your marshal training commanders.

It’s not though. At this point in the game I’ve got maxed out MAAs with stationing bonuses while maintaining a small positive gold balance while at war with all levies raised. Even if the AI has maxed out MAAs, which is doubtful, I’m still going to win because mine have +100% to their stats.

 

[ray243]

Yes, but I would really like, that the Devs fix their mess, they have created instead of removing it.

The issue is even then it's somewhat immersion breaking even if they made an AI that can station MAA as well as a player can. It's one of those magic buffs...that doesn't seem to have any real world correlation, and it makes it too "fake" as a result.

 

[Benismann]

Yep, that's why in my mod (now forever plugged in signature) I threw -2000 weights to all other buildings if the province has just 1 free slot and no "proper" military buildings (rax, stables, camps or regimental grounds are what i could as "proper" military buildings). Honestly some of the weights for buildings make no sense. IMO AI should build any "proper" economy buildings first (basically any of the terrain-specific ones since they're also the best ones available 90% of the time). Then the next 2 slots can be spent on military and fortification, and the rest AI can just build whatever idc. That's also basically how every player builds, probably, except players (i would imgine) also dont get fortifications often.
One might say AI should build what it wants to build based on the personality, but considering that AI is stuck with whatever buildings they get since they can't demolish that (and i dont see that being ever a thing too) i dont think AI being permanently crippled coz the character which happened to get build slot innovation being an economist or smth and building hospitals or cattle pastures is good for the game.

More on the building weights - blacksmiths are apparently HI+pikemen building despite having universal bonus. Camps and stables (but NOT barracks) can't be build by AI in hills before warrior lodges/hillside grazing, which are, and dont even try debating me on that, objectively shittier military buildings and should probably have straight up worse score than regular ones. Most military buildings can't be built by AI if the province can have wheat fields/orchards/manors and doesnt have them built yet, which is a fine piece of logic, weird that it only consistently applies for those 3 and not every other terrain specific economic building... Fortifications in capital have a whopping x125 weight, x5 for being a capital with less or equal than 3 free buildings slots (so always, you either have buildings generated on start or it's 867 and you at most have 3 bildings losts), x5 for having less or equal than 2 free buildings slots and another x5 for having 1 or less free buildings slot. Wow.

And another something - AI really should have some massive buff to hiring score of regiments it can actually station in a location with proper buildings.

None of this is hard to implement and none of this is controversial (i hope?). You can do it in an hour or two without focusing too hard.

 

[ShepherdOfCats]

I would rather remove the bonus from stationing MAA. It's a mechanic that made no sense narratively speaking, even for the player. Why would stationing troops make them stronger? ideally, your MAA have to be stationed at bases/provinces anyway.

I agree with this.

It exacerbates the steamroll factor.

If you're holding 5+ baronies yourself, you already have a major advantage. Then to be able to station each one of your MAA causes a huge power gap between yourself and your neighbor.

 

[Septimus the Merciful]

The worst Part about this is, that the AI is capable of stationing their Men-at-Arms correctly at a Holding, which would benefit them, but guess what is the Issue here, that the Devs have locked Buildings Slots behind Innovations and that the AI builds at random, if an Player builds the Buildings before giving a Barony to an AI Vassal, the Vassal will mostly place their Men-at-Arms correctly, for an AI at least.

The problem with this is under the current system, the AI will never be smart enough to:

a) designate certain baronies to strictly be military holdings
b) change economic buildings to military when they inherit or conquer

Even if they did the above, the AI's lack of succession management and short-termist thinking means they are like to lose those military baronies via partition or giving them away to a random count.

In my view, the easiest solution is to apply modifiers to the entire army and remove stacking. For example, if you have lvl 6 barracks in any of your baronies, it applies to all of your heavy infantry units. Having multiple does nothing. With this, at least the logic for the AI would be straightforward. Does the AI have heavy infantry? Yes. Does the AI have any barracks? No. Now the AI will build barracks. I'll admit this is not without drawbacks, but at least it takes away most of the forward-thinking logic the AI needs to min-max.

 

[InsidiousMage]

In my view, the easiest solution is to apply modifiers to the entire army and remove stacking. For example, if you have lvl 6 barracks in any of your baronies, it applies to all of your heavy infantry units. Having multiple does nothing. With this, at least the logic for the AI would be straightforward. Does the AI have heavy infantry? Yes. Does the AI have any barracks? No. Now the AI will build barracks. I'll admit this is not without drawbacks, but at least it takes away most of the forward-thinking logic the AI needs to min-max.

If you want to limit modifier stacking for MAAs I don't think limiting bonuses of the buildings is the way to go because you are still going to end up with a lot of buildings that the AI doesn't get any benefit from while the player gets maximum usage out of theirs. If buildings are going to provide MAA bonuses, they should work like Blacksmiths and Regimental Grounds where they provide bonuses to any type of MAAs that way it only matters that the AI builds them. Specific bonuses can then limited to stuff like traditions and, hopefully eventually, a law system.

 

[Septimus the Merciful]

If you want to limit modifier stacking for MAAs I don't think limiting bonuses of the buildings is the way to go because you are still going to end up with a lot of buildings that the AI doesn't get any benefit from while the player gets maximum usage out of theirs. If buildings are going to provide MAA bonuses, they should work like Blacksmiths and Regimental Grounds where they provide bonuses to any type of MAAs that way it only matters that the AI builds them. Specific bonuses can then limited to stuff like traditions and, hopefully eventually, a law system.

Actually, the logic is much simpler. Let's say the AI inherits 2 baronies. One has lvl 3 barracks and the other has lvl 1 stables. Assuming the AI has not built MAAs yet, they can easily be programmed to build heavy infantry and calvary. This is a simplistic example but you get the gist. Not only that, but it's much more difficult for the player to build space marines when they cannot stack.

Your solution doesn't address the issue as the player's forward thinking will still result in overpowered military baronies due to stacking and the AI's inevitable loss of land via partition.

 

[ray243]

At the end, if it is making MAA and levies a game of player game optimisation challenge, not only is this a game that an AI will never be able to match, it is also a game that is not really realistic so to speak. Just introducing more micro for sake of micro isn't a good approach.

 

[InsidiousMage]

Actually, the logic is much simpler. Let's say the AI inherits 2 baronies. One has lvl 3 barracks and the other has lvl 1 stables. Assuming the AI has not built MAAs yet, they can easily be programmed to build heavy infantry and calvary. This is a simplistic example but you get the gist. Not only that, but it's much more difficult for the player to build space marines when they cannot stack.

What I'm talking about is that the AI is going to inevitably end up with multiple baronies that each have a Barracks, either through conquest or inheritance or whatever, while the player is only ever going to have one Barracks and therefore no wasted building slots. That's the fundamental issue with your system, it still requires forethought that the AI isn't going have.

Your solution doesn't address the issue as the player's forward thinking will still result in overpowered military baronies due to stacking and the AI's inevitable loss of land via partition.

And you solution ignores the fact that if the AI loses their single Barracks then they will lose literally all of the bonuses for their Heavy Infantry. Any system that provides bonuses based on what buildings a ruler has is going to be inherently biased towards the player as long as partition is in the game, the differences in different systems just deal with what kind of bias the players is going to have. That's why I'd move MAA bonuses off of buildings entirely unless they provide general bonuses to all MAAs that the only thing that matters is that the AI has them and doesn't have to worry about matching specific bonuses to specific troop types.

 

[Septimus the Merciful]

What I'm talking about is that the AI is going to inevitably end up with multiple baronies that each have a Barracks, either through conquest or inheritance or whatever, while the player is only ever going to have one Barracks and therefore no wasted building slots. That's the fundamental issue with your system, it still requires forethought that the AI isn't going have.

Once again, the logic for the AI fixing this is much simpler. If the building already exists in their domain they can replace the building with something economic. It doesn't require any forethought at all. There is no stacking, and stationing doesn't matter. I think you aren't understanding what I'm proposing.

Even if the AI builds wasteful buildings, the other desired effect is that it prevents the player from building space marines that surpass the AI, as modifier stacking doesn't exist for military buildings and stationing is irrelevant.

And you solution ignores the fact that if the AI loses their single Barracks then they will lose literally all of the bonuses for their Heavy Infantry. Any system that provides bonuses based on what buildings a ruler has is going to be inherently biased towards the player as long as partition is in the game, the differences in different systems just deal with what kind of bias the players is going to have. That's why I'd move MAA bonuses off of buildings entirely unless they provide general bonuses to all MAAs that the only thing that matters is that the AI has them and doesn't have to worry about matching specific bonuses to specific troop types.

Did you read what I posted?

Even if they did the above, the AI's lack of succession management and short-termist thinking means they are like to lose those military baronies via partition or giving them away to a random count.

I'll admit this is not without drawbacks, but at least it takes away most of the forward-thinking logic the AI needs to min-max.

The drawbacks were addressed. I'm not aiming for a perfect system, just one that is better than the current system.

Providing bonuses to all MAAs doesn't fix the problem at all. The player still creates military baronies and modifiers still stack.

 

[Benismann]

The problem with this is under the current system, the AI will never be smart enough to:

a) designate certain baronies to strictly be military holdings
b) change economic buildings to military when they inherit or conquer

Even if they did the above, the AI's lack of succession management and short-termist thinking means they are like to lose those military baronies via partition or giving them away to a random count.

In my view, the easiest solution is to apply modifiers to the entire army and remove stacking. For example, if you have lvl 6 barracks in any of your baronies, it applies to all of your heavy infantry units. Having multiple does nothing. With this, at least the logic for the AI would be straightforward. Does the AI have heavy infantry? Yes. Does the AI have any barracks? No. Now the AI will build barracks. I'll admit this is not without drawbacks, but at least it takes away most of the forward-thinking logic the AI needs to min-max.

It has been tried and it did not work. In fact it makes it worse, because rn for AI to have 1 decent regiment of say archers, they would need to have just one barony with camps and idk blacksmiths. Before they would need an entire realm dedicated to them for one (and, consequently, all) regiment of archers to be good. That will also only worsen the "1 unit type army" problem which is also not good.

For current system to work, as i already said, AI just has to have big weight up for MAAs that it can actually station (and maybe disband the other ones a bit more often), and that's it.

Also - do you dedicate baronies to military? Then what are you building there? %insert_building% + blacksmiths + ???, there just isnt that many military buildings that actually stack with each other to dedicate any barony for anything

OH wait, "having multiple does nothing"? Then why would the player build any military building past the first one if they can just plop down more economy instead?

 

[Benismann]

If the building already exists in their domain they can replace the building with something economic. It doesn't require any forethought at all

Oh there're probably good reasons why AI with no long term thinking is not allowed to demolish/replace building in ANY of paradox GSGs. What you'll end up with is AI's constantly replacing buildings coz one day it's held by a duke with 2 counties, barracks in each, the other day it's held by a newly landed count, whose rax just got replaced, so they will try to replace something BACK to barracks. And then ruler above gets +1 domain limit from... idk, somewhere, and now it gets to do the fun thing with replacing newly built rax AGAIN. And that's not mentioning those rax might've been lvl 5 or smth so AI is just stack at level 1-2 rax and it sucks.

 

[Septimus the Merciful]

It has been tried and it did not work. In fact it makes it worse, because rn for AI to have 1 decent regiment of say archers, they would need to have just one barony with camps and idk blacksmiths. Before they would need an entire realm dedicated to them for one (and, consequently, all) regiment of archers to be good. That will also only worsen the "1 unit type army" problem which is also not good.

For current system to work, as i already said, AI just has to have big weight up for MAAs that it can actually station (and maybe disband the other ones a bit more often), and that's it.

I'm not sure I understand your first paragraph. Under what I'm proposing, stationing wouldn't matter and there wouldn't be any stacking. So they would need, for example, 1 militia camp for archers across all of their domains.

Also - do you dedicate baronies to military? Then what are you building there? %insert_building% + blacksmiths + ???, there just isnt that many military buildings that actually stack with each other to dedicate any barony for anything

That isn't the case. I dedicate entire baronies to military units all the time. That's how to get MAAs with +200% damage from buildings. For instance, if my barony has forest terrain and I want a "military barony", I'm building militia camps, outposts, blacksmiths, and workshops as opposed to economic buildings. If I'm in hills, warrior lodges, blacksmiths, and barracks. Many times this means

OH wait, "having multiple does nothing"? Then why would the player build any military building past the first one if they can just plop down more economy instead?

They wouldn't, but neither would the AI. The economy can already be gamed as is (once you start spamming windmills/watermills and are past 30 dev on most of your counties, you are printing money). The player trading military buildings won't make a material difference. I'm fine with this when I think more about it, b/c perhaps it would bring mercenaries into play more often than they currently are.

As I said before, the logic would allow for the AI building (and possibly even replacing) military buildings to be straightforward and servely hamper the players ability to build an army of space marines that stack wipes 10x its size.

 

[Benismann]

If I'm in hills, warrior lodges, blacksmiths, and barracks

That's just 3 slots tho, you have 6 in county capitals in high medieval anyways.

The player trading military buildings won't make a material difference

It will earlygame tho. Especially if some get-rich-quick schemes were to be addressed (cough cough the whole stewardship left tree cough).

And also idk man, every county has the potential to be the only county of a character holding it, and i would bet every county does become that at some point, so what we truly will end up with is military buildings that dont do anything everywhere all the time.

 

[Septimus the Merciful]

Oh there're probably good reasons why AI with no long term thinking is not allowed to demolish/replace building in ANY of paradox GSGs. What you'll end up with is AI's constantly replacing buildings coz one day it's held by a duke with 2 counties, barracks in each, the other day it's held by a newly landed count, whose rax just got replaced, so they will try to replace something BACK to barracks. And then ruler above gets +1 domain limit from... idk, somewhere, and now it gets to do the fun thing with replacing newly built rax AGAIN. And that's not mentioning those rax might've been lvl 5 or smth so AI is just stack at level 1-2 rax and it sucks.

I get that, but without stacking or stationing, this should be minimal. We are talking about a maximum of four buildings realm wide. This speaks more to the fact that the player usually holds their lands from start to finish while the AI regularly loses land through succession, wars, and domain limits.