Armies

[Ahriman]

On the first point, I always have at least two Patronage policies.

If people think that Patronage policies are always worth getting, over other trees, then the Patronage tree is too powerful and needs to be nerfed.

So while I may well be okay with a CS being allied to another civ, my overarching goal is to have that CS serve my purposes. More often than not, that's best achieved by allying with it (as long as I have the gold and SP's).

I think if you can easily ally all or most of the city states without a massive economic focus and advantage, then city states don't cost enough gold, or AI players aren't trying hard enough to contest them.

Just because you'd like to have all the CS allied to you doesn't mean that anything where that isn't the case is just a "transitory state", anymore than the fact that you'd eventually like to have every other player's capital means that any point in the game where you don't control all the other capitals is just a transitory state.

For most of the game, most of the CS should not be in alliance with you, and you should be ok with that, since you have other priorities.
It should not be a feasible strategy to be able to conquer most or all CSs that are not your allies.

 

[Ahriman]

I'm simply saying all three options should be equally valuable: ignore CSs, befriend them, conquer them

Here's the thing: they can be equally valuable in terms of *winning the game*, or preventing the AIs from winning the game.

But a big part of conquering a CS is that it weakens the AI player who was allied with it. So the purely private economy gains to you from conquering a CS should be much less than the gains from allying it. Otherwise, conquering it is always the better solution, because it helps you while hurting the other players.

So it is not appropriate to have the private benefits from CS conquest to be just as cost-effective as allying them.

So I still see no need for particular rewards to encourage you to conquer CSs. Most CSs should still exist in the late-game, unless the human or an AI have pursued a deliberate policy to capture them, and other players have decided to let them do so.

 

[rhammer640]

I completely agree with ahriman here. I love all of the changes so far that have increased choices, but i dont think this creates more choice and complexity, i think it reduces it.

city states were put into the game to add diplomatic complexity and strategy. Beyond the points above that ahriman says much better than i could, city states play a lot into diplomacy as is.

The added choice you are trying to add is an either or choice (ally or conquer), but it would be better to have more choices than that which is what there are now.
1. ally with the city state.
-pros: resources and type bonus
-cons: might cost a lot if another ai wants that alliance as well
2. leave them alone. (should be the case with most city states besides those few you want)
-pros: no cost to you, no diplo penalty for going after same ai
-cons: no city state benifit
3. conquer them:
-pros: other ai cant get benefits from them, usually good city location, dont have to pay for resources
-cons: no ally benefits, diplomatic hit from ai near that city state for taking away a potential ally to them

If city states are good to ally with then that means they are equally good for the ai so taking that away from your enemy could be a great strategy alone before going to war.

Basically im saying that the way they are now has enough choice and shouldn't get such drastic change as has been discussed.

PS ive been using your balance mods for a while and following the messages boards daily and want to thank you for all the hard work you put into this.

edit: not sure if this is even relevant to discussion as like 10 post happened from when i started typing until i posted, lol.

 

[Txurce]

If people think that Patronage policies are always worth getting, over other trees, then the Patronage tree is too powerful and needs to be nerfed.

I think if you can easily ally all or most of the city states without a massive economic focus and advantage, then city states don't cost enough gold, or AI players aren't trying hard enough to contest them.

Just because you'd like to have all the CS allied to you doesn't mean that anything where that isn't the case is just a "transitory state", anymore than the fact that you'd eventually like to have every other player's capital means that any point in the game where you don't control all the other capitals is just a transitory state.

For most of the game, most of the CS should not be in alliance with you, and you should be ok with that, since you have other priorities.
It should not be a feasible strategy to be able to conquer most or all CSs that are not your allies.

1. I prefer Patronage to later Liberty, and Rationalism isn't available for a while. But I suspect preferences vary.

2. As mentioned earlier, I rarely succeed in allying with all CS.

3. If I am playing for conquest, every capital I don't have is in an interim or transitory state - and this is roughly the point I was making about CS's theoretical interim status.

4. Agreed.

 

[Ahriman]

Let me try to explain why I'm worried.
I'm worried that it will be optimal for the human player to always:
a) Ally 2-3 CS; Maritime if they're a large empire, cultural or military if they're small.
b) Conquer any CS that they don't ally with.
Together, these deprive the AI of any possibility of having CS of their own.

 

[Txurce]

I completely agree with ahriman here. I love all of the changes so far that have increased choices, but i dont think this creates more choice and complexity, i think it reduces it.

city states were put into the game to add diplomatic complexity and strategy. Beyond the points above that ahriman says much better than i could, city states play a lot into diplomacy as is.

The added choice you are trying to add is an either or choice (ally or conquer), but it would be better to have more choices than that which is what there are now.
1. ally with the city state.
-pros: resources and type bonus
-cons: might cost a lot if another ai wants that alliance as well
2. leave them alone. (should be the case with most city states besides those few you want)
-pros: no cost to you, no diplo penalty for going after same ai
-cons: no city state benifit
3. conquer them:
-pros: other ai cant get benefits from them, usually good city location, dont have to pay for resources
-cons: no ally benefits, diplomatic hit from ai near that city state for taking away a potential ally to them

If city states are good to ally with then that means they are equally good for the ai so taking that away from your enemy could be a great strategy alone before going to war.

Basically im saying that the way they are now has enough choice and shouldn't get such drastic change as has been discussed.

PS ive been using your balance mods for a while and following the messages boards daily and want to thank you for all the hard work you put into this.

Based on your choice definitions, it makes sense to leave a CS alone when you don't have the gold to ally with them (con to #1) and don't want to suffer the very real cons to #3. That doesn't make #2 much of a "choice," in my opinion - again, just the de facto state when something prevents me from choosing #1 or #3. But this may just be semantics.

Thal is proposing making changes that buff #3, because it is definitely not being chosen anywhere near as much as #1 or #2. I don't think you need to worry that #2 is going to disappear as a result. Circumstances will almost always make it the most popular "choice" (or non-choice, the way I define it).

 

[Txurce]

Let me try to explain why I'm worried.
I'm worried that it will be optimal for the human player to always:
a) Ally 2-3 CS; Maritime if they're a large empire, cultural or military if they're small.
b) Conquer any CS that they don't ally with.
Together, these deprive the AI of any possibility of having CS of their own.

Thank you for figuring out how to shift the focus!

I'd worry about that, too. The problem is (b), right? And being realistic, it would be the CS that are within reach. Should the conquest buffs then keep likely potential annexation along the lines of desirable and realistic maritime alliances - to 2 or 3?

Like rhammer, I like keeping the CS in the game for their flavor. There's a reason why CS aren't included in conquest victory conditions, but have their own (diplomatic). So while I think there are a lot of circumstantial factors that stop a player from allying or conquering a CS, I agree that we should be careful in not going too far with a conquest buff. At this point I think it's safe to say that no one's being tempted to conquer them all, so wouldn't a modest buff be pretty safe?

 

[rhammer640]

Based on your choice definitions, it makes sense to leave a CS alone when you don't have the gold to ally with them (con to #1) and don't want to suffer the very real cons to #3. That doesn't make #2 much of a "choice," in my opinion - again, just the de facto state when something prevents me from choosing #1 or #3. But this may just be semantics.

Thal is proposing making changes that buff #3, because it is definitely not being chosen anywhere near as much as #1 or #2. I don't think you need to worry that #2 is going to disappear as a result. Circumstances will almost always make it the most popular "choice" (or non-choice, the way I define it).

well to be honest i use the cs diplomacy mod for dealing with cs's so it wont effect my games, i was more commenting based off of realism and that I just think #3 is already strong enough by taking away bonuses to the ai, but maybe that its not because the ai isn't prioritizing them enough.

 

[Ahriman]

I'd worry about that, too. The problem is (b), right?

Right.

Should the conquest buffs then keep likely potential annexation along the lines of desirable and realistic maritime alliances - to 2 or 3?

I don't see how you'd accomplish this without a very artificial seeming mechanic (you get a bonus for conquering the first 2 city states, but not beyond that? Why not?)

[Though on reflection, there's already a mechanic that does something like this; the diplomacy and CS warmonger hate isn't so bad for taking one CS, but if you capture 2 or 3 or 4, then you're depriving yourself of the ability to ally other CSs, because they start to hate you).]

But like Rhammer, I tend to think that the benefits from conquering a CS are large enough already without any bonuses, because they weaken the other players, and they give you an extra city in a good location with resources and luxuries that usually isn't defended by a large army, so they're easier to capture than an equivalent city owned by another AI player.
I don't like the notion that you have to be bribed with a mechanic to get you conquer them, to compensate for *not* allying with them. There's already a reason to not ally them: you don't have the gold, or you'd prefer to spend the gold on other things.

 

[Seek]

At this point I think it's safe to say that no one's being tempted to conquer them all, so wouldn't a modest buff be pretty safe?

This is the key point behind the changes, Ahriman, and I would add that it's true even if there is a quest to do so.

Though I think we all agree that the benefits shouldn't be so powerful that it'll always be the optimal play, the benefits should be enough to make taking over a CS an option to consider more often. Additionally it is a fun mechanic, imo!

 

[Thalassicus]

I think there's some misunderstanding over one reason behind choosing trait-specific bonuses, which is what I was trying to explain here:

I'm running a 200-turn autoplay game to see if citystate traits affect flavor values, it should be done in half an hour or so. I don't see anything about trait-based flavor in the xml files, but it still doesn't hurt to check with an autoplay if there's something in the C++ that might result in this. The thing is, to my knowledge it's reversed - capturing major civs gives the obvious flavor-based bonuses while capturing a citystate does not, since citystates have more limited flavor priorities.

(To avoid confusion, "flavor" is the term used by the civilization series to refer to AI priorities, in this case constructing buildings.)

The autoplay has completed, and I don't see any bonuses for one trait or another. To put it simply, this is how cities work for major civs:

1. AI chooses strategy based on personality.
2. Strategy determines flavor settings.
3. Flavor settings determine buildings built.
4. Buildings built determine what's in the city when captured.

I don't see a way to get around the way steps 1 and 4 are connected, and citystates have significantly restricted options for step 1.

Based on the results of this autoplay, searches in XML files, and playtesting experience, my conclusion is this theory of how the mod works is inaccurate:

[why should] conquering a CS give me some extra bonus when conquering an enemy city doesn't?

 

[Ahriman]

At this point I think it's safe to say that no one's being tempted to conquer them all, so wouldn't a modest buff be pretty safe?

I tend to think: no-one is being tempted to conquer them all, which is good, so there's no problem.
I definitely conquer some some of the time in vanilla and in this mod, if they're allied with an enemy who is at war with me, or who is trying to get a diplomatic victory.

In terms of the mod: I didn't know that there was a bonus in the current version, so I don't think people have really *tried* conquering them all.
A flat bonus like +3 pop has the potentially to be really unbalancing in the early game. Imagine a strat that goes: warrior, warrior, warrior, library, national college. While the 4 warriors attack a MCS, steal a worker, capture it for a +3 pop boost in the capital in the early game.

 

[Ahriman]

@Thal, I still don't understand wth you're talking about with flavor values.

Flavor values just encourage the AI to build one thing rather than another. It might encourage one AI to focus on gold buildnigs, another to focus on military buildings, another to focus on unit production.

It has no impact on the costs or benefits of any of those buildings once constructed.
So the flavor values affect *which* buildings get constructed in a city. They don't have any impact on a city that already has buildings X, Y and Z constructed.
And they don't give any production bonus that somehow means an AI city will have more stuff in it than a CS city.

So as far as I understand it, with your changes:

Situation A.
Conquer a size 10 city from another player that has buildings XYZ.
No extra bonuses.

Situation B.
Conquer a size 10 City State that has buildings XYZ.
Also get some additional, arbitrary bonus.

I have no idea what "flavor-based bonuses" you're talking about.
Are you saying that an AI city is more likely to have more structures than a CS city, or that a CS won't build any structures? I'm not sure that's true.
Just because a CS doesn't have flavor values that encourage it to be relatively more likely to build, say, barracks vs granary, or granary vs barracks, doesn't mean that the CS won't still construct buildings.

 

[Thalassicus]

If it's possible to rush with a few warriors for an easy game that's a separate issue. This was something I saw in many threads about how to easily win on deity difficulty... it doesn't seem logical for such a simple, low-skill strategy like a warrior rush to be so effective. This is why I increased defensive units AIs start with on the hardest 3 difficulty settings.

However it didn't seem to affect citystates, which is a good point. I could add some defenders manually.

To put it simply, this is how cities work for major civs:

1. AI chooses strategy based on personality.
2. Strategy determines flavor settings.
3. Flavor settings determine buildings built.
4. Buildings built determine what's in the city when captured.

@Thal, I still don't understand what you're talking about with flavor values.

Flavor values just encourage the AI to build one thing rather than another. It might encourage one AI to focus on gold buildnigs, another to focus on military buildings, another to focus on unit production.

...
Situation A.
Conquer a size 10 city from another player that has buildings XYZ.
No extra bonuses.

I think we're just using different semantics of the word "bonus." Bonus is "something given or paid over and above what is due." I consider buildings a bonus when capturing a city because the odds of their destruction are randomized, so getting lots of buildings is a bonus.

 

[Ahriman]

However It didn't seem to affect citystates, which is a good point.

I also don't think CS get the faster healing from a capital.

 

[Thalassicus]

To put it simply, this is how cities work for major civs:

1. AI chooses strategy based on personality.
2. Strategy determines flavor settings.
3. Flavor settings determine buildings built.
4. Buildings built determine what's in the city when captured.

@Thal, I still don't understand what you're talking about with flavor values.

Flavor values just encourage the AI to build one thing rather than another. It might encourage one AI to focus on gold buildnigs, another to focus on military buildings, another to focus on unit production.

...
Situation A.
Conquer a size 10 city from another player that has buildings XYZ.
No extra bonuses.

I think we're just using different semantics of the word bonus. Bonus is "something given or paid over and above what is due."

I consider buildings a bonus when capturing a city because the odds of their destruction are randomized (and we have no control over what our opponent built), so getting lots of nice buildings in addition to the city itself is an added bonus. Since step 4 follows as a direct result of step 1, capturing a major-civ city in vanilla CiV gives personality-based bonuses, while capturing citystates does not, which is the opposite of:

[why should] conquering a CS give me some extra bonus when conquering an enemy city doesn't?

Basically, the bonus is built in to the whole flavor and capture mechanic for major civs. For citystates, I could balance this inequity by creating buildings in them when captured, and that's an option I considered. I think an explicit bonus is more transparent though. It doesn't make sense for a cultural CS's to have low culture and border expansion, then suddenly a lot when sacked an pillaged.

On the other hand, looting lots of cultural artifacts from cultural CS to boost the central empire's culture does make sense.

 

[Ahriman]

capturing a major civ city in vanilla CiV gives personality-based bonuses

Can you explain what you're talking about here?
There are no personality-based bonuses that I am aware of. A high gold flavor that makes an AI build more markets makes them build less of other buildings.

A high military unit flavor makes that AI build more military units and fewer buildings.

I am not aware of any mechanic that makes AI players across the board construct more buildings than CSs do.

 

[Seek]

Can you explain what you're talking about here?
There are no personality-based bonuses that I am aware of. A high gold flavor that makes an AI build more markets makes them build less of other buildings.

A high military unit flavor makes that AI build more military units and fewer buildings.

I am not aware of any mechanic that makes AI players across the board construct more buildings than CSs do.

I think what Thal is trying to explain is that since CS operate under different rules than major civs the buildings that will survive from capture will be much fewer and/or of less value because they aren't programmed to build them.

 

[Ahriman]

I think what Thal is trying to explain is that since CS operate under different rules than major civs the buildings that will survive from capture will be much fewer and/or of less value because they aren't programmed to build them.

But I don't think thats true! Flavor values affect the type of building constructed, they don't increase the overall number of buildings.

Do some testing to check. I don't believe that AI cities end up with more buildings than CS cities.
AI cities build military units, are more likely to get into wars and lose those units and replace them, build settlers, build lots of workers, etc.
Many AI cities aren't founded until late; CS cities are founded at turn 1.

If testing shows that this *is* true (ie AI cities systematically construct more buildings than CS cities, even for the aggressive/conqueror AI types), then I'll withdraw my complaint; if this were the case, it would be fine to have some bonus to counteract that.

But the simple fact that AIs have flavor values and CSs don't isn't enough to make it true.

 

[Thalassicus]

Seek gets what I'm trying to say: citystates work differently than major civs. In vanilla we receive no bonus based on CS personality type (maritime, cultural, militaristic) when captured. Capturing major civ cities gives bonuses based on personality. This is the inequity I'm talking about.

In addition, as they point out there's a community consensus capturing citystates is not a viable strategy in vanilla. Buffing capture with personality-based bonuses is a logical solution.

There are no personality-based bonuses that I am aware of.

Well... this is why why I mentioned I think we're just getting bogged down in semantics. Here's my train of thought... and yes, I'm a programmer so I sometimes think in terms of algorithms... I hope this isn't just more confusing! I try and normally explain things in less detail, but maybe explaining differently will help.


(1)
Major-civ AI personality determines strategy.
Strategy determines flavors.
Flavors determine buildings.
Buildings are what's in a city when captured.
----------------------------------
∴ Personality determines buildings when captured.


(2)
A bonus is an unexpected extra (by definition).
Randomness is something unexpected (by definition).
Extra buildings randomly survive city capture.
----------------------------------
∴ Extra buildings when captured are a bonus.


(3)
1. Personality determines buildings when captured.
2. Extra buildings when captured are a bonus.
----------------------------------
∴ Personality determines bonuses when captured.


(4)
Culture-focused major civ cities build cultural buildings.
Cultural-focused CSs do not.
----------------------------------
∴ Capturing a culture-focused major-civ city gives a personality-dependent bonus, while capturing a cultural CS does not.


(5)
Community consensus capturing CSs is not a viable strategy.
Capturing a major civ city gives a personality-based bonus, but capturing a CS does not.
----------------------------------
∴ Conclusion: A personality-based CS capture bonus solves both problems with one clear solution.


The other solution I see to the final conclusion is creating buildings in captured cities, but as I mentioned I prefer the more transparent solution.