Slitherine

What are the reasons for not conquering cities in Pandora? I understand the reasons for not always building colonists and settling new regions; but taking cities (especially large, well-developed cities with lots of buildings and a large population) seems to me, in my few game so far, to be always well worth it. not just because of your gain, but even more so for how it can completely cripple a faction for the remainder of the game. Considering that it's usually only good to have two or three core cities, the loss of one or two of them is devastating.

So if I (or the AI for that matter), has a chance to declare war and capture a great city (or cities) from an AI faction who almost nobody likes, is there any reason not to? Other than the risk that you may be defeated militarily?

When at war with an AI faction with a diminish army, are there any reasons not to take their last cities? I never find myself making peace with an AI that I'm rolling over. I just keep conquering all his cities until he has none. Is there any reason why I should make peace with him when I know for sure that I can take his last city?

Thanks in advance.

Well, to be honest: There is no reason to not conquer cities in Pandora.

Just as you said it yourself. It's always worth it and even more so, the bigger and better developed the city is.

The only reason I can see to not take everything you can is to not make the others jealous. (a mechanism that does this was introduced with the last beta-patch)

Zak0r also critizised that the AI does not seem to defend their cities well enough, and thus making capturing them too easy.

So my current tests also involve ideas of how to ensure better defended cities.

Only one of those is in the current beta: preferring stronger units over weaker ones.

Not in the beta so far:

More defenders in general, also scaling amount with city-size.
Stop roaming, so unused units are in the cities. (but this is mostly experimental, in my current test-version the AI has a scouting problem as in: it doesn't scout at all and so getting contacts is very much delayed and it gets only very few goody-huts)
A fix to non-functional-upgrading.
And also important: The distance units are willing to walk in order to defend something was increased dramatically.
In current beta you will probably see their capital well defended but other cities much worse. That is because it builds most units in the capital, as it's usually the most developed cities. But then it is "too lazy" to walk the excess units over to the other cities that also need defenders.

The whole point of that is making capturing cities much harder, so the effort in order to do so actually becomes relevant in the consideration if you shall capture cities or not.

Xilmi wrote:The only reason I can see to not take everything you can is to not make the others jealous. (a mechanism that does this was introduced with the last beta-patch)

Because this is the only reason not to take everything you can, I think that this new mechanism needs to be very strong and have an impact as soon as you start taking any. Even if the other AIs hate the faction you're conquering, it's still very much in their best interest to realized the threat of a conqueror who has taken the core cities of other factions. Else they'll never compete. It also makes the decision to conquer another city a diplomatic one: How will the rest of the world re-act to this?

I really like your ideas for superior AI defense. As far as the "staying in" vs "moving out" of cities question goes, while I understand that staying in in many situation may be the wrong decision, I think that overall staying in is the safer and easier choice. and of course defensive buildings should be a top priority always (so far, I've noticed the AI recognizing this quite well).

Maybe there should also be other deterrents to conquering every city you can. Not necessarily a new mechanic, just maybe more conquest migration or something. I also think that war weariness (from troop losses and extended wars) could result in not every war being fought "to the death" -with the winning faction never having any reasons to make peace (although I don't know if the introduction of such a mechanic is possible). I like the way Civ 5 gives incentives for not conquering every last city (even if you can militarily). Maybe having to control all of that new hostile population should not always be worth it. I also think that the destruction of many buildings upon conquest of a city would drastically help make conquering cities more balanced for gameplay (and more realistic). Considering the destructive power of the military tech used in Pandora (as well as the lore in the pop-up message you receive when you first conquer a city), there's no reason to think that the conquered city isn't in ruins from the fighting.

Basically, for the sake of game balance, I feel that conquering cities is too efficient compared to settling new ones. The destruction of buildings upon conquest could go far in balancing this.

Also: I like your idea to program the AI to be far less likely to start a war if they are already at war. Is the AI more likely to go to war with another faction if that faction is already at war with someone else? That would give you more incentive to end a war (even if you're winning and not done taking your opponent's every last city): the increased risk of being attacked while your units are off conquering in enemy territory (possibly over-seas).

also: in my current game, I've noticed that it's far too easy to get another faction to go to war with a faction they just made peace with (or sign a non-aggression pact with). I play with a house-rule which prohibits me from demanding that another faction declares war on a faction that I'm not already at war with. I like this rule because it makes my "demand war on another" request a true call to arms and a cry for assistance, not just a way of exploiting the AI into leaving me alone. but even with this restriction, with "demand war on a third party" I can make the AI remain at war even when it's just made peace or sign a non-aggression.

In real life, I suppose the two main reasons why aggressive countries didn't conquer cities that they knew they were militarily cable of doing so were: the re-action of the rest of the world and the difficulty of governing and maintaining order in the newly conquered areas.

Thanks in advance.

JetJaguar wrote: Because this is the only reason not to take everything you can, I think that this new mechanism needs to be very strong and have an impact as soon as you start taking any. Even if the other AIs hate the faction you're conquering, it's still very much in their best interest to realized the threat of a conqueror who has taken the core cities of other factions. Else they'll never compete. It also makes the decision to conquer another city a diplomatic one: How will the rest of the world re-act to this?

Well, right now the mechanism is only indirectly reacting to city-captures and also is quite moderate.
It only becomes stronger the bigger your advantage gets.

As I mentioned somewhere else before. It is a thin line to walk.

But you just gave me the idea to try something along those lines:

When a city is captured, it could worsen relations others have towards the aggressor by the following formula:
population of the captured city/population the one who lost the city had before * relationship I had with the faction that lost the city * (1 - myAggressiveness)

This means:
If you capture a small town, noone will actually give much about it.
If you capture a big city, it will be a much bigger deal.
If you capture the last remaining city, you will gain a lot of hate for commiting genocide.
The multiplier at the end means: Factions who are aggressors themselves care less about someone conquering cities than factions who are normally peacefull. But maybe it would make even more sense the other way around.

Example: You capture a size 25 city from Terra Salvum, who still have their size 31 capital and a newly found size 2 city.
The Ambassadors liked them quite a bit, let's say 0.9.
Heid was also at war with them and had a relation of only 0.1.

Now your relationship with the ambassadors worsens by:
25/(25+31+2)*0.9*(1-0.15) = 0.33
while the relationship with the Imperium only worsens by:
25/(25+31+2)*0.1*(1-0.65) = 0.015

JetJaguar wrote: I really like your ideas for superior AI defense. As far as the staying in vs moving out of cities question goes, while I understand that staying in in many situation may be the wrong decision, I think that overall staying in is the safer and easier choice. and of course defensive buildings should be a top priority always (so far, I've noticed the AI recognizing this quite well).

Well, staying in sometimes is so bad, that I can't go with the "easier choice". Note: There's alien sieges too. Which means one or even several spitters who damage all units in the city every turn, while other alien units never attack and just give vision to the spitters. I have seen several cases of this in my test games. It is an extremely awful sight, when 3 aliens with a combined strength of 4.5 contain 12 strength worth of troops like that.
So it really should be dynamically adjusted to the strength of the units outside of the city.
When there's an easy solution for a quick big improvement, I'll temporarily use it. But in the long run, my ambition is to actually have the AI come to the same conclusions that I would in each given situation. The challenge of this kind of AI-programming lies within translating the players intuition into algorithms and formulas, that simulate it as precisely as possible.

JetJaguar wrote: Maybe there should also be other deterrents to conquering every city you can. Not necessarily new mechanic, just maybe more conquest migration or something.

Well, in this you have an exactly opposite opinion to some other players, who wanted to have it lowered.

JetJaguar wrote: Basically, for the sake of game balance, I feel that conquering cities is too efficient compared to settling new ones. The destruction of buildings upon conquest could go far in balancing this.

Once again, for a long time it was seen the other way around by many before the migration after conquest was lowered. (It once has been higher than now)
The factions with economic-bonusses also can become very powerful without having to do anything in particular to do so. So for the factions with military-bonusses it is almost a must to capture something in order to compete in the long run.
This has to be kept in mind.
When you consider multiplayer games, it can be seen there, that it's rare that someone easily loses a city to someone else before quite late into the game. The reason is, that it simply is quite a lot of effort to put in, in order to overcome the defenders advantage.
If the effort to conquer AI-cities becomes equally tough, balancing military vs. research and economy automatically becomes a harder decison.
Like if you focus und research and development while others wage war, you migh have less population but will be ahead in science for quite some time.

When a city is captured, it could worsen relations others have towards the aggressor by the following formula:
population of the captured city/population the one who lost the city had before * relationship I had with the faction that lost the city * (1 - myAggressiveness)

This means:
If you capture a small town, noone will actually give much about it.
If you capture a big city, it will be a much bigger deal.
If you capture the last remaining city, you will gain a lot of hate for commiting genocide.
The multiplier at the end means: Factions who are aggressors themselves care less about someone conquering cities than factions who are normally peacefull. But maybe it would make even more sense the other way around.

Example: You capture a size 25 city from Terra Salvum, who still have their size 31 capital and a newly found size 2 city.
The Ambassadors liked them quite a bit, let's say 0.9.
Heid was also at war with them and had a relation of only 0.1.

Now your relationship with the ambassadors worsens by:
25/(25+31+2)*0.9*(1-0.15) = 0.33
while the relationship with the Imperium only worsens by:
25/(25+31+2)*0.1*(1-0.65) = 0.015

I like this very much. I think of each faction's first two/three cities (usually heavily populated and developed) as their "core" territory. The cities they settle after these are more like outpost that are often settled in territory that could have been someone else's as much as theirs. So conquest of these "outposts" in what could be seen as "disputed" territory wouldn't upset the rest of the world nearly as much. and this just happens to be really good for balance.

Well, in this you have an exactly opposite opinion to some other players, who wanted to have it lowered

They felt that conquering cities was underpowered? but the more I think about it, I think that buildings being destroyed upon conquest is the best way to go here; like in Civ. I also like the random element of not knowing which buildings will be destroyed, as conquest should be a little messy and unpredictable; much more than colonization.

The factions with economic-bonusses also can become very powerful without having to do anything in particular to do so. So for the factions with military-bonusses it is almost a must to capture something in order to compete in the long run.
This has to be kept in mind.

What makes conquering a well developed, high population city so powerful, in my opinion, has more to do with what the other faction has lost than what the conqueror has gained. As you said, having two (maybe three) core cities is usually the best way to play. If a faction has two core cities and one of them is conquered, it can be extremely difficult to ever recover (after all, they just lost half their land). I read in one of your posts that you had recovered from something like that a while back, but that was before your improved AI patch was released; so doing so now would be far more difficult.

I've been playing with 7 factions on medium sized, continents maps and at least one faction always seems to get eliminated very early. often others never really get the chance to develop before they're conquered. I've thought of switching to playing on the "large" map size, but there are other down-sides to that. I might try a modded, hybrid medium/large map. I'm also wondering about adding something to the difficulty levels which could specifically make early, opportunistic conquests by the player more difficult. but I'm not exactly sure how. By the way, I've been playing on a hybrid, medium/hard difficulty level (AI bonuses are exactly half-way between the two).

To the topic of conquered cities being too valuable:

At first you had not a very strong negative impact from conquering, only temporary morale maluses. Then the conquest migration was introduced to simulate fleeing citizens and make conquest a little less strong. However, the migration was so strong that firstly players argued that conquest now almost doesn't make any sense if you don't capture a lot of buildings as well because most of the population captured was fleeing anyway. And secondly the strong migration could be turned off by simply razing the city. Due to low morale a newly conquered city won't produce much anything so it was a very easy thing to do. Then the conquest migration was adjusted to the current levels and I would argue that it is close to ideal. Why? Because it still hurts but using the raze button doesn't bring any benefit in most of the cases and if it does (in huge cities) it only does for a few turns.

Zak0r wrote:using the raze button doesn't bring any benefit in most of the cases and if it does (in huge cities) it only does for a few turns.

Also: Since the AI now also uses the raze-button, it at least cannot be considered an unfair exploit. ^^