AI choosing historically fitting Social Policies

Haig

Deity
Joined
Aug 3, 2010
Messages
3,172
Location
Finland
I've noticed in later era starts that certain Civs choose some particular SPs that suit them historically.

Civ V Expansion is gonna make Freedom, Order and Autocracy to matter in diplomacy during later eras, and seems to me that already AI Civs prefer specific ones.

I tried for three times to start a Industrial Era start with same Civs anc chechk out which SPs were picked, and Japan and Germany were really fond of Autocracy, Russia chose Order and England Freedom (and Order once!).

Could it have been a coincidence or AI flavors, not sure..!
 
Pretty sure Washington/Napoleon always go Liberty and Freedom. They definitely have a set flavor and some also may be forced to take policy X. Just like the devs also implemented certain Civs to absolutely wanting to build wonder X (mainly those that they had in RL) - Korea does want to build the PT more than anything else. Washington the Statue of Liberty, Catherine the Kremlin, Napoleon Notre Dame (and I think the Eiffel Tower and the Louvre as well. Probably Egypt also Pyramids (but maybe not GLighthouse (?) and GLibrary (Alexandria, wasn't it?) as that would be too much wonder focus early game.
 
Pretty sure Washington/Napoleon always go Liberty and Freedom. They definitely have a set flavor and some also may be forced to take policy X. Just like the devs also implemented certain Civs to absolutely wanting to build wonder X (mainly those that they had in RL) - Korea does want to build the PT more than anything else. Washington the Statue of Liberty, Catherine the Kremlin, Napoleon Notre Dame (and I think the Eiffel Tower and the Louvre as well. Probably Egypt also Pyramids (but maybe not GLighthouse (?) and GLibrary (Alexandria, wasn't it?) as that would be too much wonder focus early game.

Whoa, that was new info for me! I think that's very cool, and so is the Social Policy choosing.

As in Gods&Kings the ideological policies come to play, we can get for example Central powers (Germany, Ottomans) or Axis (Germany, Rome/Italy, Japan) from real history. :cool:
 
Will we see a return of bribing civs to switch social policies, like civics in IV?

I think an effect of the late game diplomacy is that warmonger autocracy civs will still be hostile with everyone and each other, due to negative diplo hits from DoWs and exterminations. But also it may strengthen the bonds between more peaceful freedom types that form a sort of coalition, especially if they share a religion. It's all speculation until we know just how much of an effect the changes have on diplomacy.
 
Will we see a return of bribing civs to switch social policies, like civics in IV?

I think an effect of the late game diplomacy is that warmonger autocracy civs will still be hostile with everyone and each other, due to negative diplo hits from DoWs and exterminations. But also it may strengthen the bonds between more peaceful freedom types that form a sort of coalition, especially if they share a religion. It's all speculation until we know just how much of an effect the changes have on diplomacy.

My problem is order and autocracy come so late that you usally take freedom


And if religion start not to matter in the renaissance you have to blaze to the industrial just to get the right policy doesn't seem right
 
I just don't understand how they're going to make the ideological rivalries work since there is no bonus to be gained by sharing the same civics as other civs.
 
I just don't understand how they're going to make the ideological rivalries work since there is no bonus to be gained by sharing the same civics as other civs.

Well for AI it's coded in, Washington with Freedom sees the growing Autocracy in Germany and Rome, and gathers allies against the evildoers.
So player might choose Freedom to ensure he's at the same side as Washington etc.

Or maybe to ensure good trading positions with Germany and Rome, a player picks up Autocracy.

There's not yet much info about how things work out, but I'd say that trading, alliances etc. will affect from those. I think that's a bonus in itself. :)

Also it's cool that AI "roleplays", gives nice flavor into things.
 
This is kind of a month old necro, but I'd like to share with ya'll a game I had as Korea on Emperor, map Pangea.

Its the 1900s, Korea is lagging behind the other powers that remain, that means America, Rome... and tiny Greece sandwiched between the two. It is also communist, like its socialist brother Rome. Against that we have FREEDOM America, and... horribly backwards Greece who might have gone autocracy if it had lasted its last war against the land of red and blue.

One thing that little North Korea did to ensure its Marxian brother to the northwest remained a Struggle Brother instead of whiny backstabby imperialist was to never let its borders open. It was, like our Real Life Counterpart, a walled-off hermit kingdom built on socialistic principles, under the enlightened rule of General Secretary Sejong the Thousand-Year Chairman.

edit: having similar social policies helped too, I guess.
 
I just don't understand how they're going to make the ideological rivalries work since there is no bonus to be gained by sharing the same civics as other civs.

It was only a diplomacy modifier in Civ 4 - there's no reason they can't add a diplomacy modifier for having the same policy branch in Civ 5, or a negative one for having opposed branches. We are, after all, promised other changes to the diplomatic system, including religion being relevant and changes to the diplomatic effects of capturing city-states. Espionage will undoubtedly have new diplo effects too. So I don't think it's productive to assume that everything currently in the game is set in stone and can't be altered to have different repercussions in the expansion.
 
I've noticed in later era starts that certain Civs choose some particular SPs that suit them historically.

Civ V Expansion is gonna make Freedom, Order and Autocracy to matter in diplomacy during later eras, and seems to me that already AI Civs prefer specific ones.

I tried for three times to start a Industrial Era start with same Civs anc chechk out which SPs were picked, and Japan and Germany were really fond of Autocracy, Russia chose Order and England Freedom (and Order once!).

Could it have been a coincidence or AI flavors, not sure..!

I think it is fairly certain they will, or they will be strongly loaded to roll the same SP. This is just from observing the fact that the AI already behave fairly consistently in the same way by Civ type. So it only makes sense they extend this to the post Renaissance era.

It also makes gameplay sense in that it makes long-term planning even more viable if you know Montezuma is bound to end up with a favoured SP. That said, I'd be open to some RNG in there where, depending on the dice roll, and flavour of the Civ, they may pick another SP anywhere from 50% to 10% of the time.

Bottom line: There will be enough Civs, between the core Civs, DLC Civs and the new Civs in the XP to allow for varied play even if Civs have a preferred SP type.

But I'd also welcome some randomness to the behavior once in a while.
 
I'm not sure that civs are designed to pick certain policies as such, so much as they are designed to follow different strategies. Alex is going to go Patronage because he's aiming for a Diplomatic victory. I don't know of any actual social policy bias. Is there one?
 
I would assume that the AI choises are based on their personality or flavour. In every game they start with a value +-2 so they will differ.

If they put alot of work into the AI on this then it should also consider it's terrain, neighbours...

An easy way to test this is to start a few late game starts like OP some with random personality, some without and see if there is a difference.
 
So this explains why Suleiman completed Patronage in my last game without City States ?
 
I'm not sure that civs are designed to pick certain policies as such, so much as they are designed to follow different strategies. Alex is going to go Patronage because he's aiming for a Diplomatic victory. I don't know of any actual social policy bias. Is there one?

I'm thinking in terms of the diplo spread where civs have loyalty/warmonger ratings and flavours towards wonder/domination/un

I can see the AI decision for SP being tied to that. Perhaps with some civs being weighted more heavily towards freedom for example, while others are more autocratic.

This as I noted will fall inline with the existing diplomatic model and will make each individual civ somewhat predictable and smooth out diplomatic planning.
Washington I'd expect would choose Freedom and be very preachy about it :)

That said, I would expect some element of Randomoness to it, as is the case with the existing AI behaviors, where in some games, Washington may go autocracy given the game conditions.

Depending on how in-depth they want to make it, AI's final choice of SP may not only be depending on seeding at game creation but also influenced by events during the game. Perhaps they want to cozy up to someone who already picked an SP and that diplomatic strategy outweighs their flavour seeding.

Either way, I think we can all agree the AI throwing a dice roll on the turn of SP selection and leaving the choice entirely up to that RNG is probably not what users would want to see.
 
Pretty sure Washington/Napoleon always go Liberty and Freedom. They definitely have a set flavor and some also may be forced to take policy X. Just like the devs also implemented certain Civs to absolutely wanting to build wonder X (mainly those that they had in RL) - Korea does want to build the PT more than anything else. Washington the Statue of Liberty, Catherine the Kremlin, Napoleon Notre Dame (and I think the Eiffel Tower and the Louvre as well. Probably Egypt also Pyramids (but maybe not GLighthouse (?) and GLibrary (Alexandria, wasn't it?) as that would be too much wonder focus early game.

And AI Inca always prioritizes Machu Pichu.

Does the Bablylon AI prioritize Hanging Gardens?
 
Didn't play against Babylon too much (or any DLC civs) as the HoF and Gauntlet games are with vanilla civs only :(

Even if they don't have a flavor to building HG, they do have a flavor to research they're unique unit which comes with archery - a pre-requisite for HG/Mathematics. So at the very least they have a little advantage towards reaching the tech earlier.
 
I didn't know civilizations had a bias towards picking wonders and policies based on real history.

This is a bad thing IMO. Going for a wonder or policy should be strategy based. In other words, if Egypt is going for an OCC and already has a worker, it'd be pretty stupid to build the Piramids instead of other, in their case better, wonders just because they build them in reality. Or if Germany picks autocracy while they're on their own continent and there are no civs to be conquered nearby.

Is it not the entire point of civilization that China can be a tiny kingdom, focussed on freedom and proud to have build the statue of liberty while the autocratic America build the statue of Zues and is fighting a war for world conquest?
 
I didn't know civilizations had a bias towards picking wonders and policies based on real history.

This is a bad thing IMO. Going for a wonder or policy should be strategy based. In other words, if Egypt is going for an OCC and already has a worker, it'd be pretty stupid to build the Piramids instead of other, in their case better, wonders just because they build them in reality. Or if Germany picks autocracy while they're on their own continent and there are no civs to be conquered nearby.

Is it not the entire point of civilization that China can be a tiny kingdom, focussed on freedom and proud to have build the statue of liberty while the autocratic America build the statue of Zues and is fighting a war for world conquest?

Current Civ SP bias is based on what will benefit them based on their grand strategy. Builders will go Tradition for growth and wonder bonuses. Warmongers will often go Honor. This is not hard to figure out as AI's SP choices are visible in the diplomacy screens.

It's a rational weighting based on the AI's grand strategy. Even if on some level there is an RNG roll involved, over many playthroughs it's clear that their choices aren't completely random.

I am arguing largely for the fact that this should carry over to the SP systems post patch where there will be diplomatic consequences to picking the Tier2 Social policies.

It shouldn't be random, but should at least be predictable and influenced by various factors. I suggested various factors, including in-game conditions and also historical flavours.
 
Back
Top Bottom