Warmongering calculation in BNW?

[Deggial]

In G&K, declaring war had an impact on your warmonger status. Eliminating a whole civilization (even a city state) was percepted even worse. The AI didn't care about single conquered cities (at least not regarding the "warmongering" status; however, the AI did worry about your increasing number of cities) or rejected peace offerings. Therefore it was a useful tactic to lure the AI into war and then conquer the civ with the exception of the last single city.

In BNW, there are voices that this calculation was changed. It is said, that eliminating a civ isn't a warcrime by itself any more. It is said, that declaring war has a lessened impact (if at all) now and it is only conquering cities that matters. I have read that the penalty depends on the percentual number of cities a civ has when a city is taken and in this respect (and in this respect only) eliminating a civ is calculated. Which is why it i worse to battle and conquer a small civilization with a total number of 4 cities than a large empire with 15 cities and thake 4 cities as well.

My question is now: Is this true? I am afraid, everything I listed above is just from hearsay and without any code based evidence.
My appeal goes to all those code miners here at the civfanatics. I know you are out there and did already analyze the XML up to the last semicolon.
Is there a new warmonger evaluation algorithm?

(As much as I appreciate well ment guesses - they don't help to solve this problem. We already have too much speculation regarding this topic. I would love to sort this problem out for now and for the future without any doubt! )

 

[Raptor Llama]

It's true: http://www.reddit.com/r/civ/comments/1i1q7l/keeping_one_ai_city_alive_no_longer_spares_you/

But they still don't care if you reject a peace offer; they never did.

 

[killmeplease]

Warmonger penalty per city:

estimated_num_cities = 13 (Duel), 26 (Tiny), 39 (Small), 52 (Standard), 80 (Large), 132 (Huge)

total_num_cities = max(total number of cities current in the game, 1)

num_old_owner_cities = max(old owner's city count, 1)

warmonger change = (10 * estimated_num_cities) / (total_num_cities * num_old_owner_cities)

So if I take one of russia's cities and she has 20 cities on Standard, then I take a warmonger hit of (10 * 52) / (20 * 48) = .54. Compared that to the old 5 for declaring war on a major civilization (+10 for conquering).

didnt get it, whats 48?
old owner's city count - what does this mean?

and why numerator rises with the world size? you get bigger penalty on larger maps given anything else is the same?
i dont understand anything

 

[Outlandish]

Thank you for finding that math for me!

I really disagree with how they implemented it, but at least now I have the formula. It just doesn't make sense that if Pacal DoWs me after thousands of years of peace, and I take 1 of his 3 cities I get warmonger. But as it's not likely to change I'm at least a little happy to know when it's going to happen.

 

[adwcta]

didnt get it, whats 48?
old owner's city count - what does this mean?

and why numerator rises with the world size? you get bigger penalty on larger maps given anything else is the same?
i dont understand anything

Examples:

1. There are 24 cities total on a standard map, everyone has 1 city, plus 16 city states.
You take out your neighbor as Atilla. (52*10)/(24*1) = ~21 This is a high diplo modifier, but since it's early game, it WILL go away by mid-game, especially because early turns are faster.

2. There are 40 cities total on a standard map, everyone has 3 cities, plus 16 city states.
You take out your neighbor as Wu and all three of his cities. (52*10)/(40*3) + 520/(40*2) + 520/40 = ~24 This is a slightly higher diplo modifier, and should also go away later, but this will feel longer, since turns get longer. By this point in the game, you only want to do this if you have DoFs, and if you can get someone to join you.

3. This one's better.
There are 60 cities total on a standard map, Hiawatha has 40, your neighbor has 2.
You take out your neighbor and both of his cities. 520/120 + 520/60 = ~13. This will go away almost twice as fast as examples 1 and 2. If you only take 1 city, that's ~4, and will go away 6x faster than examples 1 and 2.

4. This one's much better.
There are 60 cities total on a standard map, Hiawatha has 40.
You take 1 city from Hiawatha, wipe out his army, and sue for peace taking Hiawatha's gold/offering city. 520/(40*60) = ~0.2. No one will care about this.

The large map modifier is to offset the expected larger number of total cities in the world in the denominator.

edit: as a frame of reference for these numbers, declaring war on a CS to worker steal = 2.5.

 

[HerrDoktor]

So, just to state it as a simple set of rules:

1. You get warmonger for taking cities, not killing enemies or plundering. It's the same for capitals as any other city;
2. It's universal and has nothing to do with your enemy's reputation;
3. The warmonger penalty is larger the fewer cities the attacked civilization has. So the first city you take has far less impact than the last one and city states always have the huge impact of the last city;
4. You clean your name by liberating, and it generally will have a strong impact, because it's the opposite of taking the last city, when you're reviving a city-state or a dead civ.

So that explains why leaders consider you a filthy warmonger for wiping out their worst enemy, as relations are just not taken into consideration. But you can always clear your name by liberating.

A better formula would take in consideration how the other civs see your victim or at least your victim's warmonger status.

 

[Deggial]

Thank you for these interesting and enlightening posts! Very helpful!

All things considered, I think the BNW model is way more reasonable compared to the old calculation. I like the change - even if it makes my usual play-style somehow harder. Well, I'll adapt myself...

 

[kaspergm]

All things considered, I think the BNW model is way more reasonable compared to the old calculation. I like the change - even if it makes my usual play-style somehow harder. Well, I'll adapt myself...

For me it's quite the opposite, I think the old system was bad, but the new system is possibly worse. I think the Warmonger hate mechanism - in any form - has a huge potential for completely disrupting a major part of the game mechanism, namely the war game.

Imo. the Warmonger system needs a complete reworking to include some of the following features:

• A casus belli system to determine how justified you are in declaring war against someone. If they converted your cities, stole your technologies or your land, broke promises they had made to you, this should give you righteous reason to DoW them. How justified a DoW is may differ between civs - civs that share your religion may be more inclined to see converting cities as justification, for instance. If you DoW someone without justification, you get a big warmonger penalty.
• Better integration with how diplomatic game. If you DoW someone, their friends will see you as a warmonger, while civs that disliked him and/or like you will be less likely to give you a warmonger lable.
• No or only small penalty based on the number of cities you take, but penalty from razing enemy cities.
• Civs with large affinity towards war themselves will be less likely to rate you a warmonger.
• Civs that take the Freedom ideology will give you a larger warmonger penalty, while civs that take the Autocracy will give you a smaller warmonger penalty than default (Order).
• Joining someone in a war means no mutual warmonger hate between the two civs.

Features like this would actually make it possible to play a meaningful game as an early autocrat, which I think is borderline impossible at the moment.

 

[Kilplow]

[*]Civs with large affinity towards war themselves will be less likely to rate you a warmonger.

There's some type of modifier about how civs treat warmongering. I want to say Denmark was tolerate of them while most of the other civs I've run into dislike or hate them.

 

[Razor Rex]

I certainly think that the Warmongering penalty is a bit stiff. It's completely mauled any
diplomatic attempts I have made in my current game and has caused me to take part in
a two front defensive war.

 

[adwcta]

2. It's universal and has nothing to do with your enemy's reputation;

So that explains why leaders consider you a filthy warmonger for wiping out their worst enemy, as relations are just not taken into consideration. But you can always clear your name by liberating.

Not completely true. While the reputation of the AI that you are taking cities from is not taken into consideration, your relationship with the AI who's doing the judging IS a modifier.

The equation above just gives the diplo modifier. This diplo modifier is then modifier by relations. So, your friends will not hate you as much as neutral AIs, and "guarded" or "hostile" AIs will hate you more for it.

Additionally, you will not get any warmonger penalties for civs engaged in the war with you. So, it's very good to ask/bribe people to join your war.

The calculus is now changed. If there's a neighbor you don't care much for who has no friends, and is spamming missionaries and prophets at you, or with tasty workers on the borders.... Denounce then DOW, take what you need, kill his army, get one city down to 0 health, and sue for peace. This would have drastic consequences previously, but it's not so bad now. In fact, if you get the world denouncing that neighbor in a chain denounce, it will strengthen the bonds of your friendship.

 

[Lyoncet]

<smart things>

Yes, please.

 

[anewone]

• Civs that take the Freedom ideology will give you a larger warmonger penalty, while civs that take the Autocracy will give you a smaller warmonger penalty than default (Order).

Features like this would actually make it possible to play a meaningful game as an early autocrat, which I think is borderline impossible at the moment.

Why would Freedom ideology give a larger warmongering penalty than others? I have no idea how all of these mess with balance of the game but I'd prefer a system where a civ who shares ideology with you would give you lesser warmongering penalty and a civ who doesn't share ideology with you would give you a larger penalty. That'd be a bit more in line with the concept of the war of ideologies as well.

 

[Teholb]

For me it's quite the opposite, I think the old system was bad, but the new system is possibly worse. I think the Warmonger hate mechanism - in any form - has a huge potential for completely disrupting a major part of the game mechanism, namely the war game.

Imo. the Warmonger system needs a complete reworking to include some of the following features:

• A casus belli system to determine how justified you are in declaring war against someone. If they converted your cities, stole your technologies or your land, broke promises they had made to you, this should give you righteous reason to DoW them. How justified a DoW is may differ between civs - civs that share your religion may be more inclined to see converting cities as justification, for instance. If you DoW someone without justification, you get a big warmonger penalty.
• Better integration with how diplomatic game. If you DoW someone, their friends will see you as a warmonger, while civs that disliked him and/or like you will be less likely to give you a warmonger lable.
• No or only small penalty based on the number of cities you take, but penalty from razing enemy cities.
• Civs with large affinity towards war themselves will be less likely to rate you a warmonger.
• Civs that take the Freedom ideology will give you a larger warmonger penalty, while civs that take the Autocracy will give you a smaller warmonger penalty than default (Order).
• Joining someone in a war means no mutual warmonger hate between the two civs.

Features like this would actually make it possible to play a meaningful game as an early autocrat, which I think is borderline impossible at the moment.

I agree with everything quoted above. I wish more factors were taken into account- sometimes it's just frustrating to be denounced and whatnot for justly lashing out at the oppressive civ to the west.

 

[Mentatstrategy]

For me it's quite the opposite, I think the old system was bad, but the new system is possibly worse. I think the Warmonger hate mechanism - in any form - has a huge potential for completely disrupting a major part of the game mechanism, namely the war game.

Imo. the Warmonger system needs a complete reworking to include some of the following features:

• A casus belli system to determine how justified you are in declaring war against someone. If they converted your cities, stole your technologies or your land, broke promises they had made to you, this should give you righteous reason to DoW them. How justified a DoW is may differ between civs - civs that share your religion may be more inclined to see converting cities as justification, for instance. If you DoW someone without justification, you get a big warmonger penalty.
• Better integration with how diplomatic game. If you DoW someone, their friends will see you as a warmonger, while civs that disliked him and/or like you will be less likely to give you a warmonger lable.
• No or only small penalty based on the number of cities you take, but penalty from razing enemy cities.
• Civs with large affinity towards war themselves will be less likely to rate you a warmonger.
• Civs that take the Freedom ideology will give you a larger warmonger penalty, while civs that take the Autocracy will give you a smaller warmonger penalty than default (Order).
• Joining someone in a war means no mutual warmonger hate between the two civs.

Features like this would actually make it possible to play a meaningful game as an early autocrat, which I think is borderline impossible at the moment.

Nah, the way it is, is WAY more realistic than these changes would be. Real world if you go to war a lot, the whole world notices, even that countries friends.

 

[HerrDoktor]

Not completely true. While the reputation of the AI that you are taking cities from is not taken into consideration, your relationship with the AI who's doing the judging IS a modifier.

The equation above just gives the diplo modifier. This diplo modifier is then modifier by relations. So, your friends will not hate you as much as neutral AIs, and "guarded" or "hostile" AIs will hate you more for it.

Additionally, you will not get any warmonger penalties for civs engaged in the war with you. So, it's very good to ask/bribe people to join your war.

The calculus is now changed. If there's a neighbor you don't care much for who has no friends, and is spamming missionaries and prophets at you, or with tasty workers on the borders.... Denounce then DOW, take what you need, kill his army, get one city down to 0 health, and sue for peace. This would have drastic consequences previously, but it's not so bad now. In fact, if you get the world denouncing that neighbor in a chain denounce, it will strengthen the bonds of your friendship.

Well, actually no, it still has nothing to do with their relationship to your victim. You have another negative modifier for attacking a friend of the AI or a positive one for being on the same war as them. But the warmonger modifier itself is independent of who you are attacking.

 

[kaspergm]

There's some type of modifier about how civs treat warmongering. I want to say Denmark was tolerate of them while most of the other civs I've run into dislike or hate them.

Yes, thankfully, at least they did that to some extent in the new version. Still, when Genghis Kahn complains about me being a warmonger, there obviously is some tweaking that still needs to be done!

Why would Freedom ideology give a larger warmongering penalty than others?

Maybe you misunderstood me - what I mean is that if you DoW someone (without just reason) a 3rd party civ who has taken Freedom should hate you more for it than a 3rd party civ who has taken Autocracy.

Nah, the way it is, is WAY more realistic than these changes would be. Real world if you go to war a lot, the whole world notices, even that countries friends.

That may be, but question is, does it make for a better game? And I didn't say anywhere that you should be labelled as a warmonger if you go to war a lot, that may be fine - but I shouldn't be labeled a warmonger if someone DoW's me, and I then take some of his cities*, and I most certainly should not be labelled a warmonger by someone who asked me to join the war himself. Thankfully I do think they have changed the last bit in the new system.

* If you want to give me a penalty for taking other players cities, give me a penalty for expanding too aggressively, I could live with that in game context, but putting a warmonger label on me for taking cities from someone who DoW'ed me is just jarring and breaks immersion.

 

[andrewlt]

The way they did it has huge holes in the formula. I wiped out a 3-city civ around turn 100-150. The Mongols declared war on six total city states, conquering 2 around turn 250. Genghis reduced a 4-city civ to 1 around turn 300. Around turn 350-400, I still have more civs denouncing me than him and it's due to my warmonger penalties. It's a pretty huge list of greens, otherwise.

The reason? After I did my conquest, multiple civs expanded. Genghis got far lower penalties for his later wars, even though mine was already decaying for 200 turns.

 

[HerrDoktor]

A question: since the warmonger modifier is universal, you get it when joining a civ on war, correct? You'll only also get a positive one for fighting on the same side, but they may still hate you and denounce you for joining a war on their side, if you happen to capture cities.

 

[Raptor Llama]

[*]Civs with large affinity towards war themselves will be less likely to rate you a warmonger.
[*]Joining someone in a war means no mutual warmonger hate between the two civs.[/list]

These two things are already in the game. You get no warmonger penalty if the third party is already in a war with the civ you declared on. You can take all their cities and as long as the civ is at war with them, they won't care.

The first one is done with the warmonger hate bias. Some civs have higher warmonger hate modifiers (that multiplies the warmonger score mentioned previous) while others have very low ones. I think Attila's is like 3 (+/- 2 every game).

I do agree the rest need to be in, though. Reason for denouncement would be nice as well.