Peace Logic for Upcoming 1.1 Version

Discussion in 'Community Patch Project' started by Recursive, Oct 12, 2021.

  1. Recursive

    Recursive Radiating Doom At 10x The Background Level Moderator

    Joined:
    Dec 19, 2017
    Messages:
    2,808
    Gender:
    Male
    Here's the revised peace logic for the upcoming version 1.1, in great detail. Feel free to suggest how this could be improved.

    Wrote this up somewhat hastily so feel free to ask if anything seems unclear.

    Code:
    [PART 1: INITIAL CHECKS]
    
    1. Peace Blocked
    - If game options "Permanent War" or "Permanent War or Peace" are enabled, AI will never make peace with anyone.
    - If AI is a human's teammate, they will never make peace with anyone.
    - If AI is someone's vassal, they will never make peace with anyone.
    - AI can't make peace with vassals.
    - AI can't make peace if at war with a major civ for less than 10 turns.
    - AI can't make peace if at war with a City-State for less than 1 turn.
    - If player captured one of the AI's cities less than 2 turns ago and the AI has more than one combat unit near that player's cities, they can't make peace.
    - AI can't make peace with a City-State if at permanent war or at war with the City-State's ally.
    - AI can't make peace if locked into war by a trade deal or coop war.
    - AI can't make peace if the other player has no capital city.
    - AI can't make peace if CvTeam::canChangeWarPeace() returns false for some other reason.
    
    
    2. Critical State Check
    - AI is in a critical state if their empire is Super Unhappy.
    - AI is in a critical state if their capital is projected to fall in the next turn, or is at 1/2 HP or less.
    - AI is in a critical state if their War State with any major civ is "Nearly Defeated".
    - AI immediately peaces out with all valid City-States if in a critical state.
    - AI will agree to peace with all valid major civs if in a critical state.
    
    
    3. Make Peace With City-States
    - AI will make peace with all City-States if their War State with any major civ is "Defensive", except if:
      a. The City-State's capital OR a city whose original owner is the AI's team is in danger from that AI, and;
      b. That city is "under siege" (took damage last turn / is surrounded by enemies) or blockaded.
    
    - AI can only target one City-State for war at a time, updated each turn in DoUpdateMinorCivApproaches(). If the AI is at war with a City-State that isn't their current target, they will make peace unless:
      a. The City-State's capital OR a city whose original owner is the AI's team is in danger from that AI, and;
      b. That city is "under siege" (took damage last turn / is surrounded by enemies) or blockaded.
    
    - If War State is fine and the City-State is the AI's current target, then the AI won't make peace.
    
    
    [PART 2: EVALUATE EACH MAJOR CIV]
    
    These steps are repeated for each valid major civ.
    
    1. Determine War Duration
    The AI's "war duration" is the number of turns the AI and the player have been at war.
    
    Each time the AI captures a city from the player after the war begins, the war duration resets to 0.
    
    
    2. Calculate Own City Danger
    - Loop through each city that is currently endangered by the player or their allies.
    - Add that city's Danger to the AI's total Danger value.
    - The "we are in serious danger" flag is initially set to FALSE.
    
    
    [Evaluation #1]
    - City Danger starts at 1.
    - Add 1 to City Danger if the city's tactical AI "land zone" is controlled by any of the AI's enemies.
    - Add 1 to City Danger if the city's tactical AI "water zone" is controlled by any of the AI's enemies.
    
    If the city is in danger of falling next turn, is "under siege" (took damage last turn / is surrounded by enemies), or is blockaded AND has taken at least 1/4 of its max HP in damage, proceed to Evaluation #2.
    
    
    [Evaluation #2]
    - If the city is in danger of being lost next turn, add 3 to City Danger. Otherwise, add 1 to City Danger.
    
    - If the city meets one of these criteria:
      a. Is in danger of being lost next turn
      b. Is blockaded AND has taken at least 1/4 of its max HP in damage
      c. Has taken at least 1/2 of its max HP in damage
    
      ...AND meets one of these criteria:
      a. Originally founded by the AI
      b. Is an original major capital
      c. Is the AI's Holy City
      d. Is the other player's Holy City
      e. Has World Wonders
      f. Has National Wonders
    
      ...Then set "we are in serious danger" to TRUE.
    
    
    [Importance Multiplier]
    - If the city is the AI's capital, multiply City Danger by 3.
    - Otherwise, if it's an original major capital, the Holy City for the AI or the player, or has World Wonders, multiply City Danger by 2.
    - Otherwise, if it's any other player's Holy City or has National Wonders, add 1 to City Danger.
    
    
    
    3. Calculate Enemy City Danger
    - Loop through each of the enemy's cities that is currently endangered by the AI or their allies.
    - Add that city's Danger to the enemy player's total Danger value.
    - The "they are in serious danger" and "any of their cities are significantly endangered" flags are initially set to FALSE.
    
    
    [Evaluation #1]
    - City Danger starts at 1.
    - Add 1 to City Danger if the city's tactical AI "land zone" is controlled by any of the player's enemies.
    - Add 1 to City Danger if the city's tactical AI "water zone" is controlled by any of the player's enemies.
    
    If the city is in danger of falling next turn, is "under siege" (took damage last turn / is surrounded by enemies), or is blockaded AND has taken at least 1/4 of its max HP in damage, proceed to Evaluation #2.
    
    
    [Evaluation #2]
    - If the city is in danger of being lost next turn, add 3 to City Danger. Otherwise, add 1 to City Danger.
    
    - If the city meets one of these criteria:
      a. Is in danger of being lost next turn
      b. Is blockaded AND has taken at least 1/4 of its max HP in damage
      c. Has taken at least 1/2 of its max HP in damage
    
      ...Then set "they are in serious danger" flag to TRUE.
    
    
    ["Any Of Their Cities Are Significantly Endangered"]
    Set to TRUE here if City Danger is greater than 1.
    
    
    [Importance Multiplier]
    - If the city is the player's capital, multiply City Danger by 3.
    - Otherwise, if it's an original major capital, the Holy City for the AI or the player, or has World Wonders, multiply City Danger by 2.
    - Otherwise, if it's any other player's Holy City, add 1 to City Danger.
    
    
    
    4. Player Is In Serious Danger and AI is Not
    - If war duration is less than 40, refuse to make peace.
    
    
    5. AI's War Score is <= -90 and AI's Danger > 0
    - Agree to make peace.
    
    
    6. AI's War Score is <= -50 and AI Is In Serious Danger
    - Agree to make peace.
    
    
    7. AI Is In Serious Danger And Player Is Not
    - Agree to make peace.
    
    
    8. Phony War
    A war is phony if it meets all of these criteria:
    - Approach towards this player is not WAR.
    - Player hasn't captured any of the AI's cities before.
    - War State is "Stalemate" or "Calm".
    - War Score is less than 25 and greater than -25.
    - AI doesn't "want to conquer" the player.
    - AI isn't currently attacking the player with an "operation" of units.
    - None of the AI's cities are in danger from the player.
    - None of the player's cities are in danger from the AI.
    - Player passes the proximity test.
    
    The proximity test:
    - If player is the AI's neighbor, they are too close.
    - If player is CLOSE to the AI and military aggressive posture is "medium" or military strength is "powerful", they are too close.
    - If player is FAR or DISTANT from the AI and military aggressive posture is "high" or military strength is "immense", they are too close.
    
    If a war is phony and the player's danger from the AI *and its war allies* is 0, the AI will agree to make peace.
    
    
    9. War Progress
    Skip this check if "Any Of Their Cities Are Significantly Endangered" is TRUE, unless the war duration is at least 40.
    
    War Progress starts at 100.
    Decay per turn: -5 if player is stronger than AI militarily, -4 if Average, -3 if weaker.
    
    Gains:
    +10 each time AI captures a civilian (+20 for Settlers and Trade Units)
    +10 each time AI pillages an improvement (+20 if it's high value)
    +5 each time America AI steals a tile (+20 if it's high value)
    +20 each time AI steals territory with a citadel (+40 if high value tiles are stolen)
    +20 each time AI kills a combat unit or a Great General/Admiral
    +100 each time AI captures a city (+200 for the capital, +150 for the Holy City)
    
    Losses:
    -5 each time AI loses a civilian (-10 for Settlers and Trade Units)
    -5 each time AI's improvements are pillaged (-10 if it's high value)
    -2 each time America steals a tile from the AI (-10 if it's high value)
    -10 each time AI's territory is stolen with a citadel (-20 if high value tiles are stolen)
    -10 each time AI loses a combat unit or a Great General/Admiral
    -50 each time AI loses a city (-100 for the capital, -75 for the Holy City)
    
    Adjustment:
    - Subtract 1 for each point of War Weariness.
    - CBP: If Happiness < 50%, subtract (50 - Happiness %) * 2.
    - CP Only: Subtract 2 for each point of Unhappiness.
    - Subtract (5 * total strategic resource deficit), up to -50.
    
    If Adjusted War Progress is 0 or less and the check wasn't skipped, agree to make peace.
    
    
    10. "Ready For Vassalage" Check
    The AI is "ready for vassalage" if all of these criteria are met:
    - Vassalage is possible
    - War score is at least 75
    - Player isn't in possession of the AI's original capital, Holy City, or any city the AI's team originally founded (ignoring cities > 12 tiles away from AI's cities)
    - Player isn't in possession of any city that the AI is actively trying to liberate (ignoring cities > 12 tiles away from AI's cities)
    - Player isn't close to winning the game
    - Player isn't a backstabber
    
    If all of these criteria are met:
    - The AI is NOT ready for vassalage
    - War State is "Nearly Won"
    - Player's proximity is CLOSE or NEIGHBORS
    - War duration is less than 30
    - Not losing in war overall (with any player)
    - Not very unhappy
    
    ...then the AI will refuse to make peace, because they're close to winning.
    
    
    11. Should Peace Be Considered?
    Any of these are reasons to consider peace:
    - AI is "ready for vassalage"
    - AI is in serious danger
    - AI is war weary (War Weariness is >= 20% of AI's population)
    - AI's approach isn't WAR or HOSTILE
    - AI's War State is "Defensive"
    - AI is losing in war overall
    - AI is very unhappy
    - War is phony, but player is in danger from the AI's war allies (not from the AI itself)
    - War duration is at least 30
    
    If none of these apply, AI will refuse to consider peace.
    
    
    12. Peace Score
    Peace score starts at 0.
    If AI has a Tourism bonus for highest war score with this player, they will deliberately prolong the war (Prolong = TRUE).
    
    
    12A. War Score
    If war score is less than 0, subtract half of it from the peace score (War Score -20: Peace Score +10)
    
    If war score is greater than 0, add 1/5 of it to the peace score (War Score 50: Peace Score +10)
    - Don't do this if Prolong = TRUE
    
    
    12B. War Duration
    If Prolong = FALSE:
    - Duration Penalty equals the War Duration - 10.
    - If War Duration > 15, the Duration Penalty is applied.
    
    If Prolong = TRUE:
    - Duration Penalty equals the War Duration - 20.
    - If War Duration > 30, the Duration Penalty is applied.
    
    If the Duration Penalty is applied, it is done based on proximity to the AI:
    - Proximity NEIGHBORS: Add the Duration Penalty to the peace score.
    - Proximity CLOSE: Add 1.5x of the Duration Penalty to the peace score.
    - Proximity FAR: Add 2x the Duration Penalty to the peace score.
    - Proximity DISTANT: Add 3x the Duration Penalty to the peace score.
    
    
    12C. War Weariness
    - Add 1/4 of War Weariness to the peace score.
    
    - If AI is very unhappy, or unhappy AND war weary (>= 20% of AI's population), add 1/2 of War Weariness to the peace score.
    - Otherwise, if AI is unhappy OR war weary, add 1/3 of War Weariness to the peace score.
    
    
    12D. City Danger
    Danger multiplier for the player and the AI start at 1.
    
    If the AI isn't losing in war overall or its war state is "Nearly Won":
    - If player is an easy target, add 1 to player's danger multiplier.
    - If one of the player's cities is vulnerable to the AI, add 1 to player's danger multiplier.
    - If War State is "Nearly Won", add 1 to player's danger multiplier.
    
    If the player isn't losing overall in war:
    - If AI is an easy target, add 1 to AI's danger multiplier.
    - If one of the AI's cities is vulnerable to the player, add 1 to AI's danger multiplier.
    
    
    Player Is In Danger, AI Is Not
    - If war duration is less than 20 OR player's danger multiplier is at least 3 OR they are in serious danger, subtract (5 * their city danger * their danger multiplier) from the peace score.
    - Otherwise, subtract (2 * their city danger * their danger multiplier) from the peace score.
    
    AI Is In Danger, Player Is Not
    - If war score is less than -15 OR AI's danger multiplier is 3 OR AI is in serious danger, add (5 * AI's city danger * AI's danger multiplier) to the peace score.
    - Otherwise, add (2 * AI's city danger * AI's danger multiplier) to the peace score.
    
    Both Players Are In Danger
    - Subtract (2 * their city danger * their danger multiplier) from the peace score.
    - Add (2 * AI's city danger * AI's danger multiplier) to the peace score.
    
    
    12E. Overall War State
    What is the AI's overall war state against this player?
    
    "Nearly Won"
    - Subtract 0 from the peace score if ready for vassalage, or (10 * player's danger multiplier) if not.
    
    "Offensive"
    - Subtract (5 * player's danger multiplier) from the peace score.
    
    "Calm"
    - If AI is losing in war overall, add (5 * AI's danger multiplier) to peace score.
    - Otherwise, subtract (2 * player's danger multiplier) from the peace score.
    
    "Stalemate"
    - If AI is losing in war overall, add (5 * AI's danger multiplier) to peace score.
    - Otherwise, add (2 * AI's danger multiplier) to peace score.
    
    "Defensive"
    - If AI is losing in war overall, add (10 * AI's danger multiplier) to peace score.
    - Otherwise, add (5 * AI's danger multiplier) to peace score.
    
    
    12F. Special Considerations
    - If in an ongoing coop war against this player, subtract 10 from peace score.
    - If not ready for vassalage and AI wants to conquer this player, subtract 10 from peace score.
    - If the player captured the AI's original capital or Holy City, subtract 15 from peace score.
    - If the player is close to winning the game, subtract 20 from peace score.
    
    
    12G. Peace Score Modifiers
    
    [Losing In War Overall or Very Unhappy]
    - If the player hasn't captured any city from the AI, the AI is ready for vassalage, or War State is "Nearly Won", double the peace score.
    - If no city was captured AND one of the other two conditions are met, double the peace score again.
    
    
    [Not Losing In War Overall Or Very Unhappy]
    - Check if war score is either above 0, or above -15 with the player's target value being AVERAGE or better and the war state being STALEMATE or better. If neither are true, skip these next modifiers.
    
    If player captured a city from the AI:
    - If war score is greater than half the war duration, halve the peace score.
    - Otherwise, reduce the peace score by 25%.
    
    If player did not capture a city AND the AI is going for world conquest AND the player has lost their capital (or the AI is ready for vassalage):
    - If player's economic strength compared to the AI is WEAK or PATHETIC, or target value is BAD or IMPOSSIBLE, double the peace score.
    
    If neither of the above are true, but the player has captured a city the AI is trying to liberate:
    - If AI is diplomatic AND AI's war score is greater than the war duration, halve the peace score.
    - If only one of these is true, reduce the peace score by 25%.
    
    
    [Leader Flavors]
    If war score is > 0, multiply the peace score by (100 + ((Diplo Balance - 5) * 10)) and divide it by 100.
    If war score is < 0, multiply the peace score by (100 + (((Meanness * -1) + 5) * 10)) and divide it by 100.
    
    
    12H. Final Decision
    AI counts the number of "dangerous" major civs they're at war with and subtracts 1 (minimum 0).
    
    Peace Threshold = 20 - (2 * this number), minimum 1.
    
    If peace score >= peace threshold, then agree to make peace.
    Otherwise, don't agree to make peace.
    
     
    Last edited: Oct 12, 2021
    Dimmy, vyyt, Legen and 7 others like this.
  2. ridjack

    ridjack King

    Joined:
    Mar 12, 2017
    Messages:
    987
    Does this apply to players as well? I know there was a recent thread about players using the ability to attack, peace and then declare and attack again every turn as an exploit. I feel like this would be a reasonable restriction to have.

    What in the world does this mean? :lol:

    I feel like I'm misreading this; it sounds like you can't make them a vassal if you capture their capital?

    I would personally like to see war score decay significantly more slowly than it currently does in addition to this. I don't know what the current criteria are, but it feels really weird to kill tens of units, lose none and somehow still have the war end with a 0 war score.

    How does it calculate an "easy target?" Defensive settling and smart placement of units can easily make me essentially invincible with just three or four units, so I'm in no way an "easy target" even if the AI has a hundred waiting for their turn to move in.

    What determines these war states? I see you're using the terms a lot but I don't think they've been defined anywhere?

    -------------------------------------------------

    Of course it's difficult to say for sure without several test games, but my initial reaction to the (new?) war progress and duration mechanics is positive, at least in concept. Numbers are generally lost on me, but I like what this is trying to achieve.
     
  3. dostillevi

    dostillevi Warlord

    Joined:
    Sep 16, 2009
    Messages:
    217
    Also curious what this is about.

    Am I reading correctly that if the AI and the Player are adjacent, the entire "phony war" logic isn't applied and the AI will never choose peace? What's the reasoning for adjacency being a hard blocker to the resolution of a phony war?

    Is war progress going to scale on difficulty, game speed, or map size?
     
  4. Revolutionist_8

    Revolutionist_8 Prince

    Joined:
    Jan 30, 2016
    Messages:
    523
    Location:
    Hungary, Earth, M.W. Galaxy
    The vassalage sounds really restrictive, but maybe I'm interpreting it wrong.
     
  5. Stalker0

    Stalker0 Baller Magnus

    Joined:
    Dec 31, 2005
    Messages:
    8,559
    peace is illogical, peace is flawed. Logic demands WAR!!! :)
     
  6. arievillo

    arievillo Warlord

    Joined:
    Jan 28, 2020
    Messages:
    115
    ETA: Will be uploaded when it's ready.
     
  7. cpm4001

    cpm4001 Goggleman

    Joined:
    May 27, 2013
    Messages:
    2,019
    Location:
    Not here
    My only concern is that I'm not sure if an AI with a spamload of units on the borders of a player city, but unable to take it due to the player's defensive skills, would realize that the city isn't actually "in danger". Otherwise this looks good, particularly with what I assume is some extra logic to encourage the AI to wrap up unproductive long wars. I look forward to testing this out!
     
  8. Recursive

    Recursive Radiating Doom At 10x The Background Level Moderator

    Joined:
    Dec 19, 2017
    Messages:
    2,808
    Gender:
    Male
    No, it doesn't. Implementing such a restriction would require the most dreaded kind of work...UI work! So maybe later.


    Against players with 0 cities, AI will never make peace. It's meant to handle "Complete Kills".



    AI won't vassalize someone who captured AI's original capital.



    Edit: Have adjusted the calculation for next version.



    Code:
    DoUpdateEasyTargets() - updates every turn
    
    [1. Initial Checks]
    Note: when a "return" happens, the function sets Easy Target to TRUE/FALSE for that turn and skips all future checks.
    
    If War State = "Nearly Won"
      return TRUE
    
    If War State = "Defensive" or "Nearly Defeated"
      return FALSE
    
    If we're in terrible shape for war (bankrupt/very unhappy/losing in war overall/etc)
      return FALSE
    
    If we're Nuclear Gandhi and have nukes
      return TRUE
    
    If other player is FAR or DISTANT
      return FALSE
    
    If War Duration >= 30
      return FALSE
    
    If player is a City-State:
     If Boldness > 6 and Target Value >= AVERAGE, return TRUE, otherwise return FALSE
     If Boldness <= 6 and Target Value >= FAVORABLE, return TRUE, otherwise return FALSE
    
    
    [2. War State Check]
    
    Skip Phony Wars.
    
    If War State is "Defensive" in a war against any other player, return FALSE.
    If War State is "Calm" or "Stalemate" in a war against any major civ and that civ's "shouldn't start new wars" flag is set to FALSE, then return FALSE.
    
    At this point, if they're in "terrible shape for war" (bankrupt/very unhappy/losing in war overall/etc), then return TRUE.
    
    
    [3. Strength Comparisons]
    
    If Target Value is SOFT or Military/Economic strength compared to us is PATHETIC, return TRUE.
    
    "Unhappiness Boost" is true if AI is not unhappy and player is unhappy.
    
    "Wants Conquest" is true if one of these is true:
    - AI is going for or close to Domination Victory
    - War State is Offensive and Opinion is Unforgivable
    - Made a deal to go to war against them (trade deal / coop war)
    - They're a reckless expander
    - They're a wonder spammer
    - They're close to winning the game
    - They're our biggest competitor
    
    "Already At War" is true if we're NOT at war with the guy we're looking at and we ARE at war with ANOTHER dangerous major civ.
    
    If "Already At War" = False and (one of "Wants Conquest" OR "Unhappiness Boost") = True
       If player's military strength is POOR or lower and their economic strength is less than POWERFUL, return TRUE.
       If player's military strength is WEAK or lower and their economic strength is less than IMMENSE, return TRUE.
       If player's economic strength is POOR or lower and their military strength is less than STRONG, return TRUE.
       If player's economic strength is WEAK or lower and their military strength is less than IMMENSE, return TRUE.
    
    Otherwise:
       If military strength is WEAK or lower and economic strength is less than STRONG, return TRUE.
       If economic strength is WEAK or lower and military strength is less than STRONG, return TRUE.
    
    
    [4. Nothing Yet?]
    Return FALSE.
    
    Code:
    [War States]
    1. Calculate City Danger and "we/they are in serious danger" using the same calculation used for peace evaluation.
    
    - Our Danger = (Our City Danger * 100) / (Our # of Cities)
    - Their Danger = (Their City Danger * 100) / (Their # of Cities)
    - Danger Percent = (Their Danger * 100) / (Our Danger, minimum 1)
    
    War Score = Current war score
    If > 0 war weariness and very unhappy, double war score if negative, or reduce by 25% if positive.
    
    
    2A. If we're in serious danger and they're not
    War State = DEFENSIVE
    
    2B. If they're in serious danger and we're not
    If War Score > -25, War State = OFFENSIVE, otherwise War State = STALEMATE
    
    2C. If we're both in serious danger
    If War Score > -25, War State = STALEMATE
    
    
    3. If no war state has been set yet:
    If Danger Percent > 100, then War State is STALEMATE if War Score <= -25, or OFFENSIVE otherwise.
    If Danger Percent < 100, then War State is STALEMATE if War Score >= 25, or DEFENSIVE otherwise.
    If Danger Percent = 100, then War State is DEFENSIVE if War Score <= -25, OFFENSIVE if >= 25, or STALEMATE otherwise.
    
    
    4A. If War State = OFFENSIVE and War Score >= 50, War State = NEARLY WON.
    
    4B. If War State = DEFENSIVE and War Score <= -50, War State = NEARLY DEFEATED.
    
    
    5. If neither player is in serious danger, War Score is between -20 and 20 (exclusive), and no city has been captured in > 10 turns, War State = CALM.
    
    
    [Overall War State]
    War State Value starts at 0.
    
    For each major civ:
    - War State Nearly Won: Add 2.
    - War State Offensive: Add 2 if they're in serious danger and we're not, add 1 otherwise.
    - War State Defensive: Subtract 2 if we're in serious danger, subtract 1 otherwise.
    
    If nearly defeated in any war, or Defensive and our capital is at 25% HP or less, then Overall War State is LOSING.
    
    If this is not the case, then:
    - If War State Value < -1, Overall War State is LOSING.
    - If War State Value > 1, Overall War State is WINNING.
    - Otherwise, Overall War State is NEUTRAL.
    
    The second set of criteria is only checked if the AI is not ready to vassalize the player. They will vassalize as long as there aren't nearby cities taken from them or nearby cities they want to liberate, as long as the player isn't close to winning the game or a backstabber.

    Yes - because neighbors could strike at any time, and should never be dismissed as irrelevant. The "phony war" check affects other AI logic, not just peace.


    It doesn't currently.


    If they're losing units their war progress value will go down, incentivizing them to make peace. A city being in "significant" danger requires:
    - The area near the city to be dominated by enemy units, OR the city to have taken damage last turn, OR the city to be surrounded by enemies, OR the city to be blockaded + lost at least 25% of its HP
     
    Last edited: Oct 13, 2021
    Jdoug312 likes this.
  9. ridjack

    ridjack King

    Joined:
    Mar 12, 2017
    Messages:
    987
    This seems solid. Seems it like gives a few different methods to convince an AI to peace out. I wish there weren't quite as much of a focus on threatening their cities, but I don't know what I would suggest as an alternative, so.
     
  10. dostillevi

    dostillevi Warlord

    Joined:
    Sep 16, 2009
    Messages:
    217
    Yup looks pretty nice. Thanks Recursive!
     
  11. Revolutionist_8

    Revolutionist_8 Prince

    Joined:
    Jan 30, 2016
    Messages:
    523
    Location:
    Hungary, Earth, M.W. Galaxy
    So if I steamroll that AI, have 100 warscore, if I haven taken a city near their capital (less than 12 tiles, which is quite a lot to be honest) I can't force their vassalization? That sounds really... restricting and not really fun, but I don't know, just thinking out loud :think:

    But don't misunderstand me, thank you very much for your work, it's really impressive! :)
     
  12. Rekk

    Rekk Emperor

    Joined:
    Dec 9, 2017
    Messages:
    1,236
    No, if you have taken an AI city within 12 tiles of its capital, and the AI has 100 warscore against you, then the AI will refuse to vassalize you.

    In other words, the AI is told to get its cities back from you before vassalizing you, so that they aren't lost "forever".

    This is logic as to when the AI will force a vassalization on others, not rules to which it will submit to being vassalized.

    ("you" means both human and other AI players, just to be clear).
     
    Last edited: Oct 14, 2021
    Recursive and 4CV like this.
  13. Revolutionist_8

    Revolutionist_8 Prince

    Joined:
    Jan 30, 2016
    Messages:
    523
    Location:
    Hungary, Earth, M.W. Galaxy
    Aaaaaaaaaaaaaaaaaaaaaahhhh, that makes perfect sense then, thank you for the clarification! :)
     
  14. Recursive

    Recursive Radiating Doom At 10x The Background Level Moderator

    Joined:
    Dec 19, 2017
    Messages:
    2,808
    Gender:
    Male
    Correct, except that it's within 12 tiles of any of their cities, not just the capital.
     
  15. ridjack

    ridjack King

    Joined:
    Mar 12, 2017
    Messages:
    987
    After a few games played, it is definitely much easier to get and maintain warscore. Probably a little too easy, to be honest, but it feels miles better than before.

    So far, it has been very easy to get them to peace out. I haven't encountered any situations yet when they peace out when they shouldn't.

    However, I also don't feel like I've come up against any determined wars. Whether that's because they're giving up too quickly or because I just haven't come up against it yet, I can't say.
     
    Recursive likes this.
  16. Raitaki

    Raitaki Chieftain

    Joined:
    Oct 19, 2015
    Messages:
    24
    That's up to the attacking side then, no? Obviously when willingness to give up becomes reasonable, then wars would now instead be sustained by the winning side's willingness to continue. It's not like the AI just rolls over and stops producing military once it wants to give up and you reject.
     
  17. azum4roll

    azum4roll Emperor

    Joined:
    Jul 17, 2018
    Messages:
    1,781
    Gender:
    Male
    At the very least the Aztecs should be significantly more willing to peace out at above 25 warscore and significantly less willing at 0-24.
     
    black213 likes this.
  18. Little Billy

    Little Billy Chieftain

    Joined:
    Jan 25, 2021
    Messages:
    5
    Is there any logic to discourage AIs from trying to conquer city states that have their own embassy?
     
    Dimmy likes this.
  19. Recursive

    Recursive Radiating Doom At 10x The Background Level Moderator

    Joined:
    Dec 19, 2017
    Messages:
    2,808
    Gender:
    Male
    ...

    ...

    ...

    ...there will be.
     
    Dimmy and Solic like this.

Share This Page