TL;DR a war weariness system could get the AI to focus on capturing/defending realistic goals and pressure them out of pursuing simultaneous wars. There is already some kind of balance of power calculation for deciding when to declare war/offer peace, but it seems to rely on omniscience of troop totals & captive status of lords rather than being goal-oriented. The AI also has inscrutable strategy when prosecuting wars, making decisions that seem weird to the player like over-extending into enemy territory to capture settlements in weird places. This system could present goals to the player up-front when their nation declares war (and easily allow the player to set goals themselves within the same rules), direct AI behavior to achieve those goals, and then defend those achievements while seeking a conclusion to the war. Nations in multiple wars would be pressured to end them quickly, either by making lots of gains in one, some gains in both, or badly losing in one or both.
The willingness of the AI to penetrate deep into enemy territory and secure multiple settlements in a single war can be abstracted with a war weariness mechanic. If sufficiently abstracted, discrete values can be assigned to castles, cities, and armies, the sum of which can be checked against a nation's war weariness value to determine if they are willing to pursue multiple settlement sieges. In a very basic implementation, a city could have a value of 4, a castle a value of 2, and an army or village a value of 1, while war weariness could span a range of 0 to 10 (for example). Capturing or losing any of these things is costly, and a nation could only reasonably bear so much strain before experiencing widespread failures in things like production and supply lines. The loss of cities, castles, and villages reflects a loss in logistical and supply capacity, rather than stretching those resources further, but the end result is the same. Similarly, losing an army means losing a significant amount of manpower, which takes time to be replaced. Being at peace would periodically reduce war weariness level, perhaps by 1 to 3 per in-game month depending on desired pacing. New offensive wars would not be declared unless war weariness was low enough to allow for multiple territorial gains or one high-priority/value gain (more on priorities later). The level of war weariness when declaring war is determined by remaining war weariness from the previous war, or the highest level of war weariness in an active war. War weariness is independently tracked for each active war, but territorial gains or losses--as well as army losses--apply weariness to all active wars.
With the above values for cities, castles, villages and armies, a nation in a single war and with no gains or losses of territory in its previous war (which is the case for nations at the very beginning of the game) would have a war weariness of 0, and thus they would be willing to pursue multiple territory objectives in that war. Each settlement simply adds its value to the current war weariness until meeting the maximum value of 10. In this way, nations which have an extremely high war weariness level of 8 will still consider a minor territorial gain by taking a castle, but they will reasonably choose not to pursue a more intensive siege of a city and will opt instead to defend their territory. This is sensible because a war weariness level that high means they either have lots of new territory they must garrison & patrol, or they have recently lost a significant amount of territory. In our example of a nation with 0 war weariness, they would rationally prioritize settlements by these criteria in order of importance: reclamation of lost settlement, border proximity, prosperity, garrison strength (inversely proportional), prosperity of attached settlements. So, their first and highest priority is reconquering lost territory, then each step after that is used as a tie-breaker. They want to take territory which is on their border, focusing on the highest prosperity settlements, with the weakest garrisons, with the highest prosperity villages. This is all to say they want to take the highest value targets with the lowest expected cost in troops and time.
In this hypothetical situation of a fresh war in a fresh campaign, a nation might select 3 castles and a city as their objectives. So, their war weariness potential is 2 + 2 + 2 + 4, reaching the maximum value of 10, with 2 for each castle and 4 for the city. They will try to achieve these objectives and push their active war weariness to 10 (which naturally pushes their opponent's war weariness to 10 as well for losing all of those settlements). If this happened it would be a decisive, strategic victory, and naturally both sides would seek peace: one met their goals, and the other wishes to survive the ordeal. Realistically what would probably happen is one side would lay siege to a settlement with an army, and then the other side would break the siege with an army of their own or (if they arrived late) engage the invading army in a field battle before attempting to retake the settlement. When losing an army, the losing side would take 1 point of war weariness. Likewise, razing a village would impose 1 war weariness on the owner of the village. This insures that even in a territorial stalemate, wars could still end as armies are lost and villages are razed. If a nation's desired goals are no longer compatible with their current level of war weariness, then they drop the lowest-priority goal (again using the logic above, so usually the desired settlement furthest from their border). If this is a city, then they could recalculate and shift the goal to a castle instead, as long as another one is available to be targeted and their new weariness target is 10 or below.
In this way, our hypothetical nation might lose an army, drop a castle from their target list, and continue pursuing territorial gains until their actual war weariness level is 10. If they reach a point where further territorial pursuits are no longer possible (i.e. they have 9 weariness but settlements cost 2 or more) then they will focus on defending their current territory against sieges, raiding nearby hostile villages, and will become receptive to peace offers. When both sides are considered receptive, white peace is made with corresponding tribute based on the difference in war weariness imposed by both sides. A side that decisively wins, capturing several settlements and killing multiple armies will be in a position to demand more tribute. A faction which reaches 10 war weariness will proactively offer peace, and for each point of war weariness that would put them past the maximum of 10 they will offer greater tribute. This would not happen terribly often in isolation, but is a likely situation when a nation is engaged in multiple wars (recall that war weariness suffered in one war is applied to all on-going wars). Any nation which finds itself in this situation from suffering losses should be trying very hard to reduce the number of wars they are in as it is an existential crisis, so they would proactively seek peace even if it means paying significant tribute.
It is possible, however, to be in multiple wars and be winning, but war weariness will still go up in all active conflicts as the acquisition of new settlements imposes logistical, supply, security, and administrative burden on the occupier. A nation which is decisively winning a war on one front has the potential to focus their attention on that one war and resolve it in good time, and should prioritize doing that over splitting their attention with another war. In a vacuum, two nations at war may end up a situation where one is receptive to peace and the other isn't. This is realistic, as the side which is receptive to peace must logically be suffering significant troop and/or village losses to have a difference in war weariness, but not be suffering absolute strategic defeat by means of castle and city losses. In this case, the winning side will want to continue pursuing their territorial goals and are in fact incentivized to do so: more war weariness imposed on their opponent past 10 means greater tribute when they do make peace. The main consideration here is that razing villages makes them less valuable to their owner, which may be the invader at the end of the war, and so the AI should not devote much time to razing villages unless their war weariness precludes them from pursuing a siege or when they have no active armies.
Finally, when peace is declared the same difference calculation used to determine tribute can also be used to alter the lingering weariness after the war. The value can't exceed 10, but if one side was forced to pay high tribute (i.e. they had several "soft" points past 10) then some value modified by the difference in their "soft" total and the victor's war weariness can be removed from the victor's final post-war weariness. In other words, if the losing side of a war had 10 hard weariness points and received 3 more soft points from losing an army and a castle, pushing the victor to 9 weariness and the peace receptiveness threshold, then the difference in their war weariness is (10+3) - 9, or 4. Rather than the victor having 9 war weariness once the war has ended, it could instead be reduced by the difference, up to a maximum reduction like 5. It would thus be possible for a nation to decisively win a war, achieving all of their expansion goals, killing multiple armies, and razing multiple villages, then come out of the war with war weariness as low as 4. This gives them enough room to consider another offensive war with someone else and the potential for multiple territorial gains, though their potential gains are reduced because even winning isn't free.
Naturally, the more visible this is to the player the better it is. It is good for the player to be in the loop on their nation's strategic goals, so they know where they can expect to find their fellow lords and what actions they can take to reach victory. It is beneficial for the player to have a clear path to their own goals within their nation and how they can influence the outcome of a war, which shouldn't be as mindless and boring as collecting enemy lords and sticking them in a dungeon until the balance of power shifts so much they just quit. For player rulers, it should be possible to manually set target settlements to follow the player's strategic aims, though their potential weariness costs should remain the same so the player's choices don't require special handling by the AI.
The willingness of the AI to penetrate deep into enemy territory and secure multiple settlements in a single war can be abstracted with a war weariness mechanic. If sufficiently abstracted, discrete values can be assigned to castles, cities, and armies, the sum of which can be checked against a nation's war weariness value to determine if they are willing to pursue multiple settlement sieges. In a very basic implementation, a city could have a value of 4, a castle a value of 2, and an army or village a value of 1, while war weariness could span a range of 0 to 10 (for example). Capturing or losing any of these things is costly, and a nation could only reasonably bear so much strain before experiencing widespread failures in things like production and supply lines. The loss of cities, castles, and villages reflects a loss in logistical and supply capacity, rather than stretching those resources further, but the end result is the same. Similarly, losing an army means losing a significant amount of manpower, which takes time to be replaced. Being at peace would periodically reduce war weariness level, perhaps by 1 to 3 per in-game month depending on desired pacing. New offensive wars would not be declared unless war weariness was low enough to allow for multiple territorial gains or one high-priority/value gain (more on priorities later). The level of war weariness when declaring war is determined by remaining war weariness from the previous war, or the highest level of war weariness in an active war. War weariness is independently tracked for each active war, but territorial gains or losses--as well as army losses--apply weariness to all active wars.
With the above values for cities, castles, villages and armies, a nation in a single war and with no gains or losses of territory in its previous war (which is the case for nations at the very beginning of the game) would have a war weariness of 0, and thus they would be willing to pursue multiple territory objectives in that war. Each settlement simply adds its value to the current war weariness until meeting the maximum value of 10. In this way, nations which have an extremely high war weariness level of 8 will still consider a minor territorial gain by taking a castle, but they will reasonably choose not to pursue a more intensive siege of a city and will opt instead to defend their territory. This is sensible because a war weariness level that high means they either have lots of new territory they must garrison & patrol, or they have recently lost a significant amount of territory. In our example of a nation with 0 war weariness, they would rationally prioritize settlements by these criteria in order of importance: reclamation of lost settlement, border proximity, prosperity, garrison strength (inversely proportional), prosperity of attached settlements. So, their first and highest priority is reconquering lost territory, then each step after that is used as a tie-breaker. They want to take territory which is on their border, focusing on the highest prosperity settlements, with the weakest garrisons, with the highest prosperity villages. This is all to say they want to take the highest value targets with the lowest expected cost in troops and time.
In this hypothetical situation of a fresh war in a fresh campaign, a nation might select 3 castles and a city as their objectives. So, their war weariness potential is 2 + 2 + 2 + 4, reaching the maximum value of 10, with 2 for each castle and 4 for the city. They will try to achieve these objectives and push their active war weariness to 10 (which naturally pushes their opponent's war weariness to 10 as well for losing all of those settlements). If this happened it would be a decisive, strategic victory, and naturally both sides would seek peace: one met their goals, and the other wishes to survive the ordeal. Realistically what would probably happen is one side would lay siege to a settlement with an army, and then the other side would break the siege with an army of their own or (if they arrived late) engage the invading army in a field battle before attempting to retake the settlement. When losing an army, the losing side would take 1 point of war weariness. Likewise, razing a village would impose 1 war weariness on the owner of the village. This insures that even in a territorial stalemate, wars could still end as armies are lost and villages are razed. If a nation's desired goals are no longer compatible with their current level of war weariness, then they drop the lowest-priority goal (again using the logic above, so usually the desired settlement furthest from their border). If this is a city, then they could recalculate and shift the goal to a castle instead, as long as another one is available to be targeted and their new weariness target is 10 or below.
In this way, our hypothetical nation might lose an army, drop a castle from their target list, and continue pursuing territorial gains until their actual war weariness level is 10. If they reach a point where further territorial pursuits are no longer possible (i.e. they have 9 weariness but settlements cost 2 or more) then they will focus on defending their current territory against sieges, raiding nearby hostile villages, and will become receptive to peace offers. When both sides are considered receptive, white peace is made with corresponding tribute based on the difference in war weariness imposed by both sides. A side that decisively wins, capturing several settlements and killing multiple armies will be in a position to demand more tribute. A faction which reaches 10 war weariness will proactively offer peace, and for each point of war weariness that would put them past the maximum of 10 they will offer greater tribute. This would not happen terribly often in isolation, but is a likely situation when a nation is engaged in multiple wars (recall that war weariness suffered in one war is applied to all on-going wars). Any nation which finds itself in this situation from suffering losses should be trying very hard to reduce the number of wars they are in as it is an existential crisis, so they would proactively seek peace even if it means paying significant tribute.
It is possible, however, to be in multiple wars and be winning, but war weariness will still go up in all active conflicts as the acquisition of new settlements imposes logistical, supply, security, and administrative burden on the occupier. A nation which is decisively winning a war on one front has the potential to focus their attention on that one war and resolve it in good time, and should prioritize doing that over splitting their attention with another war. In a vacuum, two nations at war may end up a situation where one is receptive to peace and the other isn't. This is realistic, as the side which is receptive to peace must logically be suffering significant troop and/or village losses to have a difference in war weariness, but not be suffering absolute strategic defeat by means of castle and city losses. In this case, the winning side will want to continue pursuing their territorial goals and are in fact incentivized to do so: more war weariness imposed on their opponent past 10 means greater tribute when they do make peace. The main consideration here is that razing villages makes them less valuable to their owner, which may be the invader at the end of the war, and so the AI should not devote much time to razing villages unless their war weariness precludes them from pursuing a siege or when they have no active armies.
Finally, when peace is declared the same difference calculation used to determine tribute can also be used to alter the lingering weariness after the war. The value can't exceed 10, but if one side was forced to pay high tribute (i.e. they had several "soft" points past 10) then some value modified by the difference in their "soft" total and the victor's war weariness can be removed from the victor's final post-war weariness. In other words, if the losing side of a war had 10 hard weariness points and received 3 more soft points from losing an army and a castle, pushing the victor to 9 weariness and the peace receptiveness threshold, then the difference in their war weariness is (10+3) - 9, or 4. Rather than the victor having 9 war weariness once the war has ended, it could instead be reduced by the difference, up to a maximum reduction like 5. It would thus be possible for a nation to decisively win a war, achieving all of their expansion goals, killing multiple armies, and razing multiple villages, then come out of the war with war weariness as low as 4. This gives them enough room to consider another offensive war with someone else and the potential for multiple territorial gains, though their potential gains are reduced because even winning isn't free.
Naturally, the more visible this is to the player the better it is. It is good for the player to be in the loop on their nation's strategic goals, so they know where they can expect to find their fellow lords and what actions they can take to reach victory. It is beneficial for the player to have a clear path to their own goals within their nation and how they can influence the outcome of a war, which shouldn't be as mindless and boring as collecting enemy lords and sticking them in a dungeon until the balance of power shifts so much they just quit. For player rulers, it should be possible to manually set target settlements to follow the player's strategic aims, though their potential weariness costs should remain the same so the player's choices don't require special handling by the AI.
Last edited:
