GunsGermsandSteel
Warlord
- Joined
- Feb 27, 2020
- Messages
- 155
While I generally like to include a TLDR in long posts I want to point out that I think all these changes need to be implemented together. The main idea is to make the AI more competitive throughout the game, rebalance land combat, add interesting layers to the domination game, and solve the late medieval-early industrial siege grind. While one of the main aims of these suggestions is to improve AI combat it is also to smooth game play for the human player.
Ultimately better coding would be the 'ideal' way to make the AI more challenging militarily, but as it is the way the following units/concepts work together really make it quite hard for the AI to be competitive. For the most part cavalry and ranged units being 'better' than standard melee gives the human player a massive bonus in that both of those unit types require more positioning/skill to use. If melee and especially spear-men were better, and less vulnerable then the AI would have a more effective army.
These suggestions help fix the following problems:
-AI struggling to take cities (especially walled cities)
-AI having lopsided bonuses (excessively strong at start, overtaken by human snowballing)
-Ranged being able to snipe melee units
-Anti-Cav being ineffective at its job (both because it's too expensive, and because ranged wipes it out)
-Encampments being overly simplistic and unbalanced defensively.
Civ Improvemnts
Difficulty increases should be a more ‘staggered’ experience. As it is the AI gets the majority of their truly impactful bonuses at the beginning of the game – extra settlers/warriors/civics/techs… While the AI certainly needs some bonuses, I think the issue with frontloading so many bonuses is the game is all about surviving the early eras and then eventually steamrolling in the later eras. Early game is high tension mid-late game the tension evaporates.
Staggered Bonuses
-AI gets bonuses when attacking cities. I’m not sure the exact numbers of this but it would mainly apply to melee and siege units. Somewhere in the range of 15-33% more combat damage when attacking cities. This number would scale as the difficult scaled up 15% at king, rising to 33% at deity. As it is once walls go up the AI is virtually inept at taking cities. Also, rather than just getting bonuses at the very beginning, they would have those bonuses spread out so that at the beginning of each era they get X number of new units/techs. X would have to be balance tested.
Changes to ranged
Less combat strength about -5. (would have to be tested) Replace Arrow Storm promotion with ability to shoot over hills (this existed in real life).
Ranged should be more support/deal damage to make units get wounded before melee attacks/force multiplier because of one UPT. And not so much of a sniper unit.
Changes to Anti-Cavalry
Shield Drills – +7 combat strength against ranged attacks (replaces thrust). Pikemen and Spearmen(and UU replacements) would cost less production. The idea is that spearmen would be more numerous and their promotions are already linked to that.
Changes to Encampments
Currently encampments basically become a second city. This can be hugely problematic to assault when you have two cities and two encampments all within range of a hilly avenue of approach, with ranged garrisoned in them! It can be quite unbeatable until artillery or even bombers come around. Civs like Poland and Zulu who spam encampments can be an awful chore to clear out, even when you’ve killed off all their armies.
-Military engineers gain sapper ability. Uses 1 change to deal massive damage to walls.
-Encampments gain a new universally exclusive upgrade tree of buildings specifically tied to defense: Rampart/Citadel/Flak Tower. Encampments would still provide a base defensive bonus to units garrisoned on them, but would not share wall upgrades with city and would only gain walls/ranged attack if ramparts and/or citadel are chosen as upgrades.
Changes to Siege weapons
20% resistance to attacks from city/upgraded encampment. Has a promotion that allows it to shoot over hills called ‘Indirect Fire’ (replaces one of the early promotions).
Again these changes are intended to be combined. As individual changes they might be problematic. On the whole though the AI would be better at fighting (ranged being so good gives the human player such a huge advantage because you can pick off units without even being dealt damage).
Ultimately better coding would be the 'ideal' way to make the AI more challenging militarily, but as it is the way the following units/concepts work together really make it quite hard for the AI to be competitive. For the most part cavalry and ranged units being 'better' than standard melee gives the human player a massive bonus in that both of those unit types require more positioning/skill to use. If melee and especially spear-men were better, and less vulnerable then the AI would have a more effective army.
These suggestions help fix the following problems:
-AI struggling to take cities (especially walled cities)
-AI having lopsided bonuses (excessively strong at start, overtaken by human snowballing)
-Ranged being able to snipe melee units
-Anti-Cav being ineffective at its job (both because it's too expensive, and because ranged wipes it out)
-Encampments being overly simplistic and unbalanced defensively.
Civ Improvemnts
Difficulty increases should be a more ‘staggered’ experience. As it is the AI gets the majority of their truly impactful bonuses at the beginning of the game – extra settlers/warriors/civics/techs… While the AI certainly needs some bonuses, I think the issue with frontloading so many bonuses is the game is all about surviving the early eras and then eventually steamrolling in the later eras. Early game is high tension mid-late game the tension evaporates.
Staggered Bonuses
-AI gets bonuses when attacking cities. I’m not sure the exact numbers of this but it would mainly apply to melee and siege units. Somewhere in the range of 15-33% more combat damage when attacking cities. This number would scale as the difficult scaled up 15% at king, rising to 33% at deity. As it is once walls go up the AI is virtually inept at taking cities. Also, rather than just getting bonuses at the very beginning, they would have those bonuses spread out so that at the beginning of each era they get X number of new units/techs. X would have to be balance tested.
Changes to ranged
Less combat strength about -5. (would have to be tested) Replace Arrow Storm promotion with ability to shoot over hills (this existed in real life).
Ranged should be more support/deal damage to make units get wounded before melee attacks/force multiplier because of one UPT. And not so much of a sniper unit.
Changes to Anti-Cavalry
Shield Drills – +7 combat strength against ranged attacks (replaces thrust). Pikemen and Spearmen(and UU replacements) would cost less production. The idea is that spearmen would be more numerous and their promotions are already linked to that.
Changes to Encampments
Currently encampments basically become a second city. This can be hugely problematic to assault when you have two cities and two encampments all within range of a hilly avenue of approach, with ranged garrisoned in them! It can be quite unbeatable until artillery or even bombers come around. Civs like Poland and Zulu who spam encampments can be an awful chore to clear out, even when you’ve killed off all their armies.
-Military engineers gain sapper ability. Uses 1 change to deal massive damage to walls.
-Encampments gain a new universally exclusive upgrade tree of buildings specifically tied to defense: Rampart/Citadel/Flak Tower. Encampments would still provide a base defensive bonus to units garrisoned on them, but would not share wall upgrades with city and would only gain walls/ranged attack if ramparts and/or citadel are chosen as upgrades.
Changes to Siege weapons
20% resistance to attacks from city/upgraded encampment. Has a promotion that allows it to shoot over hills called ‘Indirect Fire’ (replaces one of the early promotions).
Again these changes are intended to be combined. As individual changes they might be problematic. On the whole though the AI would be better at fighting (ranged being so good gives the human player such a huge advantage because you can pick off units without even being dealt damage).
