feedback on AI combat behavior

Me wanting to play the game the way I like is now hyperbole? Are you kidding me?

I don't want to derail the current discussion. It's not relevant at all to VP but you know what- it's relevant to maintaining good discussions on the forums so I'm going to respond to this. Examples of hyperbole that I was referring to throughout your posts:

but you massively increase your war weariness.

You have no idea how much the player's WW goes up by when using this tactic but you are confident in calling it "massive".

Furthermore, you'd be eliminating an entire tactical aspect of the game: how do I approach a city with mounted so I can conquer it and get out on the same turn (can't cross a river or pass alongside ZOC, for example).

The simple act of capping a city and then moving the mounted melee out of the city is "an entire tactical aspect of the game". Seriously? You would think I was proposing to eliminate terrain movement penalties or something.

Moreover, one of the advantages of taking a city with mounted is that you can go for smaller cities behind enemy lines (early game or late game with air support) and see if you can hold on to them with minimal ground support; forcing the unit to stay in would make this entire tactic way too risky for my highly promoted cavalry and it's one of the most fun aspects of the game, so I'm vehemently opposed to it.

Capping cities behind enemy lines is seriously one of the most fun aspects of the game? I mean, I understand that everyone probably enjoys different parts of the game but that's really one that you truly live for? I could understand "I enjoy doing this" or "it's a fun tactic!" but "it's one of the most fun aspects of the game"? Really? You aren't exaggerating here at all?

Your proposal would needlessly deprive me of the ability to do that; needlessly because I'm not using your weird "strategy" of ultra-increasing my War Weariness just to kill some units a bit more easily.

"Ultra increasing my war weariness". lol. Again- you have no idea how much the WW goes up but you're sure it's an "ultra" amount. And it doesn't kill units a bit more easily- You can 1-shot the garrison which can be fairly consequential, especially in the early game.

The strategy is an absolute edge case that needs to be set up specifically or otherwise only happens rarely when you don't plan your invasions properly. It isn't some golden key to victory or a huge exploit everyone uses;

"absolute edge case" and has to be set up specifically... You've seriously never had the opportunity to take a city that was in danger of being retaken? That can't happen fairly naturally during the normal course of play? You don't have to go looking to set this up- it can fall into your lap just by virtue of fighting a war against a fairly even opponent.

And the downside isn't just huge increases in WW but you also pump up your warmonger score and completely destroy all infrastructure in the city, making it worthless.

Again, you don't know the WW effects but are certain they are huge. And the city is worthless because it lost infrastructure because cities can't be useful for their strategic or luxury resources? Their citadels, roads, and healing benefits? Nope- completely useless! Completely!


This has gotten way past productive exchange. The bottom line here is that you are using a strategy to score more kills than you feel you deserve, a strategy many other people aren't using (and even if many also used it you don't see them complaining about it) and to fix it, instead of just not using this strategy yourself, you want to have a change implemented that breaks other tactics which are not exploitative, and this not just for you but for everyone.

So you try to make it seem like I'm the one and only person who is annoyed at this exploit and then just assume everyone is on your side when it comes to loving the ability to move mounted melee out of a city after capping it. My complete guess is that 90% of the VP users wouldn't even notice if this was changed. A few of us would be happy to be rid of an exploit and a few would apparently be dismayed that they lost some mobility on their mounted melee after capping a city.

Then you suggest that I don't know anything about War Weariness, as if I just played this game for the first time, ignorantly assert that just because I don't use your lame tactics I've never seen cities being exchanged back and forth and demand evidence without providing any yourself.

I'm not the one using adjectives like "massive", "ultra", etc. If you're going to use that kind of language then don't whine when people call you out and want to know what counts as massive and ultra. I've said that the war weariness can be managed and I'm comfortable with that statement because it's not hyperbole.

if that doesn't matter to you then ignore it and play the game however you like and let the rest of us do the same.

Again, implying that it's me by myself disliking this exploit vs you and everyone else. How would we know if anyone else dislikes this exploit with out discussing it on the forum? Oddly enough- it turns out many people in this thread seem to agree that it's a problem. Imagine that!

And no, we don't all like to play the game exactly how you do and there are plenty of people playing this game for fun without having to min max every number or be threatened in victory until the end.

This isn't hyperbole per se, but what are you even talking about? What does min maxing anything have to do with anything? You seem to have a problem with people who play the game at higher difficulties or something. I'm sure you have reasons for that but what does that have to do with this discussion?

Ultimately this sounds like many previous attempts to turn VP into the maximally standardized and ultrabalanced, Deity-friendly Pangea number-crunching game, no matter the cost. You need to accept that most people don't play the game that way and like to keep their freedom of action.

This paragraph might be the best example. So much hyperbole. Do you think this strengthens your argument? It comes across as emotional and hurts the discussion.

The issue that is being blown up to enormous proportions

Enormous proportions? Really? Are we proposing readjusting the entire game? Is anyone claiming this is THAT big of a deal? It's an exploit some of us would like to address.

Again- I know this post wasn't helpful to the actual discussion but I still think there is some value in calling people out who aren't discussing things in good faith. I hope that if I resort to hyperbole too much in my posts someone calls me out on it, too. I probably did it at points as well so for that I apologize.
 
I'm not even going to read that line by line since it would be a waste of time. Skimming through it I see you accusing me of hyperbole again and again, while it was you, who first used all caps to claim what a huge problem this is:
This can be REALLY consequential
only then to downplay it again when it suits you:
Are we not allowed to fix minor exploits?
and then ultimately even contradicting yourself:
I don't think it's a change that has any real consequence
So don't lecture me on hyperbole, unlike you I am logically consistent in my arguments and don't flip flop around when the winds change.

But what irritates me the most is that you continuously refuse to accept that to other people other things matter because not everyone likes what you like. By doing that you're claiming that your way to play and enjoy the game is the only valid way and that pisses me off, but I'm not going to elaborate on this any further since conversing with people like you is obviously just a waste of time as you are unable to step into someone else's shoes.
 
I'm not even going to read that line by line since it would be a waste of time. Skimming through it I see you accusing me of hyperbole again and again, while it was you, who first used all caps to claim what a huge problem this is:

only then to downplay it again when it suits you:

and then ultimately even contradicting yourself:

So don't lecture me on hyperbole, unlike you I am logically consistent in my arguments and don't flip flop around when the winds change.

But what irritates me the most is that you continuously refuse to accept that to other people other things matter because not everyone likes what you like. By doing that you're claiming that your way to play and enjoy the game is the only valid way and that pisses me off, but I'm not going to elaborate on this any further since conversing with people like you is obviously just a waste of time as you are unable to step into someone else's shoes.

A minor exploit can be REALLY consequential in certain circumstances. I don't think that's flip flopping. But if that came across as hyperbole then I apologize.

And if you actually included the context of that 3rd quote it would be obvious that I was saying that I didn't see any real negative consequence to keeping mounted melee in a city after capture. It wasn't a contradiction of the earlier quotes- I was maintaining my position that fixing the exploit was worth doing. The fact that you think I was being contradictory makes me question whether or not you're actually bothering to read what I say.

I don't understand where you're getting this idea that I only view my way of playing as valid. I disagree with you on whether or not mounted melee need to be able to move after city capture and we disagree on whether or not there's an exploit to fix but I don't see how you get from there to making these assumptions about me. It's as if you have anger about some overarching trend of changes in this game and have decided to take it out on me for some reason?
 
lots of interesting ideas. i like the split roles for ranged/melee but i fear it is too complex and invasive.

how about this:

* a city's CS is always at least as high as the garrison's
I like where this is going. There should be a baseline city strength without garrison that the garrison's CS replaces (if it's higher). Any +defense buildings/effect adds to it. Any city with a garrison + walls should be stronger than that same city missing either garrison or walls. In addition, the city should never have combat strength less than the combat strength of the garrison.
* damage to the garrison is limited to the amount it would take in a direct attack
* any excess damage is forgotten (not sure about this one)
I think this is fine, but ultimately unnecessary, since the garrison's natural CS should always be less or equal to the city's strength. It gets weird if the garrison has promotions that trigger under specific conditions (eg cover). Were you suggesting taking promotions into account?
* you can"t capture a city with a garrison (not sure about that one either)
Need to be careful about this one. @065tdsa brought up hot-swappable units before; the aggressor will need to be able to bring a garrison from 100% to 0% in one turn, or the defender will just replace it with another unit at 100%.
 
Last edited:
lots of interesting ideas. i like the split roles for ranged/melee but i fear it is too complex and invasive.

how about this:

* a city's CS is always at least as high as the garrison's

This is my favorite of the ideas, as it seems pretty simple to do, and I think its intuitive from a player standpoint. That said, i'm trying to think of the current Castle values. Do you have any weird scenarios where a Knight would make a city stronger than a wall + castle does with a normal garrison now?
 
* A city's CS is always at least as high as the garrison's
* City gets 100 hp when captured

These two should be enough to prevent easy recapturing of cities.
 
The other possible issue is that if you go steel, you could have castle like defense values (21 for longsword) without having castles.
 
The other possible issue is that if you go steel, you could have castle like defense values (21 for longsword) without having castles.

yeah but the castle gives you yields. happiness and hitpoints so it's not useless.

also, i was thinking of using the base CS of the garrison as reference. so no promotions are applied.
 
21 isn't castle like. It's only slightly better than wall with garrison (17-ish I think).
 
lots of interesting ideas. i like the split roles for ranged/melee but i fear it is too complex and invasive.

how about this:

* a city's CS is always at least as high as the garrison's
* damage to the garrison is limited to the amount it would take in a direct attack
* any excess damage is forgotten (not sure about this one)
* you can"t capture a city with a garrison (not sure about that one either)

I approve the first two. They allow a conqueror to hold the city despite its lack of defensive buiildings after conquest.

I'm not sure what the third one is about, assuming the other two are implemented. You should not have situations of excess damage there.

I'm against the fourth one, it allows for situations where it is impossible to capture a city:
  • Coastal cities can have two units garrisoned, a land unit and a naval unit. These cities will be near unconquerable, as you'd have to kill two garrisons in a single turn.
  • Cities on chokepoints, which can't be attacked more than once or twice per turn, will also be near unconquerable.
  • Turn order starts to matter a lot more, as someone facing multiple civs may rotate units before the second attacker gets the chance to kill the garrison.
  • It prevents city-states allied to the human player from conquering anything, since they are last on the turn order (barbarians aside); whoever they try to attack will have the chance to rotate units to garrison the city.
 
My first brush with the 9/9 release. Its early, but already noticing the AI being a bit more aggressive on the attack (this was Mongolia). Also very pillage focused, I saw a lot of unit push in, with immediate pillaging.

On the flip side, one of my cities was unassailable vs the horseman/skirmishers the AI had. The AI just hung around the city for a few turns, doing no damage. I was able to snipe one of the units, and then the AI pulled back.

I also saw a bit of a "wave" approach against another city with a river. The AI would move in, do some damage, and pull back before it was damaged too badly, and then would "crash" back in. This was a reasonable tactic for that situation, so I was impressed.
 
I don't know if anything has changed, so my comment is on the previous patch. The AI could really dominate me at sea, with 5-7 more units (twice as much as me) in our naval battles, but instead of going in for it, seeking to destory my much smaller army, it kind of pulls back, and let me pick him off "one by one". He should beat my fleet with ease if he would only charge. I would have to pull back all over the map if that was the case.
 
I don't know if anything has changed, so my comment is on the previous patch. The AI could really dominate me at sea, with 5-7 more units (twice as much as me) in our naval battles, but instead of going in for it, seeking to destory my much smaller army, it kind of pulls back, and let me pick him off "one by one". He should beat my fleet with ease if he would only charge. I would have to pull back all over the map if that was the case.

Same experiences here - AI has a massive navy but is very shy on actually attacking me. Let's me go in and sweep them out and they barely fight back. Using latest beta version.
 
Same experiences here - AI has a massive navy but is very shy on actually attacking me. Let's me go in and sweep them out and they barely fight back. Using latest beta version.
Can confirm, I've taken AI coastal cities by land and seen no naval counter attack. Though then again that particular city never got a dedicated counter attack by land, even when I left it mostly uncovered to go fight elsewhere, so maybe it's a different problem? Songhai (the AI in question seemed to be way too distracted with a pretty irrelevant war against an already beaten down Carthage who was surviving mostly due to geography.

In general, the AI does seem somewhat shy of committing navy. They build it in one spot and only move it from there when something to kill comes near. This works perfectly for city states or civs with only a few ports close to one another. But wide empires with many ports and big coastlines tend to fail to cover all of them. The fact that the coastal city Songhai left exposed was his capital just made that instance very blatant for me.
 
Last edited:
My feedback on the AI war behavior (always play as war monger) is that the AI should better understand when its interest are better served by defending only, and not attacking the human or fellow AI territory. In almost every instance when a war is declared, even when the declaration is by the human player, the AI will send in troops to attack in the human's territory. This is true even when the AI's victory condition is peaceful, AND they have huge defensive bonuses (DofF, GW, HeCastle, Ostrogs/Arsenals, etc.) incentivizing the AI to defend only.

In fact, I believe a common war tactic is to DoWar, then sit back and allow the AI to suicide units into the human player's defensive curtain, before the human player attempts an invasion.

It makes sense that the AI would invade when the AI declares war, or is the aggressor.

But can the peacemonger AI with peaceful VC be better taught when NOT to attack/invade, and just sit back and defend?
 
I just played a game where I took a city from the Ottomans using 4 archers and a warrior (game level is on Emperor, and this was in the early game). I couldn't see past the city, I only had sight where my units were. After taking the city, it revealed that the AI had 4 archers and a spearmen unit. Why didn't the AI commit to using those units to protect the city? It seemed like they were just sitting there for awhile, as I had encountered them earlier, but then lost track of where they went in the fog. The AI could have easily defended the city and pushed my small army away, as the AI had a much better terrain advantage as well.

Despite this, overall the combat AI seems great.
 
I just played a game where Korea tried to launch an amphibious invasion while 3/4 of my navy was on the far side of the continent, but still had 2 ironclads and a cruiser watching him embark his army.

He moved 1/2 of his entire land army into the ocean, between us with absolutely no escorts. He then moved his entire navy (same size, but underpromoted compared to mine) to go and meet mine on the opposite coastline. I told him to move his units, he declared war, I used my 2 ironclads to 1-shot some of his unescorted units and he turned tail and turtled. I had a drag-out fight with his navy and came out with 3 units left and naval supremacy over him. He couldn’t float a plank of wood after that. I don’t know if landing his army and trying to tie me down on defence was the better option, but he probably should have realized moving 10+ embarked units onto my coastline with no escort was a stupid idea.
 
I just played a game where Korea tried to launch an amphibious invasion while 3/4 of my navy was on the far side of the continent, but still had 2 ironclads and a cruiser watching him embark his army.

He moved 1/2 of his entire land army into the ocean, between us with absolutely no escorts. He then moved his entire navy (same size, but underpromoted compared to mine) to go and meet mine on the opposite coastline. I told him to move his units, he declared war, I used my 2 ironclads to 1-shot some of his unescorted units and he turned tail and turtled. I had a drag-out fight with his navy and came out with 3 units left and naval supremacy over him. He couldn’t float a plank of wood after that. I don’t know if landing his army and trying to tie me down on defence was the better option, but he probably should have realized moving 10+ embarked units onto my coastline with no escort was a stupid idea.
Wow, that's weird. Can't say I've ever seen this sort of amphibious attempt at all. And I've been in situations where the AI could have actually pulled it off, too. Them launching such an invasion at all could be promising behavior, but they really do need to get escorting right.
 
Last edited:
Back
Top Bottom