Vox Populi Diplomacy Feedback

I'd like to suggest that weaker AI be more willing to band together against the top AI. Currently, in an average game of 8 players, you'll normally have 1-3 top AI, and 5 other AI trying to butter up to the top AI. I don't like this. The bottom AI buttering up the top AI only make the strong stronger. A weak AI partnering with a stronger AI to take down a strong AI isn't desirable to me, because now the surviving strong AI just "wins more". If multiple weak AI band together to take down a stronger AI, this now leave more opportunity for a weak AI to rise up.

I think it would be better if the bottom 5 AI be more willing to band together. Either forming a coalition of 5, or several groups of 2-3. The idea is that instead of the world splitting into 2-3 groups in direct compeition, now you'll have 4-5 groups.
I think vassals should band together against their mutal master too, in certain situations.
 
Perhaps there should be a cooldown on the invest 200g to build/improve/repair a luxury. As it is now with city-states at war you can pretty much just put a unit on their luxury tiles and pillage them every turn. Some AI will spend 200g to improve it, you'll pillage it and repeat. So just like there is a blocker on the diplo-espionage quests then perhaps you just shouldn't be able to sink blackhole type gold into improving a luxury tile in a warzone.
 
It shouldn't be available if the CS is under threat.
 
In my games tha AI seems to war-happy starting from medival to me. Every few turns I get asked for a joint war, everyone is always at war with someone and backstabs are common, where stable friendship/dp clusters get broken.
 
The AI diplomacy is pretty close to perfect now with a few issues.

The AI will a bit too often do suicide wars against civ's it has zero chance of actually doing anything. The AI has become WAY better than the past, and I will often be able to keep long stretches of peace with civs who have zero to gain by turning their units into experience packs for my army.

If they are going to do wars like this, at least make them more willing to do joint wars in these cases.

Of course, if the AI has nothing to gain from a war (like if they are heading towards a culture victory), it should be avoiding dumb wars when ever it can. In fact it should be kissing up to the most power civs in these cases. I am still seeing civs throwing away a large tourism lead by attacking others civs.
 
Hi I'm back trying the best game AI ever after chess :)
some feedback / suggestion (3.9 but I think should still apply?):
should the AI be less agressive towards weak civs (ahem, me) if they are far away/have no plans to conquer them?

I'm trying to beat tradition deity which I find very often quite harder than wide. I had an ideal start with GA monopoly and extremely well defended position, good start with pyramids and mausoleum. Unfortunaltey my neighbors were natural warmongers and ended up overcoming me. Maybe with a few different choices I could have managed the defense but what I found very rough is that the whole world ended up hating me, passing sanctions etc. Possible sources of hate:

- first in tourism, but I suppose still very far from danger thresholds
- got zero wonders in classical and medieval, 3 in renaissance, louvre (is that too much already ?).
- passed world religion early
- was last on overall score (but no way to invade me)
- denonce to get better relations with others... but they also denounced my agressors

again my issue is not agressive neighbors, but the pile-on from far away civs.

My impression is that a human player actively trying to win would not have joined the hate against a distant very weak civ, which obviously aims at a culture victory but is still far away. It should rather prevent it from falling at the hands of a stronger player. Of course I would still want civs to pile on an identified run-away.

If what I say makes sense, would it be warranted to increase the weight of the overall score in the threat prioritization process? Or make distance further reduce the weight of military weakness?
 
Hi I'm back trying the best game AI ever after chess :)
some feedback / suggestion (3.9 but I think should still apply?):
should the AI be less agressive towards weak civs (ahem, me) if they are far away/have no plans to conquer them?

I'm trying to beat tradition deity which I find very often quite harder than wide. I had an ideal start with GA monopoly and extremely well defended position, good start with pyramids and mausoleum. Unfortunaltey my neighbors were natural warmongers and ended up overcoming me. Maybe with a few different choices I could have managed the defense but what I found very rough is that the whole world ended up hating me, passing sanctions etc. Possible sources of hate:

- first in tourism, but I suppose still very far from danger thresholds
- got zero wonders in classical and medieval, 3 in renaissance, louvre (is that too much already ?).
- passed world religion early
- was last on overall score (but no way to invade me)
- denonce to get better relations with others... but they also denounced my agressors

again my issue is not agressive neighbors, but the pile-on from far away civs.

My impression is that a human player actively trying to win would not have joined the hate against a distant very weak civ, which obviously aims at a culture victory but is still far away. It should rather prevent it from falling at the hands of a stronger player. Of course I would still want civs to pile on an identified run-away.

If what I say makes sense, would it be warranted to increase the weight of the overall score in the threat prioritization process? Or make distance further reduce the weight of military weakness?
If you turn on Debug Mode in MODS\(1) Community Patch\Core Files\Core Changes\DiploAIOptions.sql (last option), you can see all factors affecting the AI's Opinion.

There is existing code to limit aggression if far away, but hard to say what's causing it without more information.
 
In the late game diplomacy is not really a thing anymore if you are involved in the world affairs in any way other than sitting behind your walls.

- 1st case: you are a warmonger, everybody distrusts/hates you and that's fair.
- 2nd case: you are playing very tall, never declaring war, just rushing any non-domination victory. You can usually maintain fairly good relations with others and still have diplomatic options opened to you. The closer you get to any victory though, the more distrust is growing, that's fair.
- 3rd case: you are playing wide/a mix of tall and wide and going after a non domination victory but you are/want to be engaged in world affairs. If you are attacked, you can take cities/vassalize an AI. If an AI is too menacing, you don't mind going after it. In that case I find the diplomatic game to be frustrating and unfair. The fact that you have captured some cities and/or have vassals (from the civs who attacked you repeatedly throughout the game mind you) and the fact that you get closer to a victory type makes the whole world unwilling to do any diplomacy with you (most often it's locked behind a denounciation).

I think it is a pity since it makes the late game war/diplo very clunky. You can't broker defensive pacts anymore, you can't have opened borders (since you can't have embassies for the AI is denouncing you), and if there is a domination civ in the game they can go fairly unopposed. In my most recent game I had to fend off Songhai who was crushing civs after civs. I liberated 3 French cities, 2 Chinese Cities, 2 American cities... and yet none of those AI attitude toward me changed in the least. Sometimes, without an open border you just can't project your forces at all to were they are needed. Denounciations are way too long (70 turns I think) and by the end game 70 turns of diplomatic shut down is like a death sentence for diplomacy.

I would argue that the 3rd case is the most interesting warfare wise, so it is a pity that the diplo/war system is so stiff. If you are a warmonger, by the time you reach modern/information era you are usually 100% dominant and war is totally one sided. If you are playing tall defensive, you just turtle up until you get a victory. But in the third case you usually have a strong army but the other AI, especially warmongers are a real challenge, and that's great fun. Since modern/information era is the time warfare gets the most interesting with the most potential for combined arms, it's a bit sad the game "punishes" you for wanting to participate.

The closest emulation to a "20th century world order" (other than total domination) that the game offers is vassalization. You are attacked several times, you want it it to stop, you stomp the civ and vassalize it. Problem: to do that you have to take several cities, which makes you a warmonger in the eyes of the world (even if you give back those cities right after the end of the war). On my last game I was playing tall Venice and was harassed by my neighbors. I vassalized 4 of them through the course of the game and by the end I had only kept two cities (on a huge map), all my neighbors were given back all their cities (so I gave back around 15 cities): hence, peace and freedom of commerce was ensured on our civilized continent. Here the simulation is limited: for the AI I'm sure I am a bloodthirsty warmonger (even though I never declared war myself) while I see myself as the equivalent of the US freeing Europe in 1944 and founding NATO :crazyeye: (and my vassals loved me, my only way of having good relations with anyone in the endgame)

I don't have an easy solution to all of this but I think a good start could be the following:
- if you liberate a city for a civ, the denunciation you might have one toward the other are lifted and you can resume diplomacy on friendlier grounds.
- liberating a city should be a bigger positive boost to relations (perhaps scaling with population of said city).
- make denunciation duration scale backward with era (70 turns in early game, 40 turns in mid game, 20 turns in late game).
- taking cities in a defensive war should be less impacting than in an offensive war (it might already be the case I don't know).
- if you vassalize a civ and give back their cities after the war, it should erase or mostly erase the negative penalties for taking those cities in the first place.
 
- if you vassalize a civ and give back their cities after the war, it should erase or mostly erase the negative penalties for taking those cities in the first place.
The negative opinion comes from you having 2+ vassals (in Standard). You're considered "close to winning domination".
 
The negative opinion comes from you having 2+ vassals (in Standard). You're considered "close to winning domination".
Ok, I guessed as much and I see the reasoning here. But without vassalizing them, your neighbors usually declare war on you non stop and you can't trade/have opened borders with them. Maybe the game is inherently limited and I have to live with it but it ruins a bit the end game feeling. Or maybe I should play with no victory conditions enabled then.
 
Just wait for Casus Belli to be implemented I guess. Then you'll have reasons to declare war without being hated.
 
Ok, I guessed as much and I see the reasoning here. But without vassalizing them, your neighbors usually declare war on you non stop and you can't trade/have opened borders with them. Maybe the game is inherently limited and I have to live with it but it ruins a bit the end game feeling. Or maybe I should play with no victory conditions enabled then.
You can also disable late game hatred just because you're winning.
 
You can also choose to raze cities instead of puppeting them. It makes the warscore go up and reduce the overall needed amount of captured cities to vassalize the AI.
You should get less warmonger score that way (I think you only get warmongering from taking cities and declaring war).
 
after returning from a long break, The AI is less predictable than 3.x but theres something thats bugging me the War Dec after warmongering you cant take cities from other AI without the Warmonger Penalty i think thats out of whack but not by huge amounts, my experience is the biggest warmongering AI often complains ( have modifiers of -2000) of you taking cities on the opposite contintent when they are conquering whole civs vs weaker smaller civs that dont really care

City states are often given map making explorer quests on ocean seperated landmasses pre-caravel (although on newest 4.2.7 I havent experience that)

AI diplomacy with city states seem to be 100s of influence points at times over you - if they are a particular type of ai (warmonger, less threatening vs personality type) should be weighted correctly.

In a huge map 7 ai I took 4 cities on my landmass (chieftien) and the opposite continent 2 AI eliminated more cities and destroyed 2 civs the aggressors Brazil and Rome teamed up and went nuts on me and called me a warmonger. (ironically had complained more about my warmongering)
 
I'll try to get the guides to the code updated and done soon so y'all can give more detailed suggestions.
 
I got a scenario that has occurred twice now.
In the first game I'm friends with Siam, Persia, Spain, and Germany. We are all having a merry good time as friends until religion entered the picture. Spain, understandably was upset at Siam's religion, who had converted all of Germany to his, with Persia having his own religion, and me on a island adjacent to the continent thier on, with my own religion. Spain eventually declares war on Siam, Germany, and Siam declares war on Spain (All good in my book). What's get iffy is some time later Persia then declares war on Spain, which I thought was understandable, he's taking advantage of the conflict between the other Civs. What makes me think Spain was stupid, is not to long after this, Spain backstabs me on the basis of religious differences. Which i thought was really dumb, you are already in a 3v1 war, why make more enemies? Over the next few turns, I watch from the side lines as the 3 of them carve up Spain's corpse. I originally just chalked this up to a Spain thing, being real gung-ho about religion.
Moving onto the second game, which is occurring as we speak we got Korea, Mayans, Songhai, Indonesia and Venice (Me). Again, we are living in peaceful harmony till religion strikes. Soon after, Mayans denounce Korea(prematurely ending their DOF), and a few turns later Indonesia declares war on Korea. Fast forward a few more turns, and Korea ends are DOF over territorial disputes, which wasn't a thing before this turn, where my capital expanded to touch his 5th city. Immediately after, Mayans came up to me asking to join them in declaring war on Korea. Korea is going to get reamed for this, and it's his fault in the first place.

TLDR: The bot keeps burning bridges with friendly people when they are already in a rough spot, only furthering their demise. The Ai isn't taking into account circumstances of the board when blindly lashing out at people they are on good terms with. I feel like the bots needs to be less immediately angry at people for something they did, and wait for a more opportune time to get angry at people, and not piss everyone off at the same time

Edit: I did want to mention, i do have random personalities on, so it's probably not actually Korea's personality that is like this, but any bot doing this is pretty bad in my book
Edit 2: They held out for much longer, but korea is getting torn up by Indonesia and Maya this game
 

Attachments

  • Screenshot (276).png
    Screenshot (276).png
    4 MB · Views: 15
Last edited:
Top Bottom