How to respond to AI denunciation?

[learner gamer]

Hi all, I’ve just posted the following in the vanilla strategy and tips forum - but since it relates to a VEM game, I thought it would also be worthwhile tapping into the expertise of fellow VEM gamers, to see what I could've done to better manage diplomacy in my current VEM game - and learn more about the current denunciation mechanic.

In brief, game's start saw me on a continent with Monty of the Aztecs and Harun Al-Rashid of Arabia. Very shortly after meeting the former, he denounced me...no problem I thought, that’s just Monty being Monty. I responded by denouncing him back – and Monty was subsequently denounced too by Al-Rashid.

Subsequent turns saw Monty hurl various insults my way, to which I responded “you’ll pay for this in time” to each one. I did this because (a) I did not want to appear weak by providing the alternate response (which was simply “very well” as I recall) and (b) fortunately, a CS happened to spawn at a key chokepoint. This meant that even though Monty became increasingly hostile, effectively DoWing me would mean that Monty would need to either (i) march a bunch of troops through the CS or (ii) embark a number of troops into the coast, IMHO lessening the likelihood of an early DoW.

Relations with Arabia meanwhile got off to a rocky start when I settled a very good city rather close to their capital. However, despite taking a negative diplo hit for this action, I managed to trade with Al-Rashid – and was eventually offered and accepted a DoF. I did however decline the offer of an RA – because I needed the cash rather urgently at the time for a military unit to ward off a barb unit.

Things were therefore progressing swimmingly, with me cultivating a relationship with Al-Rashid on the one hand (who indeed became friendly), whilst Monty became diplomatically isolated from both of us. Monty did try and secure me as an ally in a war against Al-Rashid, to which I replied “how dare you” – because I in fact intended to help Al-Rashid in any war, should Monty DoW, to keep my trading partner alive – at least until I met the other continent anyway.

All this however changed when my DoF with Arabia expired - although I tried renewing the DoF, they declared instead that they would wait. However, only a handful of turns later, Arabia and the Aztecs became friends / entered into a DoF. On the very next turn, Arabia then denounced me. I responded by denouncing them in turn and they went from friendly to hostile. Part of the reason given was that I had denounced a leader with whom they had a DoF in place – seemingly oblivious to the fact that my denunciation of Monty had occurred well prior to the Aztec / Arabia DoF.

The net result of all of this is that I now have two hostile leaders on my landmass. Now this won’t actually be a problem, because the war machine that I am just putting the finishing touches to will simply visit Arabia (with whom I share a border devoid of chokepoints) and then roll onto Monty – who was my original target back in the good old days when I had the DoF in place with Arabia, and Monty was the diplomatic heathen.

All of this of course brings me to my questions for you diplo gurus. Firstly, how likely is it that leaders who have previously denounced each other will subsequently enter into a DoF? I’ve seen it in a few games, but is it anything like a norm? Secondly, how do people here respond to being denounced? To put it another way, how do you decide whether to denounce the AI back? Thirdly, is there anything I could’ve done differently to help keep Al-Rashid at friendly and prevent his denunciation? FWIW, I have read Bibor’s excellent diplomacy spreadsheet, so have some idea regarding AI personalities - but perhaps I’m missing something? If BTW, you need any more information to help answer these questions, please let me know...and thanks for any and all help.

 

[Sneaks]

Denouncements fade over time, so it is incredibly likely that over a course of a game, leaders might make allies of their old enemies, and the opposite. If there are no other reasons for Monty to hate Arabia (close borders, small military, settling too much, etc), then it makes sense that Monty might see them as a solid friend against you. The easiest way to have kept Arabia friendly would have been to stay engaged in lots of trade, etc. Sometimes, leaders just change their minds though (drastically when the ally with an enemy, and all that denouncing you do really hurts).

 

[Seek]

1. Firstly, how likely is it that leaders who have previously denounced each other will subsequently enter into a DoF? I’ve seen it in a few games, but is it anything like a norm?

2. Secondly, how do people here respond to being denounced? To put it another way, how do you decide whether to denounce the AI back?

3. Thirdly, is there anything I could’ve done differently to help keep Al-Rashid at friendly and prevent his denunciation?

1. I see it occasionally also, but I have no idea what the likelihood is. Monty is easily the most "deceptive" AI, so it's likely that he convinced Harun to befriend him while planning a backstab.

2. This depends so much on circumstances that it's difficult to elaborate on - is the AI close, is it an important trading partner, is it diplomatically isolated or well-liked, etc. As a rule, I will not denounce anyone unless I'm certain to gain a positive diplo rating with someone else, and even then I often forgo denouncing. The AI sees denunciations as a big deal; as such we should as well.

3. You could have given him something you didn't need, or traded to his advantage. You also could have asked him to go to war with you against Monty - if he had accepted, that would have been a pretty large positive modifier. Keeping a good sized army is important as well, so you don't look like an easy target.

 

[Ahriman]

Somewhat on topic; do we know yet in a formal sense what the actual differences between "you'll pay for this" vs "very well" and "how dare you" vs "no thanks" responses are? Do they have any real meaning?

 

[Sneaks]

They seem to have real meaning, as they send different game event values, but what those values are is not readily apparent.

 

[learner gamer]

Thanks for your replies everyone.

Having thought about it a little more, I suspect that the denounciation was in fact initiated by Arabia (EDIT: ie. rather than Arabia denouncing me at Monty's behest), based on their propensity to forgive past transgressions and desire to enter into DoFs, according to the values in Bibor’s diplomacy spreadsheet. That said, the RNG roll which modifies these values in each game could of course have screwed with this. The clincher may well be the meaning of loyalty in Bibor’s table: some mention that a higher number indicates more loyalty, others less...so I’ve added a question to Bibor’s stickied thread in Civ 5’s strategy and tips forum. If a higher number indicates more disloyalty, the denouncement would actually be a classic Al-Rashid backstab, if lower it’s likely the RNG has obscured whether Monty or Al-Rashid’s the initiator.

@Ahriman: That was actually one of the questions I was hoping would get answered - and indeed, someone has asked the exact same question in the main strategy and tips thread. Whilst, like Sneaks, I beleive these replies do have different meanings to the AI, like you, I'll be hoping someone can provide clarification on what they are.

 

[Txurce]

They seem to have real meaning, as they send different game event values, but what those values are is not readily apparent.

I have never read any speculation, even based on limited anecdotal evidence. Aggressive answers don't seem to provoke, and deferential ones don't seem to mollify. I'm surprised that they even have different game event values; I had assumed it was unfinished business.

 

[Thalassicus]

Seems like that 2-reply system was hoping to do an intimidate/back-off sort of mechanic like Bioware's paragon/renegade dialogue system. It would be interesting... ruthless leaders might be impressed by the intimidation choice and overlook the infraction entirely, or if your military situation is weak they might get even more pissed off instead. At least that seems to be the idea... the options don't appear to actually have any effect.

 

[tlaurila]

Seems like that 2-reply system was hoping to do an intimidate/back-off sort of mechanic... the options don't appear to actually have any effect.

It would be exceedingly cool if a "Revoke that, or I will declare war" option was there, and then the AI would consider relatively military strengths and conflicts already committed to to decide whether it's what it wants. So, force peace with a CS or Civ recently declared war on, or un-settle a city built too close (with mechanics that turns city back to settler), or else it'll be war.

It might even help the player to backtrack if (s)he unintentionally provokes an AI by settling. And allows the player to effectively intimidate AIs. Right now an AI can get pissed at the player and that can deter the player, but you can't really reciprocate.

 

[Jorlem]

It would be exceedingly cool if a "Revoke that, or I will declare war" option was there, and then the AI would consider relatively military strengths and conflicts already committed to to decide whether it's what it wants. So, force peace with a CS or Civ recently declared war on, or un-settle a city built too close (with mechanics that turns city back to settler), or else it'll be war.

It might even help the player to backtrack if (s)he unintentionally provokes an AI by settling. And allows the player to effectively intimidate AIs. Right now an AI can get pissed at the player and that can deter the player, but you can't really reciprocate.

On that subject, it would also be nice to have a button that could ask the AI to stop gathering troops on your border. The AI is capable of basically asking you to move your troops away from them, or get it over with and declare war, but the player can't ask the same of the AI, as far as I know.

 

[tlaurila]

On that subject, it would also be nice to have a button that could ask the AI to stop gathering troops on your border. The AI is capable of basically asking you to move your troops away from them, or get it over with and declare war, but the player can't ask the same of the AI, as far as I know.

In a way it would, but then again it probably wouldn't. When the AI asks you to move your troops and you say "oh, don't worry about it", what it means is that you get a diplo reputation hit if you then attack. It's a counterbalance for the player being able to set up his units to spring on the AI from optimal positioning along the border. Something the AI can't do anyway.

If the player could go about asking the AI to "remove troops" as he pleases, whenever you see a DoW coming, you could use that to effect a severe diplo penalty to your enemy. Pretty much regardless how the troops actually lie. That would potentially make the game significantly easier for the peacenik player, making it much easier to prevent AI alliances from picking up your weak, ripe behind for a whooping and splitting the spoils. Something that is not desireable.

 

[Dunkah]

Jorlem's idea is excellent. (Though I doubt that it can be implemented at this time).

As soon as you initiate something that could piss off another leader the denouncing page pops up with a warning from the other Civ/s.

"Hey I see you are about to found a city on my border. I will/may denounce you if go through with this."

Then you have the option to proceed or ignore. If you choose not to take the hit you don't found city/ attack the CS etc... (Don't even need the option to respawn the settler because it never happens in the first place). It may just be that the Civ decides not to go through with the denouncement due to the hit itself.

Perhaps even have any other leaders on that screen that will jump on the band wagon if you proceed. (In the case of declaring war on a CS etc).

So in essence have Denunciations tied to a specific action.

DOW's/DOF's.
Settling Cities.
Buying tiles.
Attacking CS's.
Destroying Civs.
Liberating Civ's.
Trading with enemies.
Having too many troops on the border.
Dropping/ Building nukes.
Denounce other Civs.

It would represent the planning and dialog that would go on prior to implementing said action, and the ability to recognize that what you are about to do could piss off other Civs. Currently there is no way to know if you are too close/ have too many troops on the border to a civ, or weather they believe the territory you are about to purchase is considered sovriegn.

 

[Jorlem]

You know, instead of having the other Civ talk to you in order to try to warn you off, why not have the warning come through the built in advisers? I don't know nearly enough to know if this is actually the case, but I imagine it would be easier to add a pop up message from an appropriate adviser then to build it in to the AI. I think what loss there might be from not having a leader come onscreen and tell you to stop what you are doing would be balanced with the novelty of the advisers actually giving useful advice.

For detecting if you are about to settle in an area too close to another Civ, could it be tied into VEM's resource allocation generator? IIRC, VEM designates certain areas of land as belonging to each Civ, and then places resources so each Civ has potentially the same value, barring wars and aggressive settling. Maybe a trigger could be set up, so that if you moved a settler onto land that VEM had designated as belonging to another Civ, an adviser could pop up, and say something along the lines of
"We have just received an urgent message from [Leader], the ruler of [Civ]. He/She states that a group of settlers from our land have been spotted on territory that they have always considered as belonging to them, and asks that you ensure that they do not remain, or there shall be consequences."

Historically, most of the things on that list were things that were handled by advisers and agents of the rulers, instead of the ruler needing to speak face to face with his opposite number. One of the things I liked about Civ 4 was that the info screens were presented as being reports prepared by your various advisers, which added a nice bit of immersion. Now though, I don't think they really do anything of use at all.

 

[markusbeutel]

Seems like that 2-reply system was hoping to do an intimidate/back-off sort of mechanic like Bioware's paragon/renegade dialogue system. It would be interesting... ruthless leaders might be impressed by the intimidation choice and overlook the infraction entirely, or if your military situation is weak they might get even more pissed off instead. At least that seems to be the idea... the options don't appear to actually have any effect.

This is from DiscussionDialog.lua It seems that these responses were never supposed to have any effect at all on diplomatic relations, at least that's what I'm led to believe fluff represents. They seem more like a basic role-playing response to get the player engaged in a game - albeit without any consequences whatsoever.

-- Human did something mean, AI responded, and human responds in turn with fluff
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_BLANK_DISCUSSION_MEAN_HUMAN) then
strButton1Text = Locale.ConvertTextKey( "TXT_KEY_DIPLO_DISCUSS_MESSAGE_SORRY" );
strButton2Text = Locale.ConvertTextKey( "TXT_KEY_DIPLO_DISCUSS_MESSAGE_DEAL" );
bHideBackButton = true;

-- AI did something mean, and human responds in turn with fluff
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_BLANK_DISCUSSION_MEAN_AI) then
strButton1Text = Locale.ConvertTextKey( "TXT_KEY_DIPLO_DISCUSS_MESSAGE_OKAY" );
strButton2Text = Locale.ConvertTextKey( "TXT_KEY_DIPLO_DISCUSS_MESSAGE_YOULL_PAY" );
bHideBackButton = true;

 

[Sneaks]

The code actually sends different Diploresponses game events. Whether those game events do different things is another question altogether. You simply posted the UI display end of the code, and not the part that does anything.

-- AI seriously warning us about expansion - we tell him we mean no harm
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_DISCUSS_AGGRESSIVE_MILITARY_WARNING) then
Game.DoFromUIDiploEvent( FromUIDiploEventTypes.FROM_UI_DIPLO_EVENT_AGGRESSIVE_MILITARY_WARNING_RESPONSE, g_iAIPlayer, iButtonID, 0 );

-- AI warning us about expansion - we apologize
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_DISCUSS_YOU_EXPANSION_WARNING) then
Game.DoFromUIDiploEvent( FromUIDiploEventTypes.FROM_UI_DIPLO_EVENT_EXPANSION_WARNING_RESPONSE, g_iAIPlayer, iButtonID, 0 );

 

[markusbeutel]

The code actually sends different Diploresponses game events. Whether those game events do different things is another question altogether. You simply posted the UI display end of the code, and not the part that does anything.

-- AI seriously warning us about expansion - we tell him we mean no harm
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_DISCUSS_AGGRESSIVE_MILITARY_WARNING) then
Game.DoFromUIDiploEvent( FromUIDiploEventTypes.FROM_UI_DIPLO_EVENT_AGGRESSIVE_MILITARY_WARNING_RESPONSE, g_iAIPlayer, iButtonID, 0 );

-- AI warning us about expansion - we apologize
elseif (g_DiploUIState == DiploUIStateTypes.DIPLO_UI_STATE_DISCUSS_YOU_EXPANSION_WARNING) then
Game.DoFromUIDiploEvent( FromUIDiploEventTypes.FROM_UI_DIPLO_EVENT_EXPANSION_WARNING_RESPONSE, g_iAIPlayer, iButtonID, 0 );

I was focusing on the fluff part more than the actual code - as I don't see any "we apologize with fluff" for that specific event regarding military warnings.

 

[Sneaks]

The fluff dialog only occurs for stuff like "now the vermin are breeding!", "I see that you have become friends with X", "your army looks weak" etc. Those exist only to give the player a chance to see the AI's opinion. As for military warnings, there is actual game event code for things like armies on borders, settling close, attacking CS, buying plots, etc.

 

[Dunkah]

You know, instead of having the other Civ talk to you in order to try to warn you off, why not have the warning come through the built in advisers?

Or even better perhaps an icon that appears on the Confirm Button or perhaps three different colored Confirm Buttons (Red, Yellow, Green) that indicates you may piss someone off if you wish to investigate.

 

[Thalassicus]

Currently there is no way to know if you are too close/ have too many troops on the border to a civ, or weather they believe the territory you are about to purchase is considered sovereign.

Actually, the AI here appears to be very simple: it just counts the tiles to see if the target is closer to your capital or theirs. So if there's 20 tiles between your capitals and you settle 11 tiles from yours that causes anger, but 9 tiles from your capital is okay. It's a very hard line. I noticed this pattern from watching the effect placing cities has in various games.

 

[black213]

What are the AI changes of version 112?