Diplomacy Guide

[CountMRVHS]

I've played some Attila and Rome 2 lately, and the AI (both campaign and battle) for those games isn't too bad. Partly I'm sure the campaign AI is helped by the impassable mountains/forests that reduce expansion options.

However, even with arguably better AI in the later games, I still prefer the battles of the earlier games (MTW and modded RTW). While the later games do sometimes have an entertaining visceral quality, the bad unit cohesion and generally quick battles (especially in Attila) make it difficult for me to enjoy them.

WRT campaign and especially diplomacy, I think M2 is 'better' than RTW in that it is easier for the player to manage diplomacy thanks to the in-game cues about relationship status and trustworthiness. R2 and Attila are also relatively easy to 'game' the diplomacy system (at least on the normal difficulty level, which is how I play all my campaigns now other than MTW). You can also do some neat things in the later games; they made it easier to get client states, for example, without *quite* so much of the weirdness involved in RTW.

[Guedes]

the AI (both campaign and battle) for those games isn't too bad.

Hah! What did i said?

CA knows better that their playerbase is satisfied playing against an AI that behaves like a 5 years old kid with ADHD.

[CountMRVHS]

Have you played the newer games, and are you comparing them objectively to the older ones?

[Guedes]

If i played? What do you think?

I'm not comparing to the older titles. All the CA titles have terrible AI, no exceptions.

[CountMRVHS]

Have you found any good strategy games with better AI?

[CountMRVHS]

Periodic trolling through old forums has yielded some new data points ... potentially. I'll collect the major stuff here for your consideration.

A reddit post from 4 years ago included a couple of links to a Civ4 forum, where a user named "starlifter" had been discussing RTW and claiming to win the Imperial Campaign in 14 turns, taking many settlements through diplomacy without being at war. The posts are from 2006 or 2011, but that puts them after BI's release, and thus the info should still be relevant.

The problem is, starlifter was never very clear about how he accomplished this feat.

Here are the most relevant quotes I could find, with links:

https://forums.civfanatics.com/threa...lomacy.189973/

Do you conduct cutthroat diplomacy in your RTW Imperial Campaigns?

Does your diplomatic core "win" more for the glory of your empire than your army, in a long campaign?

For instance, have you ever gotten the AI to:
Spoiler :
- give you their Wonder provinces
- happily send you most of their gold - tens of thousands of denari!
- hand you most of their provinces --- without a fight!
- give you some of the core (original) Roman cities, whilst thanking you for your trouble?
- pay you for attacking your enemies... like rebels threatening your own land!
- submit to you as a Protectorate (even though you might not accept them as such)!
- help you attack stone walled cities like Athens -- then gotten that AI army slaughtered during the attack!


You can do these things, indeed, but not at just anytime. Timing is important, especially protectorates. Ant the AI will not give you something for nothing, e.g. you will pay somehow.

The effective diplomacy really occurs before you get involved in wars; after that, its much harder to get any major concessions out of the AI. But if you keep a good rep, you can continue to get a slow stream of gold from selling maps, and sometimes even attacks on the Rebels.

https://forums.civfanatics.com/threads/how-fast-have-you-won-rtw.189330/page-3

If you have bad luck, it can take 8 or 9 years at worst. Certain errors, particularly in diplomacy, can collapse the possibility... one need not have perfect play, but one need conduct careful, proper (e.g., think before you play) diplomacy and understand how and when to initiate war.

You negotiate. You convince them to give you some of their provinces. That part is what takes the time and planning, moreso than battle.

After you obtain as much as you can... I've obtained over 30 provinces in some games with diplomacy, then you drop the hammer and that closes the door on that sort of diplomacy for the rest of the game, and its time for conquest to get the remaining, plus Rome. Senate will never negotiate Rome away, even if you give Rome other provinces first. You must take it by force, and they have strong units and family, behind good walls... so taking Rome as Gauls in '64 for instance can require cunning and division of Purple forces.

Note that this diplomatic process is not trivial. It takes me several hours, and keeping track of who/what and replies and money with pen and paper. If you just knock on the door and say "Gimme your provinces" then you will be slapped and thrown out, and possibly ruin diplomatic contact with other factions.

Also... just getting your diplomats in position, and making contact in the "correct" order (not magic, just in a sequence of cause-effect-result, like what happens when faction A gives you a province, and that causes you to meet a Civ that you have not prepared for, and ruins your upcoming negotiations with faction B.

All this has strong strong implications for the actual battle sequence, and obtaining generals to control the rapidly expanding empire. You can choose where and exactly when you get a new general, when you get new provinces and are "short" of family members, relative to your empire state.

If you attack the Gauls, then you must fight the hard way. Diplomacy over... kaput... no workie. Ever (in that game)! Think about taking the bulk of Greece, Macedonia, Carthaginia, Gauls, Spanish, and Selucids on the 6th or 7th turn, and you will be on the right track. You need to plan your navy, as Julii... and don't go blasting north/northwest!!! You must also take a "risk" with your forces... for you will likely want many of them in other places than Italy, as not everyone is going to give you everything... you will not negotiate Sparta away from the Greeks, for instance... nor Thessalonia away from the Macedonians, for instance.

When you use diplomacy to start your empire, it is often not done by physical moe of units. It is done by negotiation. If you can keep peace, then you can "chain" reaction the contacts. For instance, if you contact the Gauls, obtain their spanish province(s), you can meet a spanish diplomat and make relations with them, and some deal(s) too. You would never have time to make and transport and move a unit all that way in early game.

Another chain is Greek... they can chain east-->west or west-->east. From the accelerated start, you immediately use your power to intimidate, and wealth to conquer in more than one thrust at the same time.

This is a good thread. Very good. What an amazing thing to learn! I read this, and figured out what starlifter was talking about, and he is right.

I have tried it using ver 1.5 (Imp/H/H), and after changing theway I think, and doing much taking of notes and planning, I have taken 51 provinces (39 by diplomacy), then Rome... in 14 turns (7 game years), as Brutii. He is right that if you plan it, you can take most of Spain for instance, by taking most gaul provinces first, then making a diplomat, and talking to Spain. If you have any war, any at all,,, if you fight any human,,, if you are attacked by any civ,,, then many, and usually all, others will not trust you. Naturally as in all war, deciet and lies are key. You must lie to the enemy, and promise to pay them huge money. And of course, you have no intent to, and when you fail, it will be war and they will try to take the provinces you bought back. The battle for Rome was full stack and massive, but fought outside the walls of Rome. Massed cavalry + generals rolling up their open field right flank from the rear turned it for me .

Now I want to try it with other civs, including nono-Roman civs. Now that I have a better idea how to conduct diplomacy, and how to extract money from I am wondering how fast it can be done with Britain, and with Scythia; I think maybe 10 to 12 years (20 to 24 turns) for those, due to the long distances, and difficulty in preventing way war or attack by an aggressor for that entire time.

Oh, and I see what you mean about the Generals, that is amazing how powerful it is to be able to choose the location you want your generals to be constructed! That was a key to being able to fight the crucial Rome battle in so short a time, with less experienced troops against those well trained SPQR troops. It was the huge number of controlled provinces versus the generals that I now understand guarantees those promotions; I have never though of that, and had never grown so fast to even observe that effect before.

https://forums.civfanatics.com/threa...al-war.168486/

Take Greece within 7 turns. 8 at the most, all Greeks. Simultaneously, take all Macedonia, except Thessalonia, and all of Eastern Selucia. Then make ATHENS your "trade center." The entire Aegean should have ports, and begin trade; gold will come in.

Do not attack the Gauls immediately. Instead, simultaneously take all their provinces except their capital, while your get the Greeks. Naturally, you should look to diplomacy, as the Julii cannot move their military fast enough physically to take the Iberian penninsula while taking Sicily, the Balkans, and Anatolia. But with diplomacy, you can set the stage and weaken their military preparation before they can field a large army. If you are cunning, you can get them to be your ally and make them PAY your to take ownership of many of their provinces, though they will have second thoughts about it in a few turns and want to fight... but by then you will be ready, and the damage has been done to them.

So GET GREECE as Julii (or Scipii, or Brutii, or Carthage, or Selecuia, or Egypt, or Gauls). As Britons, and likely as Germans & Numidians, you won't accomplish a diplomatic coup, and must slog it out like Benard Montgomery used to do in World War II, more often than not.

PS, Be sure to play Patch 1.5, if you want the "easiest" time of diplo/fight. That's because of how the province transfer is done (the favorable garrison aspect, that is).

Before you plow into your campaign, you must decide up front the sequence of contact and diplomacy. You cannot wing it. You must plan how to move your diplomats, who to avoid contact with, etc. And choreograph it to occur in a future year... typically turn 6 thru 8. And you absolutely must do it simultaneously... e.g., take Gaul, Iberia, Anatolia, Sicily, North Africa, and the Balkans. You can look small, like Sicily and Balkans... that's "OK", but you can take more if you're prepared. Once you begin combat in earnest, however, opportunities for Power Diplomacy are greatly reduced. Take it up front, or slog it out...

https://forums.civfanatics.com/threa....166844/page-6

All the factions are subject to diplomatic maneuver... while RTW AI is pretty good overall, the diplomatic AI does actually work. Its just that it does not think ahead as much as a human does, nor take certain kinds of risks (and make certain sacrifices) that can (usually) get a high reward. The AI does "nothing wrong" technically, but just as mobility with units on the battlefield is under-valued in things like AI computations and auto-resolve, so is value in the diplomatic core, which is why the Macedonians (and others) can lose control of their wonder early in the game, for instance.

The AI does not give them away... you will negotiate & pay, but not with a big battle initially. The battle will come, however, unless you can make them a Protectorate (possible in some cases, but not always probable). That's presumabley the point of diplomacy, & why they put it into the game. Just like in Real Life, when diplomacy could help create/secure empires.

Diplomacy can accelerate the game, and if you use it a lot, it can accelerate it a lot. And if you do it early, you can fight the inevitable battles necessary to win the game in diverse places on the map, simultaneously... and fund them simultaneously, instead of the slowly building steamroller. You can get about half your goal of Provinces with diplomacy, but at much higher initial cost than that of battle losses -- how you deal with the Empire's finances after that sets the tone for how/if you can hold on to your diplomatic gains (which can be quite tough to do).

https://forums.civfanatics.com/threa....172998/page-4
https://forums.civfanatics.com/threa.../#post-4677339

They just annhilate almost everything before them for the first part of the game, and are a key to rapid conquest of 40 provinces after you negotiate 6-12 provinces (or make some protectorates along the way).

After reading all of this, I made some tests, using vanilla RTW to try & see what's possible.

Based on some of the above, I figured the issue might be offering the AI tribute in exchange for cities, but then cancelling the tribute. That indeed worked - as the Scipii I was able to negotiate Syracuse and Thapsus in this manner, but I suspect the immediate cancellation made it impossible for me to continue with these types of deals - the AI wanted no part of it.

Tried again as Brutii and managed to get hold of Thermon, Larissa, Sardis, and Lilybaeum from the AI factions, usually for cash. The Greeks gave me Thermon in exchange for alliance, trade rights, and possibly map info - but no cash was needed. All the other cities cost me some money. Larissa cost me 1900/turn for around 20 turns (along with alliance, trade, & map info); Sardis cost me a one-time payment of 9,000 (plus alliance & trade); Lilybaeum cost me a little over 3,000 (plus alliance & trade). During that time I avoided war with all factions.

Not bad, I guess, but it's not exactly sweeping the map for 30 settlements...

Wambat, do you have any thoughts about this? What is this guy doing, do you think?

[Wambat]

It is mentioned more than once they are playing 1.5. That patch had a lot of bugs in diplomacy which made the AI do stupid things, like give you a million denarii in exchange for a promise to pay a similar or greater amount back as tribute. With that kind of money on turn one you could easily buy up most mediterranean provinces. I think purchased provinces also came with rather large and well equipped garrisons you could then use to expand your borders by force. So far as I recall, 1.6 put an end to that stupidity.

It is clear that some rather obvious points escaped the notice of these players. They write of not being able to trade for certain cities. Well, obviously, those were capital cities; the AI will never trade their capital. If you gifted the SPQR a province bordering Rome, and then Rome was captured by Gaul, but then SPQR recovered Rome, say by bribery, now Rome would no longer be SPQR's capital (that designation will have moved to the other province), and it should be possible to negotiate it away from them. It may have been that in 1.5 a faction's capital was still listed with other settlements in the diplomacy scroll where settlement trading would be negotiated, and this is what was causing confusion on the subject; I believe in 1.6 capital settlements are excluded from that list.

Much of the descriptions speak to things already deduced: That the AI was sensitive to the order in which these negotiations took place. Obviously these players were giving things the AI thought of value in exchange for the acquired provinces. If you give something of value to one faction, another faction which does not like that faction will now have a poor opinion of you and will not accept your proposals. They also seemed to observe that this tactic will not work after a very short number of turns; precisely because they are ruining their diplomatic reputation either by making war on factions they have implicitly promised not to; either by gifts or promising tribute; or by defaulting on their promised tribute payments, or lord knows what other possible infractions.

This discussion seems to be heavily focused on the niche pastime of pushing the envelope on the minimum number of turns to achieve victory conditions. I think this objective made it difficult, or probably irrelevant, for the players in question to develop a better theory of how diplomacy functions, or can function, over the long term. I did find it interesting they note that the tactic described does not work on the Gauls, which lends more weight to my suspicion that factions have a hard coded bias in diplomacy. If this is true it is important for modders to bear in mind.

[CountMRVHS]

I think 1.5 is the latest RTW version, with 1.6 only applying to BI. In other words, when we're playing Fourth Age, we are relying on the BI .exe patched to 1.6, but if we play RTW we are using 1.5.

If that's true, and these guys are using the same version we are, then I must be doing something different. The best I have done so far is, playing as Brutii, to get the Greeks to give me Thermon in exchange for an alliance & trade rights.

All other settlements I have gained in the early turns have been exchanged for cash, usually tribute over a number of turns. This mostly works out better than I would have thought. Promising over 1,500 denarii for 10 turns or more doesn't hurt you too badly in the early years of a vanilla RTW campaign, and of course having a settlement without having to recruit troops (or lose out on trade income) helps address the outgoing cashflow problem.

But it doesn't feel like much of diplomatic coup - I'm basically taking out a mortgage, which I do in real life anyway And the AI often doesn't seem to like it, laying siege or blockading a port shortly after they have handed over their settlement.

It would be great to figure out what the AI values.

For example, I've noticed what seems to be a pattern WRT trade rights: the AI is willing to pay for trade if they border you, but expects you to pay for trade otherwise. I can sometimes get 2,000 denarii out of trade in the former situation, but often nothing in the latter.

[athanaric]

Wambat, do you have any thoughts about this? What is this guy doing, do you think?

On one of the points: Playing FATW, I've been offered money for attacking the rebels (lol) or other factions by third parties several times. Those are just small sums though, usually not more than three digits.

[CountMRVHS]

Interesting - I've never had the AI offer me cash for attacking anyone. I have occasionally *asked* for money for such attacks, and had the AI agree, but they never initiate in my campaigns.

[athanaric]

The trick is to offer them something useless - like map information or 100 mírian - and wait for the counter-proposal.

[CountMRVHS]

Ahh, yeah, I should try that more often. I mostly save that technique for when I'm fishing for a ceasefire or protectorate.


I've got a bit of time at the moment, so off the top of my head, here is a list of things the AI might value in diplomacy, with some scattered thoughts. Let me know if I miss anything or get something wrong.


Alliance - If the AI wants it, you can usually get them to pay for an alliance. A few thousand in cash is usually the best I can do, although apparently the AI is sometimes willing to give cities as well (if early enough in the campaign, apparently).

Ceasefire - Same as above. I've found that the sort of military dominance required to get a protectorate is also useful when you want a ceasefire: beat their armies a few times, and try to bottle them up in towns. However, sometimes the circumstances surrounding the AI's appetite for peace are unclear. For example, when playing as the RK, you can get ceasefires from Harad... but when playing Harad, it seems rather difficult to get peace from the RK, even if no borders are shared. Maybe the RK/Rohan alliance plays a role here, making AI Harad feel pressured and making AI RK feel strong.

Trade Rights - AI will pay for trade rights if they border you; otherwise, they often want you to pay. If you've got the upper hand against an enemy, try the ol' Map Info offer and see if they come back with Ceasefire; request cash for it, and if accepted, make a new offer of trade rights with another cash request. Sometimes you can go from war to allies in a single negotiation this way - and make some money for your trouble.

Map Information - This seems to be the best-known moneymaking technique. I'm still paranoid about giving my map info to other factions, though, so I tend not to sell it unless it's part of a bigger deal. Doing Map Info exchanges would be a sensible thing to build reputation with AI factions (assuming there is a "reputation" system - which seems reasonable, since we can identify the AI changing the mirian value it places on certain deals over the course of a campaign).

Attack Faction - Wambat has played with this extensively, so there's probably not much to add here. You can make money by offering to attack a faction in exchange for cash, but the AI will gradually give you less & less - not sure why. Conversely, it might be possible to improve your relations with an AI faction by paying *them* to attack a common enemy, though sometimes they don't agree to it.

Military Access - The AI never seems to want this, nor do they want to give it to you, so you have to pay both ways. Maybe super early in a campaign it's possible to simply exchange Military Access agreements, since your reputation hasn't tanked yet.

Give Region - like Military Access, the AI is often reluctant to accept cities for free, so it seems not to value these too much... but on the other hand, the AI *will* demand settlements often as part of a ceasefire. If you want a settlement, the AI will often simply close negotiations unless you offer at least some cash. Not sure what to make of this - does the AI want settlements, or not? What are the circumstances under which it considers cities "worth it"?

Accept or we will Attack - I have had a tiny bit of success with this. Get a big stack right on the border of a neutral trade partner, and demand some cash with this as the threat. Without a credible show of force, though, this option seems not to impress the AI.

Become Protectorate - when the AI is asking you to become their protectorate, you can get 10,000 from them pretty reliably, which suggests that the AI values this option more highly than most anything (apart from the exorbitant amounts they sometimes demand for Ceasefire). Often they'll attack you immediately after you submit anyway. Conversely, if you want them to become your protectorate, you need to have strong military dominance and offer tens of thousands of cash.

Single Payment / Regular Tribute - The AI seems to value single payment over tribute. E.g.: a one-time payment of 10,000 is preferable to a 2-turn tribute of 5,000 each turn.

Cancel Trade Rights / Military Access / Alliance - what is the purpose of these? I mean, I know what they do, but are there any diplomatic repercussions? Is there any point? I suspect that cancelling agreements is considered a betrayal of sorts, but I'm not sure if it is considered as bad as outright backstabbing.
Also, under what circumstances might you want to cancel trade rights? I guess if you wanted to *slightly* pressure a faction, economically... but I can't imagine it would usually have that big of an impact. On the other hand, you might cancel trade rights, and then sell them back later, though the AI would probably spend less each time you pulled that trick.

[Wambat]

It has been one hell of a long time since I have played any version of RTW other than FATW. Was Rome still divided into 4 factions in 1.5? I think it must have been since the players above seem to refer to the SPQR faction. I do think there were diplomacy adjustments between RTW and RTW:BI; at least, I seem to remember it feeling that way.


The Diplomacy Guide is a little overdue for a review. I would like to see how much more this current discussion developed before publishing a revision.


@athanaric : Can you recall any details from these occasions? Did they occur early or late in the campaign? Which factions were volunteering moneys? What was your excising relationship like with the AI factions?


@CountMRVHS , taking your points:


Alliance - In my experience, the AI values settlements based on their estimation of their ability to defend them. So, farther away from their capital = less valued. Not contiguous = less valued. I have encountered scenarios where a faction will part with a settlement for 100. Several of the factions in RTW, if memory serves, are very spread out, laterally, and even non-contiguous, as with Carthage and the Greeks, so I can see getting very good rates on these settlements in diplomacy, especially if you propose before the AI can pump out a lot of units and start feeling more confident about their ability to defend their provinces.


A lot of FATW factions start out pretty compact and contiguous, while the ones that are spread out start with a pretty high military ranking. I suspect that would work against this diplomacy blitz strategy in FATW.


Ceasefire - I think the huge strength difference between RK and other mannish factions contributes to their stubbornness. The high value assigned to their units is compounded by alliances with other strong factions. I think Dale is in a similar position. Adunabar has high unit strengths, but is often diplomatically isolated. Even so, they never agreed to peace in my Rhovanion campaign. There may also be a predisposition toward/against diplomatic resolution of conflict carried over from the the RTW factions they are built upon.


Trade Rights - I have not observed the bordering vs non-bordering to be such an issue. My feeling has been that some factions intrinsically value trade more than others, and how much they favor you determines how much they will give for trade rights, but it is entirely possible all, or any combination of these factors contribute.


Map Information - My take on this has long been that the AI values map info based on how many settlements you can reveal to them multiplied by how much they like you. I feel this way because I observe that AI factions lose interest in map info later in the game, even when they really like you.


I don't notice there being any negative consequences to map trading. It may be that the AI will gun for independent settlements before you can get to them if you trade their position to the AI, but I feel like the AI already knows where settlements are and makes a bee line for them anyway.


Attack Faction - I am confident the diminishing returns on the AI paying you to attack other factions result for the unavoidable decay in faction relations over time in the recommended campaign difficulty setting. You can maintain the ability to collect payment for attacking other factions, but you need to keep the donating faction's relation level high, which means gifting them money; probably more money than you are getting back in the process.


This is what makes protectorates so valuable: You can gift them gobs of money, giving the a great opinion of you, and then get most of that money back next turn. You can then get them to give you thousands for attacking your enemies long after other factions have stopped doing so. I will try to remember to demonstrate this once I can get Harad a protectorate.


Military Access - I don't think I have ever gotten or received this without paying for it. I do not try for it very often. I can see that trying for this early in a Beorning or Dwarven campaign could make sense.


Give Region - I will try to make this section more clear. I think it is pretty well understood. The AI does not want you; and, maybe, other factions; to be in a position to grab the settlement after it is traded to them. This means zero stacks in the region; though they are fine with units being in the settlement as garrison. They much prefer the settlement being contiguous with their current territories, and the closer to their capital the better.


If the AI does not like the armies of 3rd parties to be in the region when the trade takes place, that could be a big obstacle in some of the larger and wooded provinces. I am pretty sure the AI faction will know if you have stacks in the province even if they are hidden; I am not sure if they will have the same omniscience with regard to other AI factions


Accept or we will Attack - The danger I feel with this proposal is that it would make sense if your diplomatic reputation was tarnished if you attack after a faction has paid you not to attack them, and I don't see any way of figuring out for how long you would be obliged not to attack. I would only use it in a case where you really, definitely, never intend to attack the AI faction; or you just don't care about your reputation any more.


Become Protectorate - It is curios that the experience of being attacked by the AI immediately after accepting their over-lordship is so common. Perhaps the programmers did not anticipate any self-respecting player would accept such degradation. Nevertheless, I observe the same tendency of the AI toward other AI factions: attacking them shortly after subjugating them. Silly.


Single Payment / Regular Tribute - Your observations on this point concur with my own. Money in the hand is valued above promised money. This is something that was corrected for when RTW was still getting patches. In the earliest versions of the game I can remember trading 1,000,000 denarii for a promise of 100 payments of 10,000. Obviously a bad trade for the AI.


Cancel Trade Rights / Military Access / Alliance - You would definitely want to cancel a Military Access or Alliance treaty before attacking a faction to avoid a hit to your reputation. You probably need to wait several turns after canceling the treaty to make the attack. No way to be sure how long.


Cancelling trade agreements might be helpful if you want to cozy up to a faction on one side of a conflict by severing treaties with the other side; I don't think I have ever felt the need to do so.

[CountMRVHS]

Lots of food for thought here.

I just started up a Dunland campaign, and was reminded how eager Rohan is for a ceasefire in that campaign. They approached me at their first opportunity with a ceasefire offer. Had I wanted, I could have gotten some money out of them (maybe even a settlement or two), but I was collecting an army to march on their lands, and I wanted to quickly strike a crippling blow. In the past, I've had success getting a few instances of ceasefires with Rohan over the course of the campaign as Dunland.

The reason I bring it up is that it might be a point of evidence *against* the "coalition" theory of AI diplomatic calculation. AI Rohan is in an alliance with the RK, and yet they are still very eager for a ceasefire despite what seems to be a strong diplomatic position. Yet when the player is Harad, the RK is *not* eager for ceasefire; you'd think the situation would be the same (in fact arguably it makes even less sense, given that Dunland and Rohan share a border, whereas Harad and the RK don't).

I'll take some time to digest your points and get back later.

[Wambat]

Definitely perplexing. It is hard to think of a reasonable explanation on one hand which is not countered on another when it comes to the AI and ceasefires. I do think RK is allied with the Eleves; which are, by far, the highest ranked faction at the beginning of a campaign; and Rohan is not allied with them. That may be a factor. It may also be a predisposition based on hard-coding for the vanilla faction. I notice RK will similarly opt for a ceasefire with Adunabar quite often when they are both AI. It may be very difficult to come up with a workable theory of how this is modeled.

[athanaric]

@athanaric : Can you recall any details from these occasions? Did they occur early or late in the campaign? Which factions were volunteering moneys? What was your excising relationship like with the AI factions?

Usually allied factions, for instance Beornings when I was playing Dale (Beornings asked me to attack the rebels).

Lots of food for thought here.

I just started up a Dunland campaign, and was reminded how eager Rohan is for a ceasefire in that campaign. They approached me at their first opportunity with a ceasefire offer. Had I wanted, I could have gotten some money out of them (maybe even a settlement or two), but I was collecting an army to march on their lands, and I wanted to quickly strike a crippling blow. In the past, I've had success getting a few instances of ceasefires with Rohan over the course of the campaign as Dunland.

On Normal campaign difficulty (as always), Dunland always offers me peace when I play Rohan. I accept once I've taken Dunfreca and Dol Baran, and then wait for them to take the rebel settlements in Enedwaith and build up their infrastructure, while I deal with other problems. In fact, it can sometimes be quite hard to get some factions to attack you, and this occasionally happens in that scenario, when I want to expand westwards again.

[CountMRVHS]

Interesting - I don't play Rohan very often so I wasn't aware of AI-Dunland's eagerness for peace!

I wonder if the 'faction rankings' graphs indicate the info that the AI is using when it makes its diplomatic decisions. I mean, I guess everything could be random, but wouldn't it be easy-ish to code "accept ceasefire when enemy's military ranking exceeds this threshold"?

I keep meaning (and forgetting) to look at those graphs when I'm playing, and especially when diplomatic stuff is happening. It might be good to get some data from the relative rankings.


Pipe dream, but do you suppose that with the release of Rome on Android or whatever that we might be closer to getting source code access? Then I guess we wouldn't have to wonder how all this works

[athanaric]

I'm guessing that one of the reasons is that Rohan's starting army is rather large relative to its territory, and not unexpensive. Whereas Dunland has a huge starting army, which has very cheap upkeep costs but I guess the sheer numbers is a factor, too. That might help explain the AI's stance in both scenarios.

[CountMRVHS]

A bit more weirdness WRT ceasefires - playing as Harondor, I had the RK send 2(!) units - MAA & Longbows - to besiege a coastal town right after we became allies. After beating them, I was able to get a ceasefire & trade again. (No money involved in the deal.)

Maybe this is down to the initial diplomatic stances between the factions. I would guess that RK-Harad hate each other, which could explain why it's hard for player-Harad to get a ceasefire from AI-RK. But RK and Harondor are set to be a bit more 'neutral' toward each other, I think, so the AI-RK might hate player-Harondor somewhat less.

[Wambat]

@athanaric : I am pretty sure that trespassing only affects your relations with the trespassed faction, and not what might be termed your "global reputation" if such a thing exists in RTW. If you want a reluctant party to declare war upon you I would recommend sending small single unit stacks of cheap units to trespass in Dunland's territory, and even get in the way of Dunland's stacks, so that they have to fight you in order to move. You will probably lose some of the cheap stacks, but eggs and omelets and stuff.