• We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.

Cagliostro

Charlatan or Mystic?
86 Badges
Apr 30, 2002
3.636
456
  • Stellaris: Synthetic Dawn
  • Europa Universalis IV: Dharma
  • Stellaris: Distant Stars
  • Crusader Kings II: Holy Fury Pre-order
  • Europa Universalis IV: Rule Britannia
  • Stellaris: Apocalypse
  • Stellaris: Humanoids Species Pack
  • Hearts of Iron IV: Expansion Pass
  • Crusader Kings II: Jade Dragon
  • Europa Universalis IV: Cradle of Civilization
  • Age of Wonders III
  • Shadowrun Returns
  • Hearts of Iron IV: Death or Dishonor
  • Europa Universalis IV: Third Rome
  • Europa Universalis IV: Mandate of Heaven
  • Stellaris - Path to Destruction bundle
  • Crusader Kings II: Monks and Mystics
  • Stellaris: Leviathans Story Pack
  • Stellaris: Digital Anniversary Edition
  • Tyranny: Archon Edition
  • Europa Universalis IV: Rights of Man
  • Crusader Kings II: Reapers Due
  • Stellaris: Ancient Relics
  • Victoria 3 Sign Up
  • Stellaris: Nemesis
  • Stellaris: Necroids
  • Europa Universalis 4: Emperor
  • Crusader Kings III: Royal Edition
  • Crusader Kings III
  • Imperator: Rome - Magna Graecia
  • Stellaris: Federations
  • Stellaris: Lithoids
  • Age of Wonders: Planetfall
  • Hearts of Iron IV: Cadet
  • Imperator: Rome Sign Up
  • VtM - Bloodlines 2 Blood Moon Edition
  • Imperator: Rome
  • Imperator: Rome Deluxe Edition
  • Europa Universalis IV: Golden Century
  • Crusader Kings II: Holy Fury
  • Stellaris: Megacorp
  • Shadowrun: Hong Kong
  • Shadowrun: Dragonfall
  • Divine Wind
  • Europa Universalis: Rome
  • Victoria: Revolutions
  • Europa Universalis IV: Res Publica
  • Europa Universalis III Complete
  • Magicka
  • Europa Universalis IV: Call to arms event
Just installed 1.03, first game, playing as Gisulf, Count of Salerno. I didn't repeat the non-significant history.

July 25, 1072: Murtaza King of Zirid accepted peace with Aimone Bishop of Benevento on the following terms: White peace
July 27, 1072: Kingdom of Zirid declared war on Bishopric of Benevento.
September 24, 1072: Murtaza King of Zirid accepted peace with Aimone Bishop of Benevento on the following terms: White peace
September 27, 1072: Kingdom of Zirid declared war on Bishopric of Benevento.
November 2, 1072: Murtaza King of Zirid accepted peace with Aimone Bishop of Benevento on the following terms: White peace
November 7, 1072: Kingdom of Zirid declared war on Bishopric of Benevento.
December 25, 1072: Murtaza King of Zirid accepted peace with Aimone Bishop of Benevento on the following terms: White peace
December 27, 1072: Kingdom of Zirid declared war on Bishopric of Benevento.

I noticed other Moslems doing that to Christians in the middle, but I'm not going to bother to rewrite every pair.
 
Upvote 0
If you look in the 1.03 discussion thread, someone posts this happening for Mongols. MrT says this should happen with the Horde but I doesn't say about anything else.
 
It tends to happen when X is a vassal of Y and Z is at war with X. Y then keeps declaring war on Z, making peace, then declaring war again because Z is still at war with their vassal.

There should really be a way to make peace for your entire side in a war.
 
1.03a - situation is very bad.
Played a new game on hard/aggresive and within about 20 years the game has become unplayable due to half the map seeming to be stuck on the war/peace loop.

At least with 1.03a the duke's granting revoking titles seems to be fixed but between that and this war/peace loop it really makes me wonder if the patch spent more than 5 minutes in testing before it was released. These 2 very obvious problems would have been very easy to spot if anybody had actualy played the game.

Myros
 
I saw the same thing happen in my Brittany game today. In an 18-month period, both Granada and Badajoz declared war on, signed white peaces with, and redeclared war on the Duchy of Braganza at least 5 times each. I'm fairly certain Cordoba was also involved. And the Crusades were not even active yet.

It ended after that only because Badajoz stuck around long enough to actually conquer Braganza. It's now taken out (a generation later) Braganza's overlord, Spanish Galicia, as well (which sucks as I'd lined myself up as the only heir to it just before they conquered it.)
 
Last edited:
This seems to be particularly bad in the Spains on Very Hard/Furious -- I've now started about half a dozen games as various Spanish dukes or kings, and due to the alliances which inevitably form, the Christians are almost invariably overrun within ten years or so... there's almost no ability to resist, given how much better the Muslims are than the Christians at game start.
 
Maybe this is related to the new change when AI declares war on the vassals of a duke / king it's in war with. Say, Zirid is at war with Apulia and starts thinking that dowing Benevento is a good idea. Dows it. Then it immediately gets a peace offer from Benevento, who doesn't like the idea and for some reason (the problem) Zirid accepts this peace. After that it again starts to think that dowing Benevento would be good, as it's at war with Apulia. And ta-dah: Loop.

It's painfully obvious in the other example of Braganza. Badajoz is at war with Spanish Galicia, at least I presume so as Braganza was reconquered by it. Badajoz sees Braganza in the way to Spanish Galicia's provs and sees that Braganza is a vassal of Spanish Galicia. It dows Braganza and soon peaces them. The problem isn't the AI declaring war, but the peacing.

At least I see things this way. Sorry for two examples, but I wanted to make it clear. :)
 
In my actual game, normal-normal, this happened to me 1 time with emirate of assassin (I'm count of adana); but nver after, and also I didn't see so often in other countries.
 
The_Hawk said:
This seems to be particularly bad in the Spains on Very Hard/Furious -- I've now started about half a dozen games as various Spanish dukes or kings, and due to the alliances which inevitably form, the Christians are almost invariably overrun within ten years or so... there's almost no ability to resist, given how much better the Muslims are than the Christians at game start.
Well, Very Hard/Furious is not supposed to be easy, and the only reason it was trivial to conquer Iberia from the muslims in 1.01 and 1.02 was that the AI was incapable of utilizing its vassal forces, allowing the human player to pick vassals off one by one without opposition. One can argue whether you should meet your end because of the muslim alliances or not, but you are practically begging for them to come kill you. It is certainly a deadly combination to choose both the nastiest play-settings AND one of the nastiest starting locations..

The best advice, and it is not all that good, as you cannot count on it working each time, is to ally with the French king and one (or two) of the other Spanish kings and hope they will help you.... And play on a lower difficulty and aggresivenes setting.

The Muslims will fight the Christians as often as possible in 1.03 - assuming they are stronger and feel like it (and they almost always feel like it, no matter the aggressiveness setting - because they are so very, very, strong). One of the complaints about 1.01 and 1.02 was that the AI was not aggressive enough. That has been addressed at least for religious enemies.... It is tough on the Spanish kingdoms and Georgia, though.


As such, I doubt it counts as a bug even though it is ugly to see the constant peace/dow.

...That said, reevaluating the scenario tech setup in Iberia might be an idea for 1.04. :)
 
Last edited:
The Braganza example was also on normal/normal ;) It happened about six times, and would probably have continued if Braganza had not been conquered. Interestingly, I do not think (I'll check though) that they were at war with Spanish Galicia, at least toward the end, as SG survived for about another 40 years with no change in its non-Duchy of Braganza territories.

It's not just the Muslims doing it either - upon sifting through the history files, I have similar DOWs by England and France against the Sheikdom of Annaba several times, as well as some of their own rebellious vassals (immediately after white peaces, almost always redclaring within 3 days).

Perhaps something should be done to make the AI countries respect truces, at least for a certain period of time.
 
Last edited:
I think a smarter fix would be to make AI consider more the wisdom of making peace for 100 cash or even white peace, when it's besieging the only prov of it's enemy and all enemy troops are killed.
 
Well, that too, of course. But whatever the logic - even if it was imposed by event - the AI countries should not immediately declare war with the same country they have just made peace with. With the possible exception of the Mongol hordes, because they're *supposed* to be fast moving and fierce. But with everyone doing it, it's making the geopolitical situation far more fluid than it should be.
 
Quote (missing due to my dumbness) of the_hawk post (with a phrase taken from Myros)

The situation is very good, not very bad ...everyone was aking for a stronger AI, and now people's complaining about that, incredible :confused:
You should post this on the general forum, not on the bug forum.
 
Last edited:
Did you even read the bug report??
When at least 15 AIs in my current game are constantly making peace and declaring war 1 second later OF COURSE ITS A BUG. Nothing to do with a strong AI, its a bug in whatever routine is checked to see whether peace should be asked for and/or accepted ..obviously the war routine is still in effect because as soon as peace is accepted the conditions for war are still met because they decalre war again right away.

An AI that cant decide wether to be at war or peace isnt "stronger" its dumber ;)

Myros
 
Myros said:
Did you even read the bug report??
When at least 15 AIs in my current game are constantly making peace and declaring war 1 second later OF COURSE ITS A BUG. Nothing to do with a strong AI, its a bug in whatever routine is checked to see whether peace should be asked for and/or accepted ..obviously the war routine is still in effect because as soon as peace is accepted the conditions for war are still met because they decalre war again right away.

An AI that cant decide wether to be at war or peace isnt "stronger" its dumber ;)

Myros

Apologies :eek:o I wanted to reply to the_hawk, not to you, but don't know why a I quoted your post (a bug in MY MIND)?
Sorry again. ;)
 
Hey, if we're going to bother having a truce condition for players, AIs should respect it too - I've seen plenty of ideas (most of which would even enforce more historic play) shot down on the Paradox boards over the years under the name of having the player and AI play by the same rules. So why not here?

Of course, there may be a problem of AIs making premature peace (although we should note here that in CK peace can happen by event too, so the AI isn't always making the decision there), but it seems to me that immediately redeclaring war could be dealt with seperately from that, perhaps simply by adding in a modifier to whatever equation the AI uses to determine if war is a good idea that will encourage it to respect truce conditions when they exist.
 
PE, lucaluca, my apologies; I was unclear. I'm not complaining about smart, capable AI; I love smart, capable AI. That's not what this thread is about. This thread is about the penchant of the AI under 1.03 to make peace and then immediately break it.

My theory (which I failed to state) is that this is related to alliances somehow -- that failing to make peace with the right people in the right order will drag folks back into war. This is notable in Iberia, because by and large the Spanish kings all ally with one another, and the Emirs all ally with one another as well.

Honestly, the AI routines now seem highly convoluted, to the point that any declaration of war in a mutually hostile area such as Spain leads to an explosion of wars involving virtually everybody who is or could be allied. Example: As Flanders, I warred on Barcelona, who had no allies of significance. For some reason this caused Cordoba to declare war on Navarre, who was my ally, presumably to stop me from making war in Spain. The solution? Simply not support my ally, which doesn't seem to have any negative effect for me. This just doesn't seem like the way things are supposed to work.

AI routines are beyond the ken of mortal men such as myself; but similar to the example above, while I'm sure there's some reason the code tells the AI to make peace and declare war repeatedly, it doesn't seem like the way things are supposed to work. Peace means peace; if the AI doesn't want peace, it shouldn't make it. Do you still take a BB hit for declaring war while in the target province? Because if you do, the AI is probably wrecking itself by doing this...
 
The_Hawk said:
PE, lucaluca, my apologies; I was unclear. I'm not complaining about smart, capable AI; I love smart, capable AI. That's not what this thread is about. This thread is about the penchant of the AI under 1.03 to make peace and then immediately break it.
And I probably overstated my case. Nevertheless, it would be interesting to see if somebody has the same experience with same-religion wars. As far as I have seen, it is mainly a case of wars between religious enemies - and they are governed by special rules.

My theory (which I failed to state) is that this is related to alliances somehow -- that failing to make peace with the right people in the right order will drag folks back into war. This is notable in Iberia, because by and large the Spanish kings all ally with one another, and the Emirs all ally with one another as well.
And my theory is that it is because the AI uses one set of rules for when to make peace and another for when to declare war. The point being that - at least in the Iberian case - there is very little reason to make peace in the first place, and every reason to declare war upon weaker religious enemies (next to no ill consequences) - no matter whether they are weaker because somebody else or yourself has killed off most of their manpower.

I.e. it makes peace based on something like "long war, the other party offers his entire treasury, how nice, perhaps I should get a bit of R&R" - and then, when regiments are demobilized makes war based on "how nice, I have an army ten times as large as my neighbour, who is a religious enemy, perhaps I should put him out of his misery".

A split personality of sorts. :D

Honestly, the AI routines now seem highly convoluted, to the point that any declaration of war in a mutually hostile area such as Spain leads to an explosion of wars involving virtually everybody who is or could be allied. Example: As Flanders, I warred on Barcelona, who had no allies of significance. For some reason this caused Cordoba to declare war on Navarre, who was my ally, presumably to stop me from making war in Spain. The solution? Simply not support my ally, which doesn't seem to have any negative effect for me. This just doesn't seem like the way things are supposed to work.
I would guess that Corboda attacked Navarre because Navarre was weak, not because of your actions (unless your actions weakened Navarre). Corboda often attacks Navarre.

AI routines are beyond the ken of mortal men such as myself; but similar to the example above, while I'm sure there's some reason the code tells the AI to make peace and declare war repeatedly, it doesn't seem like the way things are supposed to work. Peace means peace; if the AI doesn't want peace, it shouldn't make it. Do you still take a BB hit for declaring war while in the target province? Because if you do, the AI is probably wrecking itself by doing this...
There is NO BB for sneak attacks upon religious enemies anymore. There is, however, a small prestige hit for breaking a truce. The Muslim AI dynasties are certainly not wrecking themselves by fighting an on/off Jihad vs. the Christians - just witness how they usually manage to overrun Spain within a generation or two.

I quite agree that it would be more aesthetically pleasing if the AI did not make peace in the first place, when it is just going to declare war again (though it does earn some money this way in exactly the way human players drained AI muslims/pagans of money in 1.01 and 1.02 - it only gets smaller amounts because people actually use their money), but that does not necessarily make its declarations of war insane (as the title proclaims).

What is annoying is the cases where nobody ever gains anything and keep declaring war, failing to attack, and make white peace within a few days, as in Cagliostro's original example. That is actually damaging to the dynasties involved, as they may get to pay a minor prestige hit and possibly some monthly army upkeep.
 
Last edited:
I think the right solution could be: make the spanish christians stronger, or the spanish muslims weaker.
I don't see Byzantium falling immediately to Seljuks, or this peace-war loop, just because there is more equilibrium there.
 
Peter Ebbesen said:
I would guess that Corboda attacked Navarre because Navarre was weak, not because of your actions (unless your actions weakened Navarre). Corboda often attacks Navarre.

As would I; except that I did this twice (having saved just before my troops set out from Flanders, and my first war, where I went to Navarre's aid, having turned against me.) The situation was identical except, perhaps, for the day that I declared war, but in both cases, the same day that DoW'ed Barcelona, Cordoba attacked Navarre. This could be coincidence, but the two events seem related to me.

Anyway, this is really more a commentary on the AI generally than on the specific bug referenced above.

Peter Ebbesen said:
What is annoying is the cases where nobody ever gains anything and keep declaring war, failing to attack, and make white peace within a few days, as in Cagliostro's original example.

In my experience that's generally what's happening. The more I play, the more it seems like something is causing the AI to be dragged into wars that it doesn't really want to be in, and so it gets out of them, only to have the condition immediately trigger a redeclaration. What's unclear is whether alliances, vassalage, or this penchant for declaring war on people between the attacker and the target I hear about is to blame.