AI vs human: same attitude/reputation rules?

[Aggie]

Does de AI have the same problems with a reputation hit and attitude problems as the human players?

Example: does the AI lose the ability to trade gpt with another AI after canceling a military alliance (one-sided) or after a ROP-rape?

I always assumed that this is the case, but I would very much like to have this confirmed.

 

[Speaker]

I would highly doubt it. The AI always quits military alliances with me early by signing peace before the alliance is over, so I have to assume that they do that to the other AI as well. Additionally, the AI Civs always seem to have RoP agreements with everyone, so anytime war breaks out, they are probably breaking the agreement. Also, since the AI really only declares war via the sneak attack, they should get a rep hit for that as well. With that all said, its not really possible to track trades among the AI, but since they all usually stay in packs techwise, they must be trading amongst each other. My answer is very speculative, but seems to make sense, at least to me.

 

[DaveMcW]

I've noticed it's much easier to get an alliance against an AI that has broken deals.

 

[Grille]

Yeah, same here.

Generally, Speaker's thoughts cover mine, but...
there's indeed ways the ai declares war on you (or other ai) and not getting a rep hit:
You refuse a demand and they declare war w/o units in enemies territory etc, by signing up MA against you or MPP triggering.

 

[Aggie]

Originally posted by DaveMcW
I've noticed it's much easier to get an alliance against an AI that has broken deals.

That's something I noticed as well. But did people notice if regular deal-breakers tend to fall behind in tech?

I really don't think that the AI civs 'stay in packs tech wise'. I normally see one to three civs ahead of in tech and the rest at bit or far behind... But I didn't notice whether this had some correlation with reputation and attitude.

 

[Speaker]

Originally posted by Grille
Yeah, same here.

Generally, Speaker's thoughts cover mine, but...
there's indeed ways the ai declares war on you (or other ai) and not getting a rep hit:
You refuse a demand and they declare war w/o units in enemies territory etc, by signing up MA against you or MPP triggering.

I agree, but most of the time they sneak attack.

 

[Grille]

Originally posted by Speaker


I agree, but most of the time they sneak attack.

Yes, makes me, too, guessing that ai is not affected like human player by rep hits.

I dunno if tests are possible for confirmation to some degree,
here some inspirations:
small map, give civs different techs/resources, no money but commerce productive tiles -> chance for gpt deals, not all communications available for everyone, start war that will cause a rep hit for certain ai, donate communication to other civ(s), watch treasuries/calculate ai's income if possible etc, eventually put on different ai trade rates
(Unfortunately, I haven't just played too much civ3 -who has?- to make a test by now and must stay on the speculative road... sorry!)

 

[Aggie]

Originally posted by Grille


Yes, makes me, too, guessing that ai is not affected like human player by rep hits.

I dunno if tests are possible for confirmation to some degree,
here some inspirations:
small map, give civs different techs/resources, no money but commerce productive tiles -> chance for gpt deals, not all communications available for everyone, start war that will cause a rep hit for certain ai, donate communication to other civ(s), watch treasuries/calculate ai's income if possible etc, eventually put on different ai trade rates
(Unfortunately, I haven't just played too much civ3 -who has?- to make a test by now and must stay on the speculative road... sorry!)

I would like to try it myself, but it seems like a difficult thing to test. We're talking about at least 40 turns here that you have to play in different styles - one that causes a civ a rep hit and another that doesn't-. And how do we know that other factors, like losing important land, don't interfere with the conclusion? I've got the time, but not a clue how to do this properly. This is not the same as testing what a rep hit does to YOUR trading with the AI.

That's why I asked the question, as I hoped that at least one of us at CFC knew how this works.

 

[Siriusnight]

I've always found that after a war it's a very long time after peace is declared that they will sign a MPP. It's alot easier to see how nations get along by trying fewer civs. I should imagine that there is some kind of reason that some civs go to war with each other regularly (perhaps a poorer rep than another neighbour?)

 

[Grille]

I'm not so used to the editor.
(Excuse me & forget it, if here are dumb assumptions.)

But think of this scenario (& and do think it's very hard to test).

put ai's on (one tile?) islands (prevent spreading) - IMO, easy to keep track on
- harbor owned, trade must be possible (but no shipping allowed - maybe expensive galleys or something), maybe bonus food res to let them be able to sign gpt deals due to size (as small ai empires don't like gpt, you may put on an improved second tile with lux - or put lux resource in water?)
-due to working tiles, their commerce output could be predictable - dunno if difficulty level makes a botch of calcul?). This calculation & treasury observation may put a light on gpt deals. Constant investigation of city may provide info about lux/tax sliders.
-human player with lots of stuff travels to ais -> donate comm, different tech
-well, if there's something going on between 2 ai, try to sign MA, see what happens after that war (I hope they will ever sign peace!); therefor, you may donate different techs again
note: for MMP triggering - you may let them have a (no upkeep(?)-)unit elsewhere, hope they don't disband it
-a third ai without any comm, contacted after war by human will be informed about rep hits(?) - they might have trade only with the unblamable

The problem here is: If there's no gpt trade after the war with a certain civ, we won't know for sure the cause. Maybe there's no care about rep but also no care about trades then...

Anyway, this test (hmm... or a useful test - if I was totally wrong about anything) would eat a lot of time. I'm very corious about this topic & hope that others at least post some information/observation/conclusion.

Originally posted by Aggie
...I hoped that at least one of us at CFC knew how this works.