Ok so the AI cares Nothing about its reputation.

Last night I had a RoP agreement with England. I also was payig 5 GPT for Gems. We were Polite and had never fought a war from the beginning of the game.

(Skip the next three paragraphs if you don't care about the background)
Then the evil Romans attacked me, and I have to admit my advisors had warned me they were evil. So I sent my forces North and began capturing Roman Cities. (Despite being evil it was a poorly planned attack by the Romans, they had a quick easy capture on an isolated city but it went down hill fast after that for Rome).

About four turns in while I was fighting the Romans the Egyptions started a war with me. (Not to shocking here, though I played a nice loving Republic, whenever I needed a resource it always seemed to pop just inside the Egyption Borders so instead of trading for the resourse I'd just send my armies to take over the closest cities, I did it to them three times so I couldn't figure out why they kept giving me peace).

Anyway about four turns in England landed five cavalry on a little island I had neglected. Despite the RoP you could tell this wasn't good since there was nothing on the island but my little city.

Anyway I made peace with the Romans mostly because they didn't have anything left that I wanted at this point and a three front war seemed a bit much. And after losing a pile of workers to the Egyptions I've taken out most of their means of producing modern armies and have began taking their cities and now my armies are beginning to look at the lovely gems we used to have to pay England for and thinking about a long sea invasion of England.

(My actual question is here)
Anyway the actual question is doesn't England take a major Reputation hit for breaking a trade agreement and a RoP agreement with me? And if not why?

BTW I play Civ III Vanilla if that makes any difference.

Whythat said:

(My actual question is here)
Anyway the actual question is doesn't England take a major Reputation hit for breaking a trade agreement and a RoP agreement with me? And if not why?

BTW I play Civ III Vanilla if that makes any difference.

Click to expand...

I think it does take a rep hit, but its unknown to me exactly how much this matters to AI-AI deals.
It has been noticed by some high level players that AI that took a rep hit where still able to trade with other AI's in ways the human wouldn't be able to.

Maybe it doesn't effect the AI, maybe the effect last only a limited amount of time for the AI. maybe the effect is reduced at higher difficulty levels?

I don't know.

I do know that if the trading partner hasn't met you, it won't know of England's misdeeds. This is the same for the human btw, if you RoP-rape an AI that hasn't made contact with anyone else, no one will know what you did so your rep stays OK.

In the years that I've posted here I've never seen a serious analysis of the effects on an ai for trashing it's reputation. I and many others have assumed that when ai's lose reputation it causes their rivals to become more likely to declare war on them. If this is the case (and I have not seen it proven) then it explains why they get more and more warlike as time goes along.

If reputation counts then it is also to the player's advantage to "light the powder keg" so to speak by starting a war early and buying an alliance or two. Once they get going they have a pretty consistent habit of alternately buying alliances and then breaking the alliances early until their reputations are thoroughly trashed ... again I say, if that really matters to them.

edit> comment on mas's x-post: i would be surprised if rep affected their trade deals much, if at all. it's more the rop rep or whatever you call it that i'm commenting on here.

MAS said:

...if you RoP-rape an AI that hasn't made contact with anyone else, no one will know what you did so your rep stays OK.

Click to expand...

I'm not 100% sure, but I believe that if you keep your victim alive until it does establish contact with other civs, then the story of your treason will get past on and your reputation in this area is in tatters.

But just a few days ago Lord Emsworth performed a test where he RoP raped a civ, and he was still able to sign an RoP with another civ straight after. Lord Emsworth posted about this in the Newbie questions section.

I believe there is still a void in the strategy articles section where reputation and the effects of it are concerned.

Optional said:

But just a few days ago Lord Emsworth performed a test where he RoP raped a civ, and he was still able to sign an RoP with another civ straight after. Lord Emsworth posted about this in the Newbie questions section.

Click to expand...

He didn't RoP-rape anyone, as he didn't have a RoP. Link

Jokeslayer said:

He didn't RoP-rape anyone, as he didn't have a RoP.

Click to expand...

I've always read differently in the forums. I've just dug up a strategy article by Microbe about trading reputation. It also cover RoP reputation, and it says:

'RoP reputation is a related topic but quite independent of trading reputation. It decides if AI is willing to sign RoP with you. Your RoP rep is trashed if you declare war on the AI while you have any units with attack/defense value (including boats) inside AI territory, regardless of whether you actually have RoP deal with the AI.
Trading and RoP reputation is maintained seperately and does not affect each other.'

Things among these lines I've always come across in the forums, that's why I was surprised about the outcome of Lord Emsworth's test.
I'm planning to do a couple of tests myself now, as it's unsatisfactory to get so much conflicting information. Bear with me...

My reading agrees with Optional's. I've always read that if you DOW'd or just surprise attacked, and had had units in the now-defending AI's territory at the beginning of the turn, it counted as violating an "implied RoP."

Looking forward to the results of Optional's test.

Aabraxan and Optional are definitely right, although I believe breaking an actual ROP agreement is worse.

And as far as I know, the AI don't have reputations with each other, and it's totally up to the human player how he perceives an AI's reputation.

Optional said:

I've always read differently in the forums. I've just dug up a strategy article by Microbe about trading reputation. It also cover RoP reputation, and it says:

'RoP reputation is a related topic but quite independent of trading reputation. It decides if AI is willing to sign RoP with you. Your RoP rep is trashed if you declare war on the AI while you have any units with attack/defense value (including boats) inside AI territory, regardless of whether you actually have RoP deal with the AI.
Trading and RoP reputation is maintained seperately and does not affect each other.'

Things among these lines I've always come across in the forums, that's why I was surprised about the outcome of Lord Emsworth's test.
I'm planning to do a couple of tests myself now, as it's unsatisfactory to get so much conflicting information. Bear with me...

Click to expand...

You may well be right. I thought RoP-rape as a term only meant if you had a RoP agreement and there was a different term for declaring war with troops in enemy territory. I assume "true" RoP-rape would be worse than the other on your rep, but I don't know.

Well, I performed some tests. They were performed on a fully patched Conquests, while Lord Emsworth's tests were performed on Vanilla, but there doesn't seem to be any difference between them.

I picked up an old gamesave where I played as the Hittites and had started next to the Arabs and the Zulu. First I started moving some chariots into Arab territory. I got a message from Abdu Bakr in the interturn asking me to remove those. I said 'sure', but instead of removing them, I attacked a town of his next time I was around. Then I asked Shaka for an RoP, and... no problem!
Then I did the same thing again, but now I first made a trade deal with Abdu Bakr. I bought a tech from him for a lump sum and gold per turn. So when I now attacked him, I didn't only violate a peace treaty (and a possible 'Implied RoP') but also a sealed and signed 20 turn deal, albeit a trade one. Still Shaka had no problem signing an RoP with me.
Then, for the sake of being complete, I repeated the whole thing again, but now I signed an RoP with the Arabs first. This time, when I asked Shaka for an RoP, he would 'never consider' such a deal. Clearly I had blown my RoP reputation.

So, my conclusion is there's definitely something wrong with the generally accepted opinion around RoP reputation. Both Lord Emsworth's and my own tests prove you can attack an AI from a position where you are already in that AI's territory, even at the beginning of the turn, without damaging your RoP reputation.
I think the problem now is that there is so much anecdotal evidence around of players that have had their RoP reputation blown without ever signing an RoP, that you cannot ignore that. I believe I've also had it myself; being accused of 'sneak attacking' or something, without having an actual RoP in place. This concept of 'implied RoP' must come from somewhere, otherwise so many experienced players wouldn't believe in it. But where does it come from?

If somebody has a recent case of a blown RoP reputation without an RoP in place, or has found an example of this in the archives, then I would gladly hear about it.

By the way, I don't have internet at home anymore, so I'm even slower than usual in my forum activity.

Some very interesting results. I will search where this implied RoP rape came from, but I seem to recall an article by Bamspeedy on everything about reputation and attitude.

i have always experienced that this 'implied rop-rape' does exists and trashes my reputation. i am paranoid about signing a rop, because AI doesn't care about rep hit, i only do for increasing an offer and sometimes instead of increasing the chance of acceptance, it decreases that. i came to the implied rop-rape conclusion and also read somewhere here numerous times.

Well, that's certainly an interesting turn. Thank you for running the tests, Optional.

Optional said:

Well, I performed some tests. They were performed on a fully patched Conquests, while Lord Emsworth's tests were performed on Vanilla, but there doesn't seem to be any difference between them.

Click to expand...

Thanks a lot for the independent confirmation.

What I have been wondering though is, both our tests checked for what a third civ thinks. But what about the civ that is actually involved, i.e. the one that is attacked?

Emsworth's comment brings to mind a corollary question that I didn't think of earlier, Optional. Did the Arabs have contact with the Zulus at the time of your test? I'm wondering if that might have made a difference.

I'm very interested in the outcomes of these tests. My only "explanation" would be the difference between a "true" RoP Rape and an "implied" one. I'd like to see Aabraxan's question answered, too. That might come into play, as well....

As to another point raised, the AI *does* care about reputation. Or at least, they don't differentiate between the human and other AIs when calculating it. The difference lies only in the final calculations, after they have determined that you have trashed your rep. Only then does it check to see how *much* effect it has, based on the difficulty level.

Aabraxan said:

Emsworth's comment brings to mind a corollary question that I didn't think of earlier, Optional. Did the Arabs have contact with the Zulus at the time of your test? I'm wondering if that might have made a difference.

Click to expand...

I don't know about Optional's test, but when I tested the two civs (Russia and Germany) did have contact with each other and were not at war. (It would have been an easy trap to fall into though)

Aabraxan said:

Did the Arabs have contact with the Zulus at the time of your test?

Click to expand...

Yes, I did make sure I checked that. Remember also the third part of my test where I first signed an actual RoP with the Arabs. Then the Zulu's didn't want to know about an RoP with me anymore.

It's still strange that, after seven years of intensive play by the civ community, there's an established opinion about a game mechanism that can be disproved in an instant. Because anyone can repeat the test I did. It only takes two turns of play on a gamesave. One to move your units into another civs territory, a second one to attack, then you check a third party's response, and that's it; then you know the standard view on the 'Implied RoP' is wrong.

If I knew what further tests I could do to find out more, I would do them. Because I still have this gut feeling that the concept of 'Implied RoP' is not an illusion.
I'm also one of those players that, when he attacks, he makes sure he has no units in enemy territory. I even wait an interturn before moving units in. More players are like that, and it's based upon experience. I've never lost my RoP reputation in a long while, and I forgot about the exact circumstances when I did so in the past. I'll probably change the way I attack now, but keep an eye on my reputation.

There is certainly much that is not understood on this subject; in one game in which I purposely committed ROP Rape, the same civ later allowed me another ROP with them as part of an alliance against a third civ.

That's interesting, Bucephalus!
Hmm, would reputations perhaps come in a hierarchy? So that, when presented in a single deal, your reputation in one area could outweigh your reputation in another?