I don't see what there is to complain about. The AI is doing its job well of not being the player's lapdog
That's it, can't play this anymore.
- Thread starter jonman122
- Start date
-
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.
Would you say you normally post in threads to deny that the OP has a legitimate point of view, or do you mainly do that only when it's an opportunity to call someone a noob?
Paradox can never make a game design mistake, apparently; players can only fail to properly play their games. And, just as people see faces in clouds, they can always cook up some explanation for why random numbers, or bad design, are actually clever plans or good design.
The point is that some people disagree with you on what is good game design. I've been known to turn upon my allies without more reason than "I want your stuff" I don't see why my allies shouldn't do that to me.
It was really illustrative how many people said "-5 relations per base tax border friction is fine, you're imagining it, stop overreacting, turn the difficulty down you babies" the day before it was reduced to -1.
It was annoying in that it basically required you to ahistorically maintain a vassal wall between your lands and those of your allies. Beyond that factor it wasn't a big deal.
Claims are the new border friction, making it near impossible to maintain neighbor alliances. Pretty much anyone who gets a claim on you will turn hostile if they think they can beat you. In my Brandenburg game, Poland, Denmark and Bohemia have all turned from green hearts to flames after getting a single claim on my territory.
The +/- relation should mean something. For me, it is a simulation of a relationship with two humans. If one loves other, but other hates him, there is a space for a backstabbing. But if they are mutually at a high relationship around 200, then a backstabbing makes no sense.
I feel that maybe you should be notified if your ally is about to break an alliance. For example, if AI Spain doesn't want you colonizing Mexico, then you should be notified about it and instead of continually colonizing land that leads to a breaking of an alliance, you can stop colonizing there and go colonize Florida or something. If Bohemia is afraid that your HRE expansion has their name on the list, then you should be notified that they are worried about your recent expansions near their borders. You can stop your expansion there and let them know that you won't crush them... yet. If an alliance is about to be broken because a common enemy has been lost, then you should be notified about it and plan accordingly. It would be much easier for us to plan our next moves if we knew what our allies feel about it.
I'm confused. You want the AI to notify you when it wants to break an alliance for presumably, it's own benefit?
People keep treating the AI like it's some stupid child with a multiple personality disorder. That's EXACTLY how most players play. They start as OPM, ally big major power, beat slightly bigger enemies thanks to super awesome never warleader best buddy major power, then when big enough they screw the AI over and betray it. Alternatively, the AI breaks the alliance before the player is powerful enough, and then some players decide to whine and moan on the forums.
What did the OP expect? That Spain will become his not on paper vassal? Each nation has it's own agenda. Here's a quote from Lord Palmerstone, that pretty accurately describes the AI in this game.
"Therefore I say that it is a narrow policy to suppose that this country or that is to be marked out as the eternal ally or the perpetual enemy of England. We have no eternal allies, and we have no perpetual enemies. Our interests are eternal and perpetual, and those interests it is our duty to follow."
I think a large part of the problem stems from the fact that Paradox is (seemingly) trying to make the AI act like a human player, instead of building a game in which both humans and AI act in a historical manner. It seems that due to the limitations of the system the game is built on, our only options are a dumb-as-cattle AI or a dumb-as-cattle AI that has been gamily engineered to provide challenge.
I think the ultimate goal of AI is so that it is basically an artificial human. (I'm probably completely wrong on this one, I know nothing about AI.) I'd rather not them turn the other way.
I don't think anyone's arguing against that. It's just that Paradox is missing out on the primary reason we play with actual humans, which is to say that we can interact with them on a level other than "ally" and "enemy." If the reason for the Spanish betrayal of OP was his intrusion on their South American interests, that's a step in the right direction. But I think between two humans, they would have worked something out and divided the land between them instead of trashing their century old alliance and instantly warring.
I can't really argue with you there. Although I would admit, if I was in a similar situation as Spain, I would've demanded all his SA colonies or it would've been war. The AI just skipped the demanding and negotiating part.
I have noticed the same thing. The problem being that the AI over reacts to minor changes. As a human I do not care about every single claim and CB I have. But the AI thinks a single claim on a province belonging to an ally is enough reason to start hating.
Diplomacy needs some tweaking. It´s not that the AI starts hating you for no reason. The problem is that the AI starts hating you for very minor reasons.
Yeah, if the AI is going to hate me for eating the last cupcake, it should at least demand the cupcake and threaten war. Of course, if the AI was to do that it would seem petty. Do the AI nations really break off century-old alliances and go to war in the period of a few months over random claims? How do we ever get centuries-old alliances if this is enough to break one?
Broadly, this is completely wrong, at least as far as video game AI. The goal of programming an AI to play a computer game is to provide a structured and scalable challenge to the human player of the game, within the design goals of the project. For instance, it would be really very straightforward to develop a fighting game AI that could 100% perfect any human player. But this isn't desirable - except, sometimes, in arcade games, it IS desirable, so that you can make somebody credit feed. Another example: the leading edge of AI design for Chess and Go isn't trying to make them play like a human. It's trying to make them play very well, by harnessing the strengths of computer architecture.