They need to hotfix AI agression now

[glider1]

There is now evidence that a player has won on Deity with a single city. That is not good. Players should be pushed into at least 3-4 city expansion in my opinion.

I don't think it has been talked about much, does anyone think less ancient age war means less opportunity cost for establishing embassies it seems like a no brainer to establish them?

 

[Aristos]

So, no one has tested what I proposed a few pages (and days) ago? Increasing the AIDeclareWarProb value for the higher diff levels to see what happens?

I said I will not test it because I am still seeing good and balanced games in my experience, plus I am so committed now that even if I test it and arrive at the same conclusion as I already have, who would take it as confirmation?

It's not that hard to change that value. Anyone?

 

[glider1]

Thanks Aristos. I'll test it for sure considering I've modded the DLL to make the AI's attack more potent.

 

[Aristos]

Thanks Aristos. I'll test it for sure considering I've modded the DLL to make the AI's attack more potent.

Out of curiosity, what did you change there? Someone proposed a few pages ago (or maybe it was in another thread, I don't remember) to kickstart BetterBNWAI Project...

Perhaps we should do that.

 

[Buccaneer]

You have no idea what cascading effects changing one value would be. No one should make gameplay changes to the files without studying and thoroughly testing the results, esp, in effecting the AI use and behaviors, and of course balance/unbalance. Only firaxis and their test teams should do that.

I am happy that you like the current balance; the rest of us will wait for the fall and winter patches to see what changes they bring in making more AI civs play more to their (combat) strengths instead easily being pacified and not winning effective at higher levels.

 

[Aristos]

You have no idea what cascading effects changing one value would be. No one should make gameplay changes to the files without studying and thoroughly testing the results, esp, in effecting the AI use and behaviors, and of course balance/unbalance. Only firaxis and their test teams should do that.

I am happy that you like the current balance; the rest of us will wait for the fall and winter patches to see what changes they bring in making more AI civs play more to their (combat) strengths instead easily being pacified and not winning effective at higher levels.

"The rest of us"??? Holy C, I did not know I was alone... should have told me earlier man, so I stop wasting my time here.

Of course I have no idea of the effects; that is why it's called testing. I find it truly amazing that such an acid reaction can come due to an intended attempt to lend a hand... what happened to this community?

 

[Buccaneer]

Didn't mean to imply you were the only one. I don't know if you are in the majority (which I doubt) or minority.

The response was timed to what just happened in my current game. I was going along peacefully building, getting ready to start the Radio buildings and I get backstabbed by the Huns. They won't be that effective because they have to come by sea and I have lots of good city-states in the area. But it reminded me - again - what an equalizer attacks like this can be. Now I have to stop my peaceful building and work more on my military units, and putting me behind on tourism and culture that I was planning. Once the war is done, I'm going to have to work on catching up, if I can. That is my definition of.a challenging game which aggressive wars from the AI can do best, more than the usual few that are known to do.

 

[glider1]

No probs, we can move discussing changes over to the modding forum Aristos

 

[Kordanor]

Well, modding this file brings two problems I see:

1. I only play multiplayer games (with AI). And modding in MP is still not possible. I am not sure if it worked if all players would do the changes in the files though. So I guess there is no way around an official solution.

2. Even the current situation is unclear. I mean, after playing maybe 50 hours in 9 different games I am very sure that the AI is much to passive. But as you can see in this thread lots of people don't share this experience.
In addition I guess it would take me more than 5-10h to be able to judge any change in the settings more or less objectively. And I don't have the motivation to spend that much time on testing for a result which would probably not help me anyways due to 1.

 

[95soram]

I have been on this thread in the past complaining that the ai was too passive. Now to be fair, I have to revise my story a little bit. I have played five games so far all small/emperor/normal speed. In my first game I was only attacked only once at around turn 160. Then I had three games in row were I was never attacked the whole game. In two of those I deliberately kept a tiny army and still wasn't attacked. The story changed in my fifth game. I was attacked repeatedly by multiple ai civ's starting on turn 170. There was so much fighting I decided to switch to and got a domination win for the first time in a long time.

So I guess in the end the amount of ai hostility is somewhat random. I still think we need more DOW's in the early part of the game, especially for the civ's that have an early conquest bias.

That is just my two cents.

 

[IdiotsOpposite]

Why not just create a separate AI dll that's activated when the player turns on an "Aggressive AI" option, that's programmed to make things like early war and otherwise-non-warmongerers like the Maya ready to war more often in the early game? That way everyone wins.

 

[Enrico Swagolo]

AI is really weird right now. Emperor Venice game - I'm sandwiched very closely between Shaka and Atilla. Despite me having 1 warrior 1 scout for a long time, a good location and a lot of best-of-the-best wonders, they didn't attack me, neither alone (the advisor said they could both wipe me from the face of earth) nor together, as would've typically happened in Civ 5 G&K.

They didn't even desire my lands! And our capitals were very close to each other.

Only once I saw Shaka's forces approaching, but it was late Medieval - I had so many wonders in my capital and good gold/gpt with which I easily rushbought two Composites, built one and he kinda... walked around with his catapults/spearmen/composites before finally giving up.

Not only that, Shaka didn't settle a second city until around 300-500 AD, Atilla didn't settle his second at all. He didn't attack neither me nor Sejong, whose army is now worse than mine.

Firaxis should really fix that and disregard them "ai declares war on me when I have no army despite me having thuosands of valuable wonders in my lands? SO UNFAIR" people. I think only once an AI declared a war on me since BNW (and it took a provocation while having much, much smaller army), it's mostly me having to warmonger if I want any conflicts...

 

[pilot00]

So, no one has tested what I proposed a few pages (and days) ago? Increasing the AIDeclareWarProb value for the higher diff levels to see what happens?

I said I will not test it because I am still seeing good and balanced games in my experience, plus my position is so strong now that even if I test it and arrive at the same conclusion as I already have, who would take it as confirmation?

It's not that hard to change that value. Anyone?

I said I wont post again here but I have a question: What does the bold text above mean exactly?

 

[Aristos]

I said I wont post again here but I have a question: What does the bold text above mean exactly?

yes, I was not sure how to word it... I was thinking "polarized" but did not like the sound and does not represent what I think... it means that if I test it, and my own test confirms that my initial impression was right, who would take my confirmation seriously as it is only confirming what I was already advocating here (including some code evidence that apparently, some don't like)?

 

[pilot00]

yes, I was not sure how to word it... I was thinking "polarized" but did not like the sound and does not represent what I think... it means that if I test it, and my own test confirms that my initial impression was right, who would take my confirmation seriously as it is only confirming what I was already advocating here (including some code evidence that apparently, some don't like)?

It just seemed by the way you posted that you crescent yourself the one that brought the undisputed answer. Just pointing out how someone might read it, no hard feelings here from me.

I don't believe that there is someone that 'doesn't like it'. The problem is that it doesn't match with what many (including me) are experiencing in the game right now. And for that we are called stupid and liars. Because lets face it: If I can convince you that the AI is faulty I will win the prize!

Anyway probably shouldn't have posted again.

EDIT: BTW what are the chances of the code existing as it should (Which IMHO is faulty logic as well but thats another matter) but not working as it should?

I am happy that you like the current balance; the rest of us will wait for the fall and winter patches to see what changes they bring in making more AI civs play more to their (combat) strengths instead easily being pacified and not winning effective at higher levels.

Apparently I grow tired of hell freezing again

 

[Aristos]

It just seemed by the way you posted that you crescent yourself the one that brought the undisputed answer. Just pointing out how someone might read it, no hard feelings here from me.

I don't believe that there is someone that 'doesn't like it'. The problem is that it doesn't match with what many (including me) are experiencing in the game right now. And for that we are called stupid and liars. Because lets face it: If I can convince you that the AI is faulty I will win the prize!

Anyway probably shouldn't have posted again.

EDIT: BTW what are the chances of the code existing as it should (Which IMHO is faulty logic as well but thats another matter) but not working as it should?

Who called you stupid or liar?

Who is trying to convince anyone?

All AIs are faulty or we would not be having this discussion.

Faulty logic is a possibility, but to find that we (?) need to dig into the code, at least before we happily declare it broken. Or let the devs do their job, without declaring it broken without knowledge.

 

[Buccaneer]

Apparently I grow tired of hell freezing again

I'm glad we're on the same page again, I don't even remember where we were a while back.

 

[pilot00]

I'm glad we're on the same page again, I don't even remember where we were a while back.

Must have been a couple more times recently

 

[Aristos]

@pilot,

yes, you are right, now that I looked it sounded like what you said. Nothing like I was trying to say. I just wanted to try and help; I guess it's really not worth it (not that I float in time anyways...).

Maybe I should just leave.

 

[ww2commander]

Incase people are not aware, this issue is discussed in the latest Polycast by Ed Beach and Dennis Shirk

- 18m05s | Interview
Stemming from the chat still, whether the BNW Artificial Intelligence (AI) considers trade routes when declaring war, the absence of their scaling for map size and a seeming disconnect between the East India Company National Wonder and the Dutch civilization; then, AI considerations for war and a civilization's selection and relations in light of Ideological adoption.

EDIT: "Its geared to make smarter decisions"...i.e. will war impact its long term goals and economy too much.