AI diplomacy cheat

I'm now playing an Emperor tiny epic game with 18 civs. Great fun and such, but I've suddenly noticed something.

Almost everyone is Hindu, except for Catherine and Cyrus (Buddhist) and me (no religion).

So I checked Catherine's diplo modifiers with everyone, to see if someone didn't like her.
Turned out everyone was pleased, but I discovered two things.

First thing is that Cyrus gets a disproportionally huge same religion modifier for a leader who has Free Religion as a favourite civic. But that doesn't matter.

Second thing is : noone has a malus diplo modifier for 'You refused to
help us' or 'You refused to help us during wartime.' Only I had those.

I only checked about 50 diplo relations in a single game, so this might be coincidence. But is it true the AI doesn't get those modifiers towards other AIs, only towards the player?

You are right, AIs never get those demerits. Either they don't make demands that get refused, or they get no hits when they do.

Ok, the cyrus thing the a leader definition by firaxis , in a attempt to mimic the RL Cyrus ( Ancient persians ( that is, pre-Alexander ) were extremely tolerant in terms of religion of their conquered foes, but were quite adamant in enforcing Ahura-Mazda cult to the persians themselfes ). Not a good fit, but it is more a roleplay vs gameplay stuff than anything else.

The refuse thing is diferent, in spite of not being exactly a AI cheat ... it is more a AI shortcome Due to some reason ( mos likely lack of coding time by the devs ), the inter AI deals are never tried, unlike the human-AI and AI-human ones ... that is, the inter AI trade tries only happen when both parts would agree, making them have a 100 % success rate

Why i don't call this a cheat ? Simple, because the AI, due to the same reason, can't get "fair trade" ( aka I ripped off you ) diplo boni and the magnitude of the bonus is similar. Like i said, it is more a question of AI stupidity that anything close of a cheat.

Pretty sure they get fair trade bonus between them, maybe it's hidden unless you have certain mods installed?

The real diplo cheating is detecting worst enemy trades with civs they haven't met yet.

I really hate the base peaceweight thing too as it makes AIs love/hate each other and often screw itself in the process, but that fits better under r_rolo1's definition of "stupid" ai rather than cheating. Maybe some games monty/shaka should have it out, and cooperate in others. Current model gives them an extreme bias towards cooperating, even when doing so eliminates victory chance almost entirely (say, a 4th civ out there is going culture, or a civ with low peaceweight like they have)

I also dislike hidden modifiers. They are essentially a tax on newbies; good players learn how the mechanic actually works and beginners are left in the cold to get screwed unless they read on here a lot.

nfw said:

Pretty sure they get fair trade bonus between them, maybe it's hidden unless you have certain mods installed?

Click to expand...

Probably has to do with gifting. The AI can gift techs to the player when (s)he is really far behind (that's something I learned in my first Deity game ).
I was talking about the maluses though.

The real diplo cheating is detecting worst enemy trades with civs they haven't met yet.

Click to expand...

I was defining 'cheat' as an unfair advantage over the player, although I see how you can call this a cheat too. It does influence the AI as much as the player, though, as far as I know.

I really hate the base peaceweight thing too as it makes AIs love/hate each other and often screw itself in the process, but that fits better under r_rolo1's definition of "stupid" ai rather than cheating. Maybe some games monty/shaka should have it out, and cooperate in others. Current model gives them an extreme bias towards cooperating, even when doing so eliminates victory chance almost entirely (say, a 4th civ out there is going culture, or a civ with low peaceweight like they have)

Click to expand...

Do you mean leaders which do not like war (Joao, Gandhi, Elizabeth) will get a positive diplo modifier towards each other, and vice versa?
I too think the model should be more randomised, even without random personalities on. And random personalities, in my opinion, should completely randomise the values, instead of switching the personalities of the different leaders.

I also dislike hidden modifiers. They are essentially a tax on newbies; good players learn how the mechanic actually works and beginners are left in the cold to get screwed unless they read on here a lot.

Click to expand...

Can't do anything but agree with this.

Peaceweight is actually semi-separate from how much of a warmonger an AI is, making it somewhat misleading. What it DOES do, however, is provide a bonus with other AI of similar peaceweight.

Humans get no peaceweight bonuses, though AI have their own hidden modifiers with the human (zara/mansa/gandhi have positive hidden bonuses with the player, while some of our favorite warmongers have negatives ).

Sitting bull has an incredibly high peaceweight, while monty/shaka/alex are low. However, sitting bull still builds a lot of units and will easily declare on people annoyed or worse. In a typical game he and the warmongers will always hate each other on contact, and only start liking each other if religions or some random shared war combine to do so.

From ROLO (sorry, haven't learned to do the "quote" thing right.....
The refuse thing is diferent, in spite of not being exactly a AI cheat ... it is more a AI shortcome Due to some reason ( mos likely lack of coding time by the devs ), the inter AI deals are never tried, unlike the human-AI and AI-human ones ... that is, the inter AI trade tries only happen when both parts would agree, making them have a 100 % success rate

ok, fair enough from a role-playing aspect, there's no POINT in the AIs having to "randomly" determine if negotiations fail or not (I think....)...however, I have read that some of you guys talk about things such as hidden diplo modifiers between the AIs (warmonger respect)...so...to emulate the fact that say...um...Mongolia, Korea and Japan tended to war with each other in real history (or I should say Korea got to Defend a lot against Mongolia and Japan) are there hidden "hate" diplos pre-coded in to represent that? Or should the human player assume that all civs are kindasorta vanilla towards one another and the best way to tweak them is to get them into wars with one another and jack with their religions and civics?
TMIT kinda answered with peaceweights and warmonger respect....so I'll focus my question...are said peaceweights and WM respects actually kindasorta BASED on real world history? Can I expect, say, Korea and Japan and China to have a little enmity towards each other before even checking the modifiers the game WILL tell me about?

Peaceweights also make colonies even more undesirable, as if you suddenly spawn a vassal everything hates... well derp. Your colony's initial attitude should be the same as the current attitude of civs towards you...

In any case hidden modifiers are stupid. The game should flat us tell us that AIs inherently despise each other. I should not see a +4 and "friendly" without any real explanation.