Tinto Talks #56: Yet more changes

[Marquoz]

The link to this week's DD is https://forum.paradoxplaza.com/forum/developer-diary/tinto-talks-56-26th-of-march-2025.1733172/

Today's topic is more changes to already discussed systems. For a change, I found parts of this one interesting. I think Antagonism will be a good replacement for the old Bad Boy and the current Aggressive Expansion:

Antagonism indicates how other countries are likely to view us. If they feel a lot of antagonism towards us, countries that consider us as relevant to their interests will be less inclined to engage in diplomacy and may act against our interests. Antagonism is caused by basic differences between countries' societal values, government types, religion, culture and language, and actions can cause an antagonism 'bomb' in a location that affects the countries near it to varying degrees depending on how much they care about that location and about the antagonistic country. Antagonism 'bomb' effects will generally dissipate with time. Antagonism also affects a country's opinion of you.

The fact that you will have a baseline antagonism score against countries with different societal values than yours could have far-reaching effects, depending on how it's implemented, and may lead to a more hostile and aggressive AI.

[Triple G]

Originally posted by Marquoz:

may lead to a more hostile and aggressive AI.

How could the AI be more hostile and aggressive than it already is? Like it goes to war with anyone who is weaker - in the moment they have the opportunity (which excludes truce breaking, or taking any stab hits)...

Given the quote - i guess it´s some random element, based on these things like culture, which make wars more likely, or also to abandon alliances, or royal marriages, and on the other hand harder to forge these...

It´s then probably more unlikely to get certain alliances, even if we also have "baseline antagonism", if we see it as the modifier for different / same religion. This sounds like that - but more detailed, as it´s based on more stuff.

I´m not on speed with the development of EU5, as i usually wait some years after release to see how it works out, so i don´t know what they mean with "societal values". In EU4 most AI countries have world domination whatever it costs as their only societal value, and those who don´t have this, will be erased from the planet...

[Totally Innocent Chatbot]

Originally posted by Triple G:

I´m not on speed with the development of EU5, as i usually wait some years after release to see how it works out, so i don´t know what they mean with "societal values". In EU4 most AI countries have world domination whatever it costs as their only societal value, and those who don´t have this, will be erased from the planet...

It's an old concept mostly recycled from EU3. It replaces some of the mechanics currently encompassed by EU4's Idea Groups and Government Reforms, except that they are gradual "sliders" that take effect over years or even decades rather than an all-or-nothing bonus applied instantly. Centralization vs. Decentralization, Spiritualist vs. Humanist, Serfdom vs. Free Subjects, Quality vs. Quantity, Offensive vs. Defensive - all different areas for your nation to specialize in, and unlike in EU4, each is a spectrum of two mutually exclusive polar opposites (no Quality AND Quantity for you).

[Ashling]

Originally posted by Triple G:

How could the AI be more hostile and aggressive than it already is? Like it goes to war with anyone who is weaker - in the moment they have the opportunity (which excludes truce breaking, or taking any stab hits)...

In EU4 most AI countries have world domination whatever it costs as their only societal value, and those who don´t have this, will be erased from the planet...

It kind of feels like a codification of an emergent behavior we see in EU4. So, instead of the ai working against you (even if allied) because of an unseen "threatened/worry" ethereal type of thing it may now just be written in plain text why they're kinda annoyed even though you've done nothing directly against them.

I also think the ai "acting against our interests" might be a thing? The ai in EU4 is a bit self-absorbed, I don't think it would really consider what other nation's interests may be (they don't do things like guarantee the independence of a weaker nation to stop your expansion like they do in HoI; no idea if or should they do it in EU5)

[Triple G]

Thanks for the answer. Hmm - yes - sounds indeed better.

(I´m usually more curious how the AI would handle more complex systems, so i´m not really enthusiastic about features, which sound better, or more interesting, or more flexible, on paper.)

[Marquoz]

Originally posted by Triple G:

Originally posted by Marquoz:

may lead to a more hostile and aggressive AI.

How could the AI be more hostile and aggressive than it already is?

The AI often forms alliance hugboxes that last for a very long time because it can't find a way to overcome them (skilled players get around them with ease, of course). But if the AI has built-in antagonisms to nations with different societal values (the new sliders), cultures, governments, language, and so on, these hugboxes will be less likely than in the current system.

Obviously, until I see the game in action, this is just speculation.

[Triple G]

Originally posted by Ashling:

I also think the ai "acting against our interests" might be a thing? The ai in EU4 is a bit self-absorbed, I don't think it would really consider what other nation's interests may be (they don't do things like guarantee the independence of a weaker nation to stop your expansion like they do in HoI; no idea if or should they do it in EU5)

I don´t think they could solve this for the next couple of year or decades, as i think there would be limits to the AI.

Like the AI is only a code. If this is the case - do that. That´s about it. Even some machine learning AI does nothing different - the only difference is that it would try all possible moves in all possible orders - and then sort them by success chance. Which again results in: If this is the case - do that. Like the hardware only functions with on and off. There´s nothing in between.

I don´t imagine that the AI in EU5 would "feel more lifelike" - for me it would be good enough if it´s able to somehow play the game. In EU4 it´s somehow able to play it, which is why we spend many hours in it i guess. (and the amount of achievements)

[Triple G]

Originally posted by Marquoz:

The AI often forms alliance hugboxes that last for a very long time because it can't find a way to overcome them (skilled players get around them with ease, of course). But if the AI has built-in antagonisms to nations with different societal values (the new sliders), cultures, governments, language, and so on, these hugboxes will be less likely than in the current system.

Obviously, until I see the game in action, this is just speculation.

Yes - this would be something which could be solved with it - like when there are the set alliances - and an area comes to a halt, because nobody does anything, till the player attacks any one of the countries, or their allies...

Haven´t thought of this.

[Dumah]

I wonder if positive biases will exist, sorta "protagonism". If for example Poland will beat up Ottomans, would balkan countries cheer for them? They are slavs, christians, defeated alien invader, this surely should amount to some diplomatic boons. Or system will just serve as a punishment, just like BB and AE before it... in that case i dont see much improvement, really.

[Marquoz]

Good question, Dumah. You might want to ask it in the official forum thread.

[Octopuses]

It's like more strict AE.

Anyway, for me this confirms game will be released late this year, November probably.

[Triple G]

Originally posted by Octopuses:

Anyway, for me this confirms game will be released late this year, November probably.

Usually to release a game like this for the winter sales is a good idea - from a business point of view.

But i for myself rather wait some two years after the release, to see how it is like. I was super disappointed by Stellaris, and that the AI couldn´t play it - even after waiting some time to buy it as i (mostly) never buy a game at release . (actually "mostly", because i bought one early access game - which means i technically bought it before the release, but was also disappointed by Bannerlord)

[cayenne_spicy]

Originally posted by Totally Innocent Chatbot:

Originally posted by Triple G:

I´m not on speed with the development of EU5, as i usually wait some years after release to see how it works out, so i don´t know what they mean with "societal values". In EU4 most AI countries have world domination whatever it costs as their only societal value, and those who don´t have this, will be erased from the planet...

It's an old concept mostly recycled from EU3. It replaces some of the mechanics currently encompassed by EU4's Idea Groups and Government Reforms, except that they are gradual "sliders" that take effect over years or even decades rather than an all-or-nothing bonus applied instantly. Centralization vs. Decentralization, Spiritualist vs. Humanist, Serfdom vs. Free Subjects, Quality vs. Quantity, Offensive vs. Defensive - all different areas for your nation to specialize in, and unlike in EU4, each is a spectrum of two mutually exclusive polar opposites (no Quality AND Quantity for you).

Me when I'm like: "but quantity is a quality of its own"

Everyone: "shut up Pepper Stalin"

[cayenne_spicy]

Originally posted by Dumah:

I wonder if positive biases will exist, sorta "protagonism". If for example Poland will beat up Ottomans, would balkan countries cheer for them? They are slavs, christians, defeated alien invader, this surely should amount to some diplomatic boons. Or system will just serve as a punishment, just like BB and AE before it... in that case i dont see much improvement, really.

This is what it made me think, too, that perhaps what they're getting at is that the undercurrent and hidden modifiers which affect the spread of AE as it's implemented in EU4 (such as religion, culture, etc) will instead be more malleable and affected by things like the extent of good relations with various countries of those demographics and the duration of time those relations have been maintained. So for example, a Muslim country which has gone out of their way to maintain good diplomatic relations with a million Catholic HRE princes would find that all of those German Catholics won't be as mad at them over them taking over Malta as they might otherwise be.

[Ashling]

Originally posted by Triple G:

Originally posted by Ashling:

I also think the ai "acting against our interests" might be a thing? The ai in EU4 is a bit self-absorbed, I don't think it would really consider what other nation's interests may be (they don't do things like guarantee the independence of a weaker nation to stop your expansion like they do in HoI; no idea if or should they do it in EU5)

I don´t think they could solve this for the next couple of year or decades, as i think there would be limits to the AI.

Like the AI is only a code. If this is the case - do that. That´s about it. Even some machine learning AI does nothing different - the only difference is that it would try all possible moves in all possible orders - and then sort them by success chance. Which again results in: If this is the case - do that. Like the hardware only functions with on and off. There´s nothing in between.

I don´t imagine that the AI in EU5 would "feel more lifelike" - for me it would be good enough if it´s able to somehow play the game. In EU4 it´s somehow able to play it, which is why we spend many hours in it i guess. (and the amount of achievements)

They did it in HoI (and I believe Imperator), I've seen it done well in other games, it's not really about being life-like. It's the difference between a chess-bot sacrificing a rook now for a better position later vs a bot who never sacrifices a piece ever: not all self-interested actions are beneficial in the long run.