Civilization 5 AI ? Civilization 6 AI ?

[Barghaest]

Tourism is both very deterministic in its accumulation, with little player input other than archaeologists, and forced on you by its overwhelming importance to ideology. This also makes ideology an "I win" for the player with the lion's share of tourism-related Wonders. If you aren't a tourism leader, all you can do is select the most influential ideology among everyone else's (and actively try to avoid being first to ideology), and become a bit player in the late game blocs constantly being bothered by allies' invitations to declare war on the other lot, and with no ability to direct the late-game diplomatic landscape, even if you're otherwise a world leader and positioned to win, say, science victory.

Ah yes. I generally get a decent tourism output even when not going for CV because the theming bonus for Great Works also applies to Culture as well so I concentrate on accumulating them whether I go Aesthetics/Exploration on not. I close borders to high tourism civs to reduce their impact on my empire and sometimes I wage wars to acquire more Great Works through conquest (razing the previous city down and moving the GWs to empty slots in my existing empire).

In my Mongolia game, I did end up completing the Exploration tree (went primarily for the Happiness/Gold from Lighthouse/Harbor/Seaport to fix my early issues) because no other civ did and that gave me sole access to Hidden Sites, which I used to boost my culture gain by filling my empty Museums and even started a culture renaissance from one to lower the influence impact of Egypt on my empire.

 

[mitsho]

I understand this is a list of what the AI shouldn't be, but this is also what is commonly critiqued about the Civ5 AI so my points remain the same.

Yeah, I wasn't specifically speaking of civ5 here, but more generally on strategic games. I wasn't saying that personality doesn't exist in civ5 at all f.e. I do feel like I should lose some more words on "Role Playing" which may be a bit wrong choice of terminology here.

If a civ is "roleplaying", it will make decisions based on its personality (Egypt going for wonders), its environment (horses -> going for knights and maybe be more aggressive), its set-up (we share religion with X, so I'll attack Y) and its history (I've been trading with Z for centuries -> I like them). In this kind of sense, every action of a civ looks logical or rational (if you knew everything about them). The opposite is when a civ suddenly declares war on you to prevent you from reaching culture victory. Every victory is 'alien' in that it doesn't make sense for the government in a real world to believe it "loses the game".

Simulation is a better term for this than role-playing. My point is that a good game needs both (but simulation is more important here than in a fantasy world like starcraft) and that to achieve a good AI one first needs to define the goals of the game. If it's victory conditions, what are they and if one thus wants for the players to "gang up" on the leader? My personal opinion is no since I believe in civ as more of a simulation and that would be as strange as - real world example - England declaring war on America in 1990 because "we fear you are becoming too strong".

And lastly, all my comments above are 'relative'. Of course there needs to be direct bonuses to the AI, but the devs should try to reduce them. It's a bad excuse if you use this technique of bonuses to band-aid problems every time.

As this thread is about AI, my main argument is that one needs to develop the game and its AI simultaneously as they interact so heavily (see 1upt rule).

 

[jokii]

Many people play peaceful games in Civilization , building the economy, trading and diplomacy.
If you try playing the game from the beginning in "always war fashion", you will realise the awesome stupidity of the AI.

If you take the AI settler, and declare the war at the beginning of the game, AI will not build units to defend.
Instead he will build one more settlers and send it do death. Or worker, or he will built and try to attack you with only one warrior.

Only thing that AI can use in his advantage is city bombardment and archer inside the city.
Really disappointing.... I want to play war most of all.... If I wanted to play economy, city construction, I would play SimCity

Only lack of happiness which is connected with the fact you cannot raze capital cities is what stops you to raze the whole map in the early eras of the game.

 

[jokii]

In my last game I attacked Egypt at the beginning of the game by taking his settler.
I had one warrior and two scouts around his borders.
After the declaration of war Egypt made worker which I took with my scout. Then he made another worker which was in city the whole time because I was close.
Next he made War chariot which was in the city the whole time. After that he made two cargo ships and Trireme ship.
After that nothing, for the entire era. He was constructing buildings or trying to build wonders.
I still have only two scouts and warrior, and his worker is still in the city.
When I came with catapults , an era later, he had level 10 city without terrain improvements but he did not made second land unit.
Really dumb, dumb, non existent AI. Even CIV 4 AI is far better and much more competitive.

 

[dexters]

I never used to have that issue: late-game war was commonplace in my older games.

My issue with the ideology system isn't merely in the way it distorts diplomacy (and indeed Congress resolutions - at least in principle tactical voting is an interesting development while it lasts, but even that is subsumed beneath the overriding influence of ideology).

It's the way it forces tourism spam.

Yeah, but to clarify you defend with culture. You get so much happiness in the late game however that it really should not be an issue. There are plenty of modifiers to defend against tourism anyways.

The usual risk of dissidents is if you're first to hit ideology and you take a blind leap of faith, but if you're first, you usually have a few things going for you. Most AI so rarely choose freedom its usually a toss up between order and autocracy.

If you're playing from behind, you can sort of guage where you want to lean to avoid dissidents and or have a good diplomatic shot at getting it set as the world ideology to avoid it.

If you're in a situation where dissent unhappines is so crushing that you're cursing at the game, you've probably already lost the game even without the ideology mechanic.

In my last game I attacked Egypt at the beginning of the game by taking his settler.
I had one warrior and two scouts around his borders.
Really dumb, dumb, non existent AI. Even CIV 4 AI is far better and much more competitive.

Nah, you probably just picked the one AI civ in the game that has a builder flavour seed in that particular game, and yes an early war generally cripples builder AIs in terrible ways.

One thing you have to realize is that unlike Civ4, where all AI is essentially the same AI with different skins, the AI in Civ5, while building from largely the same base, do follow different strategies. This is why you can see the same Civ sit on a few cities in one game, but expand rapidly and wage war in another.

And no, Civ4 did not really have a better AI. Civ4 AI is fairly predictable. Massive early rexing, consolidation, more rexing. Every AI did the same thing so it's just a rush to build as many cities as the maintenance formula will allow you until all the land is filled up.

 

[player1 fanatic]

Lets talk about AI bonuses.
The idea is to compensate AI bad reasoning with bonuses to make it appear more successful.

One place where Civ5, as well as Warlock game, failed in this, is combat, where 1upt is used.

Giving more units to AI is bad way to compensate for lack of decent tactics when using 1upt system.

1upt requires stonger units on side of tactically weaker player to add challenge.

Thats why deity in civ5 can be fun when waring with AI that is one era of military tech above you.

What civ5 really needed is giving free promotion bonuses on higher difficulties to AI to compensate weaker play, not just adding more weaker units as cannon fodder for human player to train his elites and increase military gap in his favour.


Anyway the idea is when adding bonuses to AI is to know what kind of bonusess to add that will improve AI strength. Sometimes one kind of bonus is not effective due to the way how game rules are designed.

 

[PhilBowles]

Yeah, but to clarify you defend with culture.

As far as I can gather, you defend against influence with culture. Not against ideology. Ideological pressure is related to the influence civs with ideology X have over you, and so culture buffers it to that extent, but the determining factor in which ideology influences you to the greatest degree is whose ideology has the greatest combined influence. And that means that in order to remain competitive your own ideology has to have high tourism.

This is partly why it's a poor design - this clumsy way in which an offensive resource acts as a defensive resource against ideology.

But beyond that, tourism spam is still forced, because tourism is itself the major source of culture now that the base output of culture buildings is so low. So in order to build up a cultural defence against foreign ideology, you still need to focus on tourism generation to almost the same extent you do if actively using tourism as an aggressive resource.

You get so much happiness in the late game however that it really should not be an issue. There are plenty of modifiers to defend against tourism anyways.

There is only one modifier that defends against tourism: differing ideology. Moreover, war is also the only way to expel diplomats, which you can't otherwise control, since although they need embassies to be installed, denouncing (and so kicking out the embassy) does not kick out the diplomat. Aside from that, you can try to deny some of the positive modifiers - you have control over open borders and shared religion, but not over trade or necessarily over ideology (if you choose one before they do).

Most AI so rarely choose freedom its usually a toss up between order and autocracy.

In my games, the first three AIs to ideologies tend to adopt one of each in sequence, unless one of the earlier players to ideology has an overwhelming level of influence. I certainly haven't noticed them avoiding Freedom - if anything Autocracy seems to be the rarest.

 

[dexters]

What civ5 really needed is giving free promotion bonuses on higher difficulties to AI to compensate weaker play, not just adding more weaker units as cannon fodder for human player to train his elites and

Don't they already get free 'cover' promotions at higher levels?

 

[APCY]

They just need better programming, i dont know what programmers works there but i doubt they are good...