AI+ v13.1

[Littlebob86]

What annoys me about firaxis is there lack of help post release.. Like if they had one employee helping a few of the bigger modders, you guys would get the answers you need.. Same with xcom 2, just nothing really helpful!

 

[Zuizgond]

Thanks Siesta. I was about to write that AI doesn't seem to build districts or get a religion in v6 when I saw that you released the v7 fixing those issues. Going to try this new version asap.

Keep up the good work!

 

[Siesta Guru]

...it seems that AI is so fixed on settling city spots(as hilghighted by the game itself) that it will keep sending settlers entire game until that exact spot is taken...no matter where that spot is!!
...seeing lack of farms...but too soon to confirm as a rule

That is pretty much accurate, yes. Luckily there are some factors that can make it so that some city spots, like those further away, will end up being marked as not worth it.But in some cases that doesn't really work, most notably when the spot is right in the middle of a bunch of barbarians. I think it may also go wrong if a spot is visible and appears close, but is hard to reach, thanks to mountains / enemy terrain / oceans in between.

What annoys me about firaxis is there lack of help post release.. Like if they had one employee helping a few of the bigger modders, you guys would get the answers you need.. Same with xcom 2, just nothing really helpful!

Hmm, I suppose I could attempt to actually contact them instead of passively waiting for them to contact me. This isn't really an official forum after all. I'm not sure what the proper channels for that are though.

Thanks Siesta. I was about to write that AI doesn't seem to build districts or get a religion in v6 when I saw that you released the v7 fixing those issues. Going to try this new version asap.

Keep up the good work!

Hope you'll find it to be better than in the previous version! If it's still not appropriate I may have to buff these desires even further later.

 

[Mynex]

The only change I've been able to find that can positively affect this, is that I can set a diplomatic 'cost' and 'worth' to the peace making diplomatic action. Sadly testing so far suggests that it doesn't affect (or barely affects) how much AIs are willing to offer in their trades, and would at most affect their willingness to get to peace. But it's not the willingness where the problem is, it's them tossing cities at you like they're luxuries.
I've already added a conservative change to these values in v6, and it's still present in v7 because it at least doesn't seem to affect things negatively. Let me know if you happen to notice a change in this area, it's unfortunately very hard to test this, as I can't see it in my battle royale games (which may be why the Civ AI guy hasn't noticed).

Where is this change made at? I'm getting ready to play a marathon game here soon and I'll test out what you've got... if it's good I'll let you know, if not, I'll tinker until I can find the sweet spot, or close to it.

 

[Siesta Guru]

Where is this change made at? I'm getting ready to play a marathon game here soon and I'll test out what you've got... if it's good I'll let you know, if not, I'll tinker until I can find the sweet spot, or close to it.

in diplomacy.xml look for


<Update>
<Set Worth="-20"></Set>
<Where DiplomaticActionType="DIPLOACTION_MAKE_PEACE"></Where>
</Update>
<Update>
<Set Worth="-10"></Set>
<Where DiplomaticActionType="DIPLOACTION_PROPOSE_PEACE_DEAL"></Where>
</Update>

You can also try setting the Cost variable the same way. The vanilla default for all 4 values is 0, and it probably only takes integers.
My hunch is that 'worth' how much general use it thinks it can get out of there, while 'cost' related to a 'diplomatic cost', aka, will this make me friends.

 

[Horem]

I use you mod along side another mod that adds a new District, the AI builds this new District in every City. But the vanilla Districts it picks and chooses what ones to build where. I know the new district doesnt add anything in regards to the AI other than setting AIFlavoredItem/Favored to 1(true) for Minor Civs. Wondering why the AI struggles with the vanilla Districts.

 

[Siesta Guru]

I use you mod along side another mod that adds a new District, the AI builds this new District in every City. But the vanilla Districts it picks and chooses what ones to build where. I know the new district doesnt add anything in regards to the AI other than setting AIFlavoredItem/Favored to 1(true) for Minor Civs. Wondering why the AI struggles with the vanilla Districts.

It doesn't really struggle with vanilla districts in any of my testing games. It tends to build them to the limit, with some variation on which it builds based on the civs, usually with an ok balance. Do you just mean it builds the new district more than vanilla ones? What are its yields/great person points?

 

[Salmuth]

This issue is also a little frustrating in the sense that I'm getting feedback in terms of a relative lack of districts on basically every district type. AIs simply can't be competitive in all areas of the game simultaneously. Unlike in civ 5, these things don't scale well into higher difficulty levels because the AI is limited by the same district numbers on every difficulty. If all AIs would hyperfocus religion, then suddenly a culture victory may be super easy as they're all neglecting that. If they focus both, they'll fall behind on science quickly. If they try to keep science up too, they'll end up with ****ty cities with 0 production. If they work on that too, they may have too few troops to fight off anything. If that is solved too, then were back to where we started as it's spreading so thin now that it's going to be mediocre on all issues.

I guess it is not possible to focus AI district production according to each city's possibilities like :

• great science/faith adjacencies (+5/+6) : make it a priority in this city.
• You got warmonger neighbours, build encampement in your closest city
• etc...

I wonder about their use of traderoutes as well. Could they prioritize them to go internal to make sure they grow instead of looking for the biggest gold output ? I guess they do the last considering their amounts of gold they often have.
I think about it because that could make them competitive since it is the main strategy now in civ6.

 

[Siesta Guru]

I guess it is not possible to focus AI district production according to each city's possibilities like :

• great science/faith adjacencies (+5/+6) : make it a priority in this city.
• You got warmonger neighbours, build encampement in your closest city
• etc...

Unfortunately not without the dll. Those kind of methods would probably indeed give the best results with proper tweaking.

I wonder about their use of traderoutes as well. Could they prioritize them to go internal to make sure they grow instead of looking for the biggest gold output ? I guess they do the last considering their amounts of gold they often have.
I think about it because that could make them competitive since it is the main strategy now in civ6.

There's nothing I can directly tweak to increase internal desire. It may be that they're affected by the preferences for yields though, in which case they would already be doing internal traderoutes more than in vanilla (not entirely sure whether that's happening). Sadly, even if they did do more internals, they would do it a very non-optimal way, as they won't relocate them tactically, which is the main reason that tactic works out so well anyway. The food is probably mostly wasted on them too, considering they do terribly at housing management. I don't ever see them build aqueducts on cities without fresh water and they barely build neighborhoods, suggesting they simply don't care. There's sadly no housing desire knob I can tweak either.

 

[Horem]

Do you just mean it builds the new district more than vanilla ones? What are its yields/great person points?

Alot more than the Vanilla ones. It has mixed yields of Gold/Food/Production. Great Merchant points.

 

[Siesta Guru]

Alot more than the Vanilla ones. It has mixed yields of Gold/Food/Production. Great Merchant points.

District choices are mostly made based on a combination of yields and great people points, it doesn't look at the buildings that it allows or special benefits. With those yields, it will spam it all the time because it'll think it's the best one in almost any situation based on those yields. Which then of course means it isn't building a vanilla district there.

 

[Horem]

District choices are mostly made based on a combination of yields and great people points, it doesn't look at the buildings that it allows or special benefits. With those yields, it will spam it all the time because it'll think it's the best one in almost any situation based on those yields. Which then of course means it isn't building a vanilla district there.

Oh I took into account the Buildings, the District base Yield is Gold from Adjacenices (Commerce, Harbor, River) and 1 GP(Merchant) point per turn. Thats it.

 

[Siesta Guru]

Oh I took into account the Buildings, the District base Yield is Gold from Adjacenices (Commerce, Harbor, River) and 1 GP(Merchant) point per turn. Thats it.

Which mod is this exactly? That definitely doesn't sound like it's something it should be picking most of the time.

 

[Horem]

The mod is unreleased atm. We are still doing some tweaks to it before releasing it to the community. Off to work now, be back in 4-5hrs time, I will have a chat with the rest of the Team and see if erveryone is happy to release. If not atleast send ya a link via PM so you can test/see for yourself.

 

[Siesta Guru]

The mod is unreleased atm. We are still doing some tweaks to it before releasing it to the community. Off to work now, be back in 4-5hrs time, I will have a chat with the rest of the Team and see if erveryone is happy to release. If not atleast send ya a link via PM so you can test/see for yourself.

Alright, let me know when you have it ready. I'm definitely curious what it could be.

 

[V. Soma]

Ok, had a game with AI+v7, Omnibus mod Complete ruleset, Smoother difficulty
emperor, 4-leaf map, standard
me: France, AI: Spain, Brazil, Sumeria

I won by religion, ad1400 - Spain had religion but not much rel. units, only miss. and some inq.s

I saw Spain having 4 settlers at the same time, 3 escorted, searching for settling place in vain, as map was filled...
same with other AIs...

cities

Brazil 7, me 5, Sumer 4, Spain 4

buildings

me 23, Brazil 11, Spain 10, Sumer 0 (?)

districts

me 6, Brazil 3, Spain 2, Sumer 1 (?)

great people

Brazil 4, me 4, Spain 1, Sumer 0

wonders

me 4, AI 0 (???)

 

[Siesta Guru]

Ok, had a game with AI+v7

Thanks for the feedback!
Hmm, Sumer having no districts/buildings doesn't really surprise me, but it's definitely something that needs work. Sumeria, Scythia and to a lesser extent Egypt have this annoying tendency to just spam out their unique improvements and do little else. Will try to do more next version.

Seems the AI religion game is still rather underwhelming in v7. My current tests for v8 are a little more promising, but it might still not be enough, they're still not really getting astrology fast enough, and it's very hard to get them to without breaking other things again. I'll probably have to sacrifice the changes that made them be able to pick archery at least.

Will improve wonder desire a little next version, although it should be noted that it's kind of by design that they shouldn't really bother much with wonders on difficulties below immortal. The AI is terrible at selecting which wonders are good for it, and I can't change anything about that atm. So just not building wonders is probably better for them than wasting their precious hammers on a huey teocalli on their only lake tile. Especially since most wonders this game are either rather underwhelming (great library) or highly specialized (mont st michel). It shouldn't be 0 wonders though, even on 4 players, that looks and feels wrong, even if it may actually be the best way for AIs to keep up.

The settlers.. oh the bloody settlers.. I don't really see this issue going away anytime soon if at all unfortunately. Why it keeps on plopping new settlers out when there's clearly no need for them is one of the great mysteries of the modern era. It feels like there must either be something seriously wrong with how behaviortrees in general work (which I've been suspecting in other areas too), or there is some alternate way besides the settler behaviortree that can end up producing settlers.

 

[Seven05]

Hmm, Sumer having no districts/buildings doesn't really surprise me, but it's definitely something that needs work. Sumeria, Scythia and to a lesser extent Egypt have this annoying tendency to just spam out their unique improvements and do little else. Will try to do more next version.

I fear that will require changes beyond pure AI for now. I was able to reign it in with my personal tweak mod by limiting their placement and it works great but its not an AI change.

The settlers.. oh the bloody settlers.. I don't really see this issue going away anytime soon if at all unfortunately. Why it keeps on plopping new settlers out when there's clearly no need for them is one of the great mysteries of the modern era. It feels like there must either be something seriously wrong with how behaviortrees in general work (which I've been suspecting in other areas too), or there is some alternate way besides the settler behaviortree that can end up producing settlers.

I've been testing out some AI tweaks in isolation and found they do a decent job of settling with the stock AI by tweaking the value of plot yields alone. The problem is that every change has a cascading effect on other changes so fixing one thing breaks another, like your issue with archery vs. astrology. I'm also finding that quite a few values appear to work the opposite of what I expected or seemingly not at all (likely small differences that are hard to see). But I'm going to keep trying to force this square peg into the round hole until they give us the DLL source

I've tried radical changes to the tactics in AIFavoredItems with no visible effect in game unless combined with other changes but even small changes with some PseudoYields make a significant impact.

I have a bunch of changes I'll be happy to share with you so you can test them too, I do everything in SQL though since I find it much easier to manage on the mod level than a bunch of XML files.

 

[Horem]

Alright, let me know when you have it ready. I'm definitely curious what it could be.

This

 

[Siesta Guru]

I fear that will require changes beyond pure AI for now. I was able to reign it in with my personal tweak mod by limiting their placement and it works great but its not an AI change.

I've been testing out some AI tweaks in isolation and found they do a decent job of settling with the stock AI by tweaking the value of plot yields alone. The problem is that every change has a cascading effect on other changes so fixing one thing breaks another, like your issue with archery vs. astrology. I'm also finding that quite a few values appear to work the opposite of what I expected or seemingly not at all (likely small differences that are hard to see). But I'm going to keep trying to force this square peg into the round hole until they give us the DLL source

I've tried radical changes to the tactics in AIFavoredItems with no visible effect in game unless combined with other changes but even small changes with some PseudoYields make a significant impact.

I have a bunch of changes I'll be happy to share with you so you can test them too, I do everything in SQL though since I find it much easier to manage on the mod level than a bunch of XML files.

The cascading effect is indeed a massive pain. Most of the negative effects people have reported here so far are due to that. You give a minor boost to food desire, and suddenly they don't build any culture districts anymore. The old civ 5 flavor system felt a lot nicer in this regard, allowing fine control without blowing everything else up.
I can confirm that some of the plot evaluation stuff appears to work the opposite way, and some require you to set favored to false for some reason. And yeah some other favored items like the favored techs appear to do very little if anything at all, while in the districts system setting them to be favored affects it so much that they'll now build these things every single time.

Yeah I'd love to see your changes, and of course feel free to use some of my stuff. You may at least find some of the 'strategies' to be useful.

This

Awesome! So, having looked at the files, I think it's indeed just that it values the yields highly enough to have it considered one of the best districts in all cases. The one point in food on the citizen yield is probably what pushes it over the edge. I say probably because I can't know for sure right now, the weights I've given in AI+ to value certain yields appear to just modify some already present internal desires that I can't see. Does the AI also prioritize it in vanilla?