1.18 Civics Changes

[AOS9001]

the AI merely thinks that these civics are too strong

Seconding Sultan. These two civics are fine as they are, it's just the historically correct alternatives are not attractive enough to the AI... or player, in some cases.

 

[Leoreth]

Had a test run with 0 AI value for building production modifiers. The outcome was that Caste System lost popularity to a mix of Slavery and Manorialism, but Citizenship remained dominant. So they were part of the problem but not the only cause.

 

[Apple111111]

Had a test run with 0 AI value for building production modifiers. The outcome was that Caste System lost popularity to a mix of Slavery and Manorialism, but Citizenship remained dominant. So they were part of the problem but not the only cause.

I wonder if the AI properly considers the value of happiness for state religion as much as it would consider general happiness, because if not, theocracy would be notably undervalued.

I’m imagining the ai doesn’t really care about the military food production that vassalage has either.

So the problem might not lie in citizenship itself, but the low favouring of other civics in the same branch. It’s also possible that the ai considers gold-whipping very high-priority. (Perhaps on the same level of population-whipping)

 

[Leoreth]

Yeah, that is also a possibility. I will do a more in depth evaluation of the civic values.

 

[AOS9001]

I have noticed the AI will gold whip hordes of units when it is threatened.

 

[banzay13]

I have noticed the AI will gold whip hordes of units when it is threatened.

It always was so. Just not much civs was running citizenship in ME

 

[Dracosolon]

Speaking of Citizenship, does anyone use spies to bribe barbarians with it? Does the AI know how to use that mechanism?

 

[banzay13]

Speaking of Citizenship, does anyone use spies to bribe barbarians with it? Does the AI know how to use that mechanism?

Even I don't know how to used it
Especially for mounted units

 

[Leoreth]

Speaking of Citizenship, does anyone use spies to bribe barbarians with it? Does the AI know how to use that mechanism?

No, the AI won't use it, and it won't factor that ability into its selection of the civic either.

 

[AOS9001]

Speaking of Citizenship, does anyone use spies to bribe barbarians with it?

Only when I'm playing Byzantium, and only then for Bulgarian lancers. And even then, 100+ gold is a hefty price to pay.

 

[banzay13]

Only when I'm playing Byzantium, and only then for Bulgarian lancers. And even then, 100+ gold is a hefty price to pay.

Should spy and lancer be on same title? How it works?

 

[AOS9001]

Should spy and lancer be on same title? How it works?

Yes, the spy needs to be on the same tile as the barbarians you want to bribe. You can choose which barb in the stack you want to bribe. It's actually very annoying because getting a spy to the barb tile is harder than it sounds, especially because barbs pillage every road they find, and many barbs have Steppe Adaptation, making them teleport around the map.

 

[SultanRedSnake]

I have noticed the AI will gold whip hordes of units when it is threatened.

Looking at you, Cordoba.

 

[SultanRedSnake]

The Middle Ages are back in Medieval Europe. The canonical Monarchy-Elective/Vassalage/Manorialism combo is now the AI meta once again. Citizenship fad is over.

Redistribution has seen massive adoption across the board. I would attribute this to both moving Reg Trade back in tech tree and increasing AI Civic Food value, no? Personal Sidebar: Merchant Trade still Meh.

 

[triplebassist]

One civic I've been thinking about recently is Syncretism. I think it works well as far as gameplay goes, but I struggle with what it's supposed to represent. The obvious thoughts would be Hindu-Buddhist kingdoms in SE Asia and the interplay of indigenous religions with Catholicism and Islam, but neither of those fit mechanically since Hindus and Buddhists tolerate each other and there's no representation of pagan religions once an organized religion is spread to a city. Because of that, it feels like the civic exists for the Kushans and Korea and not really anybody else. I'm not sure whether it would be better to go back to the Tolerance civic that we had on the smaller map or try to find more situations where syncretism makes logical and gameplay sense--the only other example I can think of is a China with a lot of Buddhist cities.

 

[SultanRedSnake]

Latest Civic Report: Redistribution is the new global favorite for first 600 years or so. Instances of civs reverting to default civics (mainly Reciprocity) is also commonplace. The following snapshots of a 600AD/Regent/Normal autoplay may be considered typical.

• 1210 - 29 Civs / 16 Redist / 7 Recip
• 1500 - 31 Civs / 7 Redist / 10 Recip / - Companies have been widely discovered for some time now; was expecting widespread switch to Reg Trade. Several former Redist civs have been on Merch Trade
• 1550 - bookmarked this date just to make a brief note: Portugal has had Companies for 10 turns now and is still on Merch Trade. They also love Republic now (almost every game if not every game). If Reg Trade ends up staying at a later tech I'd advise making Alfonso's fave civic Monarchy and Joao's Reg Trade.
• 1565 - suddenly noticed several four instances of Personalism that were definite reverts (Mali, France, Japan, Norse). France had founded Protestantism a few years prior so I switched to them to check log. Curiously, the last "Revolution has begun entry" was in 620AD. Same went for Norse, which had converted to Prot. and was also on Personalism; last Revolution was from 880AD. This seems to indicate that conversion caused the civic change, but this didnt seem to happen to the other four converts. I checked Japan since conversion can be ruled out as a factor; last Revolution was two turns prior (Merch to Reg Trade).
• Loaded 1500 save - civs in question had been already been running Personalism by this point. Not sure what to make of any of this now. One thing to note is that this means Japan didn't (or couldn't?) switch from Personalism during the aforementioned Revolution.
• Loaded 1375 save - at this moment I I realized these civs started with Personalism #Whoopsie!. Still worth noting that they still hadn't switched when alternatives became available.
• 1600 - Redistribution has by now been moved on from, mostly in favor of Reg Trade. I highlight France again, which had adopted Fanaticism (makes sense, Prot shrine and at war with several Heretics) in 1565 but stayed on Personalism and Recip. Surely Anarchy is factored into AI decisions, however on Normal speed switching two civics is only one turn anyway so, like, WTH France? Civic Upkeep would remain constant from a change to Vassalage, and only marginally increase with Reg Trade which of course would be offset by capital commerce (players knows, maybe AI doesnt?)
• 1650 - general trend still in effect, paused here for France again: was expecting at least a switch to Bureaucracy (Louis' favorite) after Statecraft but not as of yet.
• 1677 - France to Bureaucracy but still on Recip even with a Stab malice now in effect.
• 1700 - Ending here. Individualism and Free Ent. starting to appear. Slavery going strong w/ 11 out of about 20 civs. Recip. still run by France, HRE, Vietnam, Java, Malay, Norse, and China. Also good to see Vassalage not being automatically dumped for Bureaucracy anymore. Italy (subject of Spain; reduced to Milan and Florence) caught me eye with a whacky combo of Rep/Vass/Caste/Reg/Fan/Kin.

I'm trying to track down the discussion that contributed to Reg. Trade being moved to Companies. IIRC the intent was to give Merchant Trade a longer timeline or was it just too widely adopted? Regardless it might be worth considering move it back one column. (My personal choice would be Patronage since that would also un-nerf by dearest Portugal - more seriously its no more than a prereq for most civs). Other notes: Netherlands to start with Thalassocracy, Ottos w/ Syncretism or Clerg, Elective and Merch Trade still have space for improvement.

Last, I recognize the need to put together a more organized format and methodology for posts like these so thanks for putting up with it meanwhile.

 

[AOS9001]

It's possible the AI is avoiding merchant trade while running vassalage, they are not compatible civics after all (though I'm not sure why).

And Reg. Trade got moved to Companies because the AI was making super science cities with National College + Reg. Trade in the year 1000. Reg. trade was pushed to the late middle ages to stop such super science cities from being made too early and snowballing a civ (cough Korea cough) into the Renaissance prematurely.

 

[SultanRedSnake]

It's possible the AI is avoiding merchant trade while running vassalage, they are not compatible civics after all (though I'm not sure why).

Good one. Though I don't have it firmly in mind as to how the AI calcs work (if at all) with respect to stability (time for a Forum dive). Just for Sheez-and-Geez my stab would be that having free-wheeling, cash-chucking used-car salesmen weaving in and out of your rigidly stratified LARP fantasy leads to obvious calamities (randos getting ripped off, peasants getting uppity ideas, heralds and minstrels spamming ox-cart insurance offers).

And Reg. Trade got moved to Companies because the AI was making super science cities with National College + Reg. Trade in the year 1000. Reg. trade was pushed to the late middle ages to stop such super science cities from being made too early and snowballing a civ (cough Korea cough) into the Renaissance prematurely.

Ahhhhhhhhhhhh that's what it was - kind of a Duh moment that I didn't even suppose it was a balance issue. I recall agreeing 100% in principle that the overserved phenomenon warranted attention. Now that we've seen the change in action would anyone agree perhaps a different method to tackle the issue would be called for? Surely messing with the Building on the tech tree would risk far fewer unforseeable and/or wide-reaching consequences as opposed to shifting a Civic. In other words perhaps we swung too heavy a cudgel at this nuisance.

 

[Alexius08]

I've been playing as Italy and Sweden recently. Playing as either of them became much harder when extra food no longer went to unit production while running vassalage, and monarchy's bonus for producing with food has been nerfed, now that it only applies to settlers and workers. Despotism is still a must-have for building up a military strong enough for achieving their area-based UHV goals.

 

[SultanRedSnake]

I tried a few crude stunts to try to find a tipping point in disfavor to Redistribution's appeal. I didn't try a whole lot and thus no breakthroughs to share other than returning a null result on my main hypothesis that the 12/7 commit "Civics AI: increase food value" would be a certain source of insight on the matter if not the primary factor. In fact, I went further and reduced the value in steps down to 1 (release value is 4 - increased from 3 in commit) without producing an evident change in behavior as expected. Next move was double the Trade income penalty and give Merchant Trade some juice, doubling commerce to 2 and throwing in double merchant slots (I'm limited in my competence with XML so have to stick with easy copy/pastes). Big Meh from AI, no bites. I'd intended to play around with a possible new home for Regulated Trade (back one tech column to split the difference - considered Patronage a safe start as its close to the vertical middle). Expected and certain outcome was hasty adoption as with prior versions, but observations showed lethargic response at best from all but the most sensible and familiar cases. *However*, I suspect a major monkey wrench present in the machine was playing further tricks with the AI's range of decisions and thus muddying the results among other potential unknown effects: the "forced" reverts to Reciprocity. Could it some how be the case that the "forced" switch to default is related, perhaps adjacent to the ostensibly conscious preference for Redist? What raises this suspicion is that the revert has been observed occurring alongside a switch in another Civic Option, but then following a reload and rerun the same civ will end up with Redist. instead.

As a last YOLO I doubled Merchant Trade extra routes to 2 *and* winged in a 100% yield bonus. Just when once again no results seemed evident a few conspicuous holdouts prompted an insight: the civs who stuck with Merchant Trade either spawned in 600 with multiple cities, or flipped several cities on birth. That is, these actually had *some* trade commerce upon which to base decisions within the brief Free Switch window. The common single-city, blank slate spawn might see a max of 1-3 trade commerce on these first few turns if any at all, yet have likely just poofed in a nice swathe of juicy Food tiles with their Advanced Start points. Too these add the instability add the spectre of instability alongside Manoralism. Viewed from this angle and taking away the human long view, we can now see an AI no brainer. From here, even if sufficient expansion has taken place to unlock notable latent wealth from trade, if instability wasn't a consideration on spawn, it probably is now, and of course Anarchy is now added to the cost.

Hehe yeesh so now what?