Tips Wanted [Deity] - How To Slow Down another AI's Culture-Snowball/Run?

[Pengwin7]

Just finished a Deity game where I painfully watched one AI (Khmer) just absolutely run away with the culture lead.
I try to win in 200 turns... and just couldn't keep him close. By the time I hit Rock Bands (turn 188 with Tamar/Georgia), I had less than 100 foreign tourists to the Khmer's 276 domestic count.
[For clarity, it's not my own "100" that I'm concerned about... obv - that's for me. But is there ways to keep him from getting as high as 276 domestic tourists in 188 turns?!]


I do buy all GreatWorks - as soon as available - from the lead-culture AI.
Khmer barely had any gold or luxuries all game...
I'm guessing it was the pantheon(Sacred Path) + religion(Choral Music) that zipped him up his own culture tree.

[War, pillaging, yes, one answer... it's something I haven't tried, because I've had issue with the grievances damaging other aspects of cultural victory push.]


One interesting reddit thread I found suggests that 'wiping out Religion of the Choral Music AI' (which is/was Khmer here) is a good tact.
But... on Deity, his religion was up 10 turns before mine and his populations are so populous that it seemed like a fruitless thing to try when my faith could be going towards empire/infrastructure.
[Anybody that has played Khmer probably knows well how nicely his population/religion can explode.]

https://www.reddit.com/r/civ/comments/urq876/advanced_tip_for_fast_culture_victories

Are there things people do to SIGNIFICANTLY slow another AI from running away with their own culture/tourism?!?

[For those that like numbers, I didn't catch Khmer for Domestic Tourists until turn 189, when I/Tamar had 275 and he/Khmer had 276. This is obviously way too late to hope to win a fast culture victory, I like to win in sub200 turns on Standard/500.]
The '275 domestic tourists' at turn 189 is a really high number in itself.
I think it's the highest number I've seen in my games... and I like to track the numbers.
First time I've seen Deity AI beat me at sending Rock Bands. His were in my land by turn 186... mine in his at turn 188. And he had locked my Rock Bands out by turn 198. Those are firsts, for me, within 200/500 standard turns.

Spoiler :

 

[Oberinspektor Derrick]

I understand that you have said that you don't like to war much, but the only consistent answer in this case is war, even when considering grievances.
Not using this option is essentially removing a very important tool in your kit to get the job done, which is particularly important if you are trying to push your deity win times to shorter and shorter time frames.
You can somewhat compare the situation here to an AI that builds a big wooden door that stands between you and victory, and the longer you spend your time getting to that door, the stronger that door gets.
And in order to break that door, you need the correct tool.
The correct tool to use in this situation is a sledgehammer to break through, hitting hard and fast to knock it down to reach your victory.
Sure you can technically use a different tool like an axe, a carpenter's hammer or possibly even a drill or a saw to try to gain entry as you slowly grind away at the door, but the most efficient tool here is to just use your sledge sledgehammer and demolish that door immediately.

In order to stop the AI in this case, you want to just take them out (or at the very least cripple them extremely hard, leaving them with just a bad city or two), ideally before they get the chance to run away on culture.
Taking their cities lowers their culture (which fuels their rise in domestic tourist count), lets you pillage them (which sets them back and gives you a ton of faith for more rock bands or GMC purchased units) and lets you actively "fight" their rock bands (they dont get killed, but pushed back to their nearest city).
Screw the diplomatic penalty in this case - You can set up trade and open borders before you go to war (which lasts 30 turns anyway, which is "forever" at this point), and as long as you keep renewing friendships when they run out, most people will stay friendly to you and let you continue trading for open borders.
Either way, going to war against one enemy is usually not killing your diplomatic relations that much, especially if you use a casus belli.
And if that enemy is your only remaining roadblock (only AI that you aren't dominant over), even ruined relations are worth it since you will be winning the CV right after you take them out (since their pool of domestic tourists that prevented the CV, is now gone, thus granting you the win).

Either way the tourism modifier from open borders (25%) is really overrated in this particular case (one AI runaway).
Don't get me wrong, its a really good bonus in almost all cases, but its not worth fretting over and then have an AI run away on culture, just to keep that 25% bonus with everyone (especially not if you're dominant over them already, or about to be).
Either way, trading open borders for 30 turns with others before declaring war, is more than enough to finish off an AI and win the CV.

An additional exploit here is to offer other AI to join your war, in which case you can actually get strengthened relations with them out of that war.
A lot of AI will happily join you in that endeavour for 1g, which has the positive side effect of also killing off any trade routes and open borders with the runaway AI culture civ and your new wartime buddies.

At your stage of the game, bombers and tanks/cavalry units (to take out the city after the bombers reduced them to 1 hp) is usually the quickest way to delete an AI within a handful of turns, and its probably wise to kill them off entirely so that their domestic tourist roadblock disappears completely.
If you want to declare on a runaway AI earlier though (right before they start to become a real runaway), you can "just" cripple them with heavy pillaging and taking a few cities before declaring peace, for a relatively low diplomatic cost.
You probably can't get open borders with them any time soon, but you don't need to since they are now crippled culture wise for the rest of the game (and its "just" 25% anyway).

One interesting reddit thread I found suggests that 'wiping out Religion of the Choral Music AI' (which is/was Khmer here) is a good tact.
But... on Deity, his religion was up 10 turns before mine and his populations are so populous that it seemed like a fruitless thing to try when my faith could be going towards empire/infrastructure.

The correct use of faith here (seeing as you wanna pool up faith for naturalists and rock bands as much as possible) is to blow it on military units.
It's way cheaper and much less tedious than trying to wage religious warfare, which will just bleed you dry on faith even with Yerevan suzerained.

 

[DeckerdJames]

I have built ski resorts all over to prevent a culture victory by another. I have not tried it and I don’t know if the grievances are high for doing it but you could bomb their tourism improvements or build enough of your own that they don’t become dominant.

 

[Oberinspektor Derrick]

I have built ski resorts all over to prevent a culture victory by another. I have not tried it and I don’t know if the grievances are high for doing it but you could bomb their tourism improvements or build enough of your own that they don’t become dominant.

That does not prevent a CV by someone else, as placing ski resorts only increases your foreign tourism pressure, not your domestic tourism (which is fuelled by culture, and what you want to increase to guard against someone else's offensive tourism).
Generally speaking, tourism (as a yield) is your offensive weapon, while culture (as a yield) is your defence.

 

[Pengwin7]

If you want to declare on a runaway AI earlier though (right before they start to become a real runaway), you can "just" cripple them with heavy pillaging and taking a few cities before declaring peace, for a relatively low diplomatic cost.
You probably can't get open borders with them any time soon, but you don't need to since they are now crippled culture wise for the rest of the game (and its "just" 25% anyway).

Yes - this is the part of my game I probably (sigh) have to commit to a 'learn' on.
Generally, the first 55-60 turns of my games are the same - trying to get 2-3 cities to have good holy sites (at least 1 w/shrine).
In my last run w/Tamar (always /500 standard speed), my key marks (per the Era log):
turn 47 - Golden Age at end of Ancient Era
turn 56 - Holy Prophet [*I'm basically using production ONLY towards holy sites/prayers until my religion.]
turn 57 - Religion
turn 60 - Political Philosophy (I assume this is always a key mark for war, with the +4 strength gov)
[I don't have save files to check what I used my first post-HolySite production on... but I did build Apadana - which is a LOT of prod, chops w/Magnus - bc I had relic and wanted to create Relic spots and get Envoys for Tamar. Apadana + WonderHeavy capital is sometimes a strategy-path... and one has to abandon, I think, all other paths (military) - to commit to it fully. I also went harbor + Maus in that same capital to take advantage of Imhotep charges. And it was a great, great game for ME... but it also meant that I couldn't, in this case, commit to military push on Khmer...]


As an estimate, how many units are required to:
i) Take a city - assuming they don't get Ancient Walls up
ii) Take a city - assuming they get walls up and have some defense (couple archers)
iii) Take a city with some heavy defensive terrain (behind a river, or maybe mountains two sides, or coast on two tiles)


As an example, I think this guy (boesthius) is pretty popular for Civ6 videos.
I watched this series hoping for tips... only to find out that he didn't complete it...
I speculate he didn't complete it... because the war on Germany didn't result in any 'keeper' cities and he used a LOT of production/chops on creating military units.
Units created - without cultural benefit - is something you look back on as "Well - I could've used those turns differently".

[2nd to last comment shows he didn't make a "Part 3"...]

*Unfortunately, this is one of the only sub200 culture videos I've ever seen that include 'war'. Maybe there are more out there (?!?)... but I haven't found them.
So I don't have a good tutorial to be doing TWO paths with an empire - balancing a military with continuing on cultural path.

 

[Oberinspektor Derrick]

As an estimate, how many units are required to:
i) Take a city - assuming they don't get Ancient Walls up
ii) Take a city - assuming they get walls up and have some defense (couple archers)
iii) Take a city with some heavy defensive terrain (behind a river, or maybe mountains two sides, or coast on two tiles)

See my response in the other thread you have going, regarding what I'm looking for specifically when invading a target.
The thought process relies on a lot of factors, but with some practice this feels like second nature to evaluate as you go.

That being said, what you are asking for here is an early game rush, whereas the situation you commented above is a late game invasion with the aim of killing off another civ, and those are two different things.
In the latter case you pretty much only need access to bombers to guarantee taking out the AI (they are incapable of defending against it, and they level a city incredibly fast), the rest takes care of itself.

As for an early game rush, your timings regarding a religion already show that you are strong at the early religion opener, and if you want to learn early game rushes I suggest you play around with crusade and ideally an OP civ for a few games to practice taking out your first neighbour (the hardest target in the game, the rest are a cakewalk because you snowball at that stage).
There is probably no better candidate for this than Theodora, because not only are you almost guaranteed a religion, you also get "free" combat strength from just having a religion (+3), and the real kicker - you passively spread when you defeat an enemy unit, making crusade spread in no time and guaranteeing you enormous amounts of free era score (+3 every time you flip a city passively through combat).
You also get tons of free culture from her holy sites, so it lends itself extremely well to practicing a cultural playstyle with warfare mixed in.
Then after you master the religious rush opener, you can transition into other civs that don't have her OP kit, and find out that with some adjustments you can do just as well through abuse of the crusade belief.
Essentially you're just doing your regular religious opener (2-3 cities, straight to HS in all, prayers when needed), then abusing work ethic to pump out tons of cheap units, place them around their city while you "seed" that AI city with crusade, and just jump them to take them out before they can react.

Regarding your question of "how many units", that is impossible to answer in a quick manner, so check out my post in your other thread to pick up the signs you need to look for.
Mostly its about their city combat strength (telling you which units you face and how many you need of your own), their military strength (which tells you what initial resistance will be like and the amount of forces that are hidden in the fog of war), and their science output (telling you how soon you can expect a nasty surprise in terms of higher tech units).
Very generally speaking though, I take at the very least enough warriors +1 to siege their city (having at least one extra lets your rotate warriors in and out easier and still maintain a siege), pre-build a battering ram in case I sense that walls are about to come up (you can tell from the city animation that a wall is being built), and two archers (because worst case, if they get walls and you have not pre-built a battering ram, the archers can rotate to keep shooting the walls every 5 turns and less, which prevents them from repairing the wall).

The most important skill is playing the tactical game though - meaning that you get an intuitive feeling of what a relative difference in combat strength means between units, and being able to calculate what sort of combat strength you get from defending versus attacking (you generally want the AI attacking you when fortified on rough terrain because you trade immensely well then, even on deity).

As an example, I think this guy (boesthius) is pretty popular for Civ6 videos.
I watched this series hoping for tips... only to find out that he didn't complete it...
I speculate he didn't complete it... because the war on Germany didn't result in any 'keeper' cities and he used a LOT of production/chops on creating military units.
Units created - without cultural benefit - is something you look back on as "Well - I could've used those turns differently".

Boesthius isnt the worst of the bunch, but he does indeed do some questionable things as you point out.
Generally I am not a fan of chopping out units early on, because it is an extremely risky strategy that puts you all in against your target.
If it fails its basically game over right there and then, since you cannot recover from such an extreme investment.
Which is why I always advocate for the religious opener because of how consistent it is at wiping out the AI (unless you play some civ that gets an insane early power spike without it, like Nubia, Gaul, Babylon etc.).
I have even used the opener on Deity++ (a mod that simulates two difficulties higher than deity), which goes to show how consistent it is at taking out target civs that have a starting lead (assuming you get an adjacency pantheon, which is the hardest part if you cant get lucky on spawn yields, tribal huts or CS pillaging).

 

[Pengwin7]

Great reply.
Theodora/Crusade is 100% on my list of things to try - with a pivot to culture victory.
I think it would water-down this thread to go back to that... and what other CIVs that works well for.

Few notes for clarity:
1. I have no desire/intent for 'late war' with bombers or other such. (not sure on how that was interpreted - my apologies if my language was cloudy)
2. Yes, definitely possibly meant 'early war', with pillaging where beneficial.*
3. *Tough thing here is, for me, it's very conditional. Like - this SPECIFIC play-through... on retrospect I can say: Khmer was close (15-tiles capital-to-capital... and 10-tiles in a Stonehenge city he forward-settled towards me... AND he was the one who benefitted from his rainforests/SacredPath/Choral Music. With experience... i.e. now! and going-forward!... I will look out for these types of things and want to prepare myself for a way to hurt his development. But... if Khmer was two AI away... 25+ tiles... or wasn't even the Khmer... an early-war would just bee too far a travel. By the time my units arrive... they'd likely be outdated by turns to the quick-start of AI-DeityScience-push. (*I kind of want/hope that the whole Crusade, take-out-neighborAI isn't so "doable"... every game... because then... I'd lose interest if there is a 'routine' to winning sub200 culture games consistently.)


As is... I still haven't seen/found a set of videos where somebody wins Deity CV, consistently, different leaders, *with* intentional early-war. (i.e. 'Routinely')
I'd really like to watch a video like that. I'm a visual learner.

(I'm not doubting any advice... I can vibe that it's possible with somebody as skilled as you... but I also know that I can't/won't learn it unless I see every-single-step-along-the-way... which basically means I need a video to learn it. As an example, one has to watch somebody... eval their target... and see how many - what type of units - and how they place and attach with those units - when they pull back, heal, shuffle placement... etc... I've tried - and I was just incredibly far from being effective... so far... it feels... impossible on deity unless one lucks out or can get that 'help' on war from other-side neighbor of the target).

Thanks again though!

 

[Oberinspektor Derrick]

Few notes for clarity:
1. I have no desire/intent for 'late war' with bombers or other such. (not sure on how that was interpreted - my apologies if my language was cloudy)
2. Yes, definitely possibly meant 'early war', with pillaging where beneficial.*

I dragged up the example with bombers because in a late game scenario like the one you described (where you want to make the t200 mark but a single runaway AI is preventing you from it), bombing them into oblivion while performing more rock concerts is the only proper answer.
Usually you cant tell for sure in the ancient/classical era if someone specific is going to be a runaway in the late game around the t200 mark that is your only remaining roadblock for a CV.
Thus the point of early war (ancient/classical) is generally not to stop someone from being a late game threat, but to set up a snowball for yourself.

Especially with the religious crusade opener, you can afford to spew out a ton of cheap units in record time after establishing your religion, due to work ethic (so that your opportunity cost for waging war is very low).
That production invested in warriors/archers (assuming you take over your neighbour and/or a few city states) pays massive dividends almost immediately, since you are getting free and developed cities (the AI has often built a few districts in them, improved the lands and usually built a monument+granary in them).
Contrast this to building your own settlers and developing those settled cities, and the cost of spending 5-10 turns building warriors and archers is a joke of an investment if you can get 3-5 already developed cities out of it.
And it synergizes well with your peaceful plan anyway - at this stage (after you just founded your religion), there isn't much to build anyway (you have HS in place, and probably not teched CH, TS or campus yet), and the only worthwhile items are really just military units and/or monuments.
Assuming you secured monumentality and have a work ethic/crusade religion and a nearby neighbour that is ripe for attack, there is zero incentive for you to not be at war, because you can just spend that production on a bunch of warriors/archers/battering ram, while you spend your faith on settlers and rotating Moksha to essentially do what you would do on a peaceful play (Moksha also fantastic on a push where your city borders theirs, since you instaheal).
Sure you gotta spend faith on at least one missionary, but pillaging will soon make all that faith back and then some, which fuels more settlers/districts.

Also FYI, assuming you just take out 1 neighbouring AI and not go on a huge rampage later on, the diplomatic cost of taking out one AI is quite manageable (many AI dont care, and the ones that do sober up after a short while as grievances decay fast this early in the game), especially if you already had friendly relations with others before the DOW (and especially if you pay them 1g to join in, they often love you for it).

 

[Oberinspektor Derrick]

I'll see if I can make some screenshots explaining my thought process one day.
Haven't played civ for over a week now, but I still have the autosaves from the game where I played as Mansa Musa (to showcase the crusade opener), and I guess I could take some screenshots from different turns and explain my thought process on a tactical level.
Essentially my goals (tactically) is to either overwhelm a city fast (so not to take losses, usually being the first city that I jump), OR inflict slow but steady damage on the enemy (whether his units or his cities), where I try to take zero losses in units in order to maintain consistent pressure (zero losses to mitigate the defender's advantage).
Health losses are fine as long as I can rotate the unit out and still maintain the siege, but unit losses are only acceptable if they are necessary to gain a large advantage (such as taking a particular city before walls are up).
Farms are your friend for health pillaging purposes, and fortunately the idiot AI tends to build a lot of them right next to their cities, which is usually a big green sign for me saying "please invade".

 

[Pengwin7]

Now - with the Crusade - the whole assumption is that you CAN convert the neighbor AI city with missionary(s) prior to your attack.
And... with Khmer being the discussed-neighbor - or currently I'm playing a game with Inca as neighbors - we are talking 3-4 missionaries... bc the pop is, typically, 7-9 and religion is established there.
[In my current game, playing Eleanor-England, I needed 4 charges to convert a 6 pop Inca city... and the Inca are already returning 3 missionairies to re-charge the city... and have two other neighboring cities, their religion, and Holy Sites - I've typically found/learned that the bigger cities, with earlier established religions, are just going to push missionairies back and undo any conversions.]

Also - as you surely know - a (likely) runaway culture CIV is apt to be that sort of faith/culture/ChoralMusic/high-pop-city type of AI that has the missionairies coming to re-charge a converted city.
[In my current game, the grievances from converting the one city - and being asked NOT to - ... are already being met with disdain from some of the other AI...]

So... if I spend all my faith in trying to convert neighbor city, while producing units, and the attack city isn't converted on units-arrival...
Then what?

 

[Oberinspektor Derrick]

Now - with the Crusade - the whole assumption is that you CAN convert the neighbor AI city with missionary(s) prior to your attack.
And... with Khmer being the discussed-neighbor - or currently I'm playing a game with Inca as neighbors - we are talking 3-4 missionaries... bc the pop is, typically, 7-9 and religion is established there.
[In my current game, playing Eleanor-England, I needed 4 charges to convert a 6 pop Inca city... and the Inca are already returning 3 missionairies to re-charge the city... and have two other neighboring cities, their religion, and Holy Sites - I've typically found/learned that the bigger cities, with earlier established religions, are just going to push missionairies back and undo any conversions.]

Also - as you surely know - a (likely) runaway culture CIV is apt to be that sort of faith/culture/ChoralMusic/high-pop-city type of AI that has the missionairies coming to re-charge a converted city.
[In my current game, the grievances from converting the one city - and being asked NOT to - ... are already being met with disdain from some of the other AI...]

So... if I spend all my faith in trying to convert neighbor city, while producing units, and the attack city isn't converted on units-arrival...
Then what?

The assumption holds though.
As long as you have decent faith output and a shrine while getting your religion around T50-60 as you say, you are in a position where you can convert the nearest city.
The AI usually doesnt have much if any missionaries at this stage, and they are not converting their own cities beyond the minimum stage anyway (and religion is in place in most of their cities when they founded one in a city with a HS) but rather they use any missionaries they do have (which is rare) to spread their religion to non-religious cities.
The only thing you need to worry about when jumping the nearest city is to convert it (should take 1-3 turns depending on whether you use 1 or 2 missionaries), while having your units outside the city ready to jump it.
Once the city is flipped you immediately jump in, and the AI will have no time to react and re-convert it by sending in its own missionaries (in fact, doing so would have their missionaries at risk of being killed, which would aid your religious spread even further).
Missionaries are either way subject to zone of control from your military units, and cannot spread back immediately without losing a turn, so even if they try to you just kill them with military units as they stand there helplessly.

Remember, you are not waging religious warfare here (where they will convert back), you are waging an actual war, and missionaries just cannot re-convert their cities when you have military units there.
At most you need 2 missionaries around T60ish to convert the nearby city in the case where someone like Khmer has a high pop city, but thats the absolute maximum.
Once you invade you only need to pillage a bit and grab the city, and the rest of the cities tend to fall pretty fast (though in all honesty, the second city is usually the hardest to take out, because you lost the element of surprise and you probably struggle on loyalty in the first city - this is where you need to pay attention and not mess up).
Further cities can easily be converted by running a missionary on top of a warrior that is sieging the city though.
Even if the next city isnt converted yet for crusade, you should have multiple warriors that you can rotate in and out while the missionary converts, and once converted its just a matter of a few turns until the city falls.
If you have Moksha in the first city that you took, you can probably just rotate wounded warriors back in to instaheal them and keep up the pressure.

Again, I'll see if I can make some screenshots to explain the process, as it's rather straightforward when you get good practice at it.
After that you can pivot into fully peaceful play, enjoy your guaranteed golden age, the free developed real estate you just took, and no future contest on nearby settling spots.

 

[Pengwin7]

I love your tips - and I did a huge trial last night - very, very carefully.
I think it would derail this thread to explain what I like and dislike about it... so I won't post it here.
But it is appreciated.


If the runaway culture leader is NOT a reachable neighbor by military... this war effort, I think, does/can derail from expansion.
Every unit produced for the war effort could've gone to a different infrastructure/builder/settler.
And where I normally take the 30% savings on missionary/apostle... to go WorkEthic/Crusade means you aren't saving on your first of those... losing faith.
Can't say I'm a big fan... because in an 8-leader game, there's only 1/7 chance... maybe 2/7... that the culture leader is war-able early.


[will post my Theodora WorkEthic/Crusade effort in my sub200 thread for your review]