An observation about Cultural Reversion

[Golden Bear]

I guess that I am one of the few that is not too bugged about culture flipping. I actually think that it adds a lot to game. After reading the strategy note on culture flipping I have been able to handle it easily.

however...

One of my observations is that a city is almost guaranteed to revert if the human player stacks a ton of units, like units that are healing or a stack of bombers in a city which has recently been conquered. I have tested this repeatedly by rerunning the turn (from autosave). The first time, I was just POd about losing 8 or 10 wounded cavalry that were going to heal while pacifying a city so I decided to cheat. When I replayed, I took all the units out of the city and left a single infantry in it. To my surprise, the city NEVER reverted. Since then, I have tried the experiment in every game that I have played at one time or another - pile a bunch of units in a recently conquered city, no matter how big it is and wait to see if it reverts. Chances are pretty good that it will. Replay the turn and take the units out.

Just an observation. It might help people to avoid losing cities on a flip.

Golden Bear

[Arrian]

Hmm, I've stacked wounded troops in a newly captured city many times w/o it flipping, but I have definitely noticed the "reload, remove troops from city - no flip" thing. This was a couple of patches ago, when I was still trying to figure out how this stuff worked. I would remove all troops, even the infantry, and no flip. It changes something in the random number sequence for the turn, I think. Anyway, if I feel a city is a flip risk (rare, since I tend to be heavy on culture), I raze the sucker and rebuild one of my own. I often bring settlers along with my invasion forces, particularly late in the game.

-Arrian

[Demerzel]

^^ was back in early patch days.

this idea still works for 1.29f, I had a damn city flip on me regardless what I did. As I'm about to go out into the rain to get some food, I thought sod it, let's see what will keep the city for me.

Building a temple did not. Stacking large numbers of troops in the city did not. Removing all my troops from the city did.

Are the citizens of this rapidly shrinking former chinese city really that annoyed with hosting troops that when I remove them, they get all happy and don't need to culture flip back to the damn chinese? Or are they just particulary stupid and think they're now free?

This strategy seems to be quite successful in avoiding culture flips of recently conquered border cities when the front isn't moving that fast. If the city does actually flip too, at least I've got troops next to it safe from being "vanished" and ready to conquer it back.

Anyway, what I wanted to know is, is there any particular reason why removing troops actually prevents a c.f. whereas piling troops in does not?

Apologies if the answer is buried deep in this thread ( http://apolyton.net/forums/showthrea...threadid=39825 ) but its 8 pages or so long and as i said im hungry and off to restock my cupboard.

[Dis]

this is proof that the game does cheat.

I can call it nothing else. Wiping out offensive forces like that.

[DeepO]

Originally posted by Demerzel

Anyway, what I wanted to know is, is there any particular reason why removing troops actually prevents a c.f. whereas piling troops in does not?

The answer is quite simple. The flipping percentages are always quite low, one of the highest chances you can get in a reasonable game is 10% (and this means you've got no culture in your empire, or are playing to let a city flip instead of protecting it).

So, at most there is just a 1 in 10 chance of flipping. If, at that moment, you reload and move all your troops out, you do two things: 1) you worsen the percentage chance (e.g. from 10 to 11%) because there aren't that many troops available. 2) you change the random number that is compared to the flipchance, which most likely is better than what you got.

an example: a given city has a 10% chance of flipping, with 5 troops inside. at the beginning of the turn, a number is generated, say 4 (on a scale of 100), and compared to the chance. 4 is smaller then 10, so the city flips.
You reload, and move your troops out. The chance becomes e.g. 11%. Again, another random number is generated (the situation has changed), e.g. 64 (out of 100). This is again compared to the flip chance, where 64 is bigger then 11, so the city stays yours.

So, while the chance becomes worse, you get 2 different shots at it when reloading. Note that thiis will also happen if you move e.g. a completely other unit into enemy territory, or if you move one unit differently then the first time: the RNG seed changes, thus you change the flipping behaviour of that city. It is false that it prevents flips, as if you test a few games where you don't move any troops into cities, the number of cities that flip will certainly be higher...

But again, remember that troops are the smallest factor of preventing flips, it is the only thing that will get the chance to zero, but total culture of your empire, number of foreign citizens and foreign tiles, and WLTKD are the biggest factors. If you make sure these are good, adding 1 or 2 troops will prevent all flips.

DeepO

[Demerzel]

I'm topdog at culture ( first time thats happened since i played emperor ) and my culture is about twice as good as the nearest AI the chinese who I'm fighting. so that might help i guess but obviously i need to either shift the front lines forward some ( which takes so long in the middle ages ) or starve meh new citizens a little more

cheers for the explanation. much appreciated.

[DeepO]

Well, twice the culture as the next AI is good on Emperor, that should keep flips to a minimum. So... you need to do something about those borders asap, these will be the easiest to deal with.

Do keep in mind that a resisting citizens counts double, so quelling the resistance fast is certainly also important... Starving can help, but I don't ever do it on purpose, or I should be in despotism and can use the foreign populace to rush cultural buildings (to help with the borders). It works too slowly, and after adding troops is the least effective way to avoid cultural flips. But there is no denying that it can help in the long run.

Oh, and at all costs avoid being in disorder, it will more or less double the chances (while WLTKD halves it).

DeepO

[aahz_capone]

I tend to do the same thing DeepO. However, I remember reading somewhere that a patch made it that you could flood a city with so many troops it will never flip (maybe this wasn't for conquered cities.)

I also remeber a thread whinning about the following but I just can't keep from saying it again: During a culture flip (or GIVING a city to someone else) you loose all you units in there. that sucks the sweat of a dead dog's paws. can those useles peasants REALLY have killed my mech inf off? I think not. And peacefully giving the city away should atleast count for something.

btw, as long as I'm on about it, when rushing using city pop in despotism, the manual says that you scare people away from your city to live somewhere else after making them work so hard. Who actually believes this? They don't go anywhere else, that much is clear from the city sizes around it. Nor do they defect to other civs. Lets be frank and admit that we just work the to death á la Stalin.

[DeepO]

aanhz_capone, I'm not going to comment on your 'troops shouldn't die' thing, this has been discussed to death in here. Let's say that I'm a big supporter of the idea of Culture flips, and that I know in large how it happens, and how to deal with it. This doesn't mean I agree with all the implementation details. It never bothers me, though, as it is extremely rare that I lose a bunch of troops in a flip.

As to the troops: it was possible from the first patch on to use many troops to zero the chances. However, as mentioned in the Culture Flip Formula thread (look at page 4) you do not want to do it blindly. In some cases you need over 100 troops to negate the chances, in others only a few are enough. Garrison troops in a city is not the best way to counter your culture flip problems, on the contrary, it is the worst.

DeepO

[Dis]

I'm not sure what level you guys play on, but I've not had a problem with culture flips.

Before I never commented on it because most of my warfare was in the ancient age. But lately I've been doing war in the modern or late industrial age. And I have been taking big cities.

But many of my cities haven't been that big because of several reasons. First off I use massive bombardment sometimes getting the cities below pop. of 7. Second I'm in communism often and rush build temples, libraries after reistance ends. Thirdly I starve the population down. Ethnic cleansing if you will. Fourthly I vacate most of my attack force after the resistance has ended. Fifthly I have huge amounts of culture in my empire. Even cities close to the enemies capitol have not flipped.

[Playa del Carmen]

I haven't had much of a problem with culture flipping either. Usually, if anything's going to flip, it's going to be the AI's city flipping to me. (I play on Regent, so that may make a difference .) Usually, the only time I lose cities TO the AI is when I capture one of their cities and just can't hold it.

So I really haven't had a chance to see what makes culture flipping so annoying. It adds a nice, different dimension to the game. One thing you learn quickly about captured AI cities is that you CANNOT just leave them on their own and forget about them (especially if they're close to the AI's capital). You'll have to take some intiative on your part and either rush improvements if you have the money or stack it with troops or something.

[DeepO]

Well, CF is mechanism to make the game balanced, so that warring alone won't cut it. If you are naturally a balanced player, you won't ever have problems with it... but people who see Civ3 as a war game are pissed because they need to divert some attention to the cultural part of the game, and most likely lost quite a few cities and troops when they didn't.

What level is concerned: it has no direct impact on the culture flipping, but of course, it becomes harder to build up culture on higher levels. So, when you go up, CF becomes more of an issue

DeepO

[Demerzel]

it's especially hard on higher difficulty levels when you don't choose to play a religious civ. i often play the romans and so find my culture seriously behind the others, not that I'm not trying to keep up but they get to build most GWs before me and temples have to fight with legions for precedence of building.

[fittstim]

Originally posted by Dissident
this is proof that the game does cheat.

I can call it nothing else. Wiping out offensive forces like that.

This isn't cheating. It's just Firaxis' way of informing Americans of the oft-forgotten 4th Amendment to the US Constitution which explicitely forbids the quartering of military soldiers in civilian households.

See. Their just trying to educate people.

[DeepO]

Demerzel, have you tried the ultra early temple yet? with the Persians (I don't play the Romans often, but the Persians are amongst my favorites) you have the same problem of choosing between immortals and temples, but you don't have to: build veteran warriors instead, build temples in between (rushing where possible), and once you feel like attacking connect the iron, upgrade everything, and of you go. Doing it like that means you'll have both culture and a good army, and you don't have to worry about culture before immortals (or legions) become obsolete. Plus, if you upgrade and plan on that from the beginning, there is a good chance your army will be a lot bigger: it's easier to safe on cash then it is to build 30 shields units in despotism.

DeepO

[Sarxis]

Originally posted by Golden Bear
...culture flipping. I actually think that it adds a lot to game.

I wouldn't go so far as to say 'a lot'.

[Mattaba]

If you just put a large enough garrison in the city it will never flip.
I mostly have good culture and with 50 units in a town sized 15, flip chances are zero or almost zero and I always starve new conquered towns.

[Puupertti Ruma]

I also use massive bombardments before taking a modern or indrustrious city, and I we're always bit worried on how effectively bombardment ate citizens. With 'only' 20 artillery units you could turn 20 city into rubbles no more than two or three turns.

I have noticed, though, that in v. 1.29f the possibility to hit on a protectin unit garrisoning in a city has risen dramatically. When I had to have at least 20-30 and lot's of good luck to reduce vet infrantry to 1 Hp, it now takes 'only' 15-20 artys, and just slight luck. Of course the city will still suffer but not even nearly so much.

I think that is a step forwards in Civ3 warfare.