Make City Flipping an option

[DeepO]

No problem about the quotes, it just confused me a little

Originally posted by darthx86
I disagree... somewhat. While the diplomatic goal is already built in, the editing of units, GW, SW, etc. would be much more difficult for the AI to work around.

No it is not. I don't know exactly how the AI works, but speaking from my own AI programming experience (which was no game AI, BTW) I expect this to be a rule based, database driven AI. So; general concepts are hard coded, while specific parameters can be changed.

An example: the AI knows that it is best to counter defensive units with mobile units. It doesn't know the reason for it, it was hard coded by Soren to do it like this. However, when it has to pick which unit it uses to attack a defender, it does this based on changeable parameters. It will use knigths to attack spearmen over horseman, as knights have a far better chance of winning.

If, however, you edit the attack value of a horseman to be 8, while a knight still has 4, it will pick the horseman over a knight to attack. This example might seem unrealistic, but it shows that the AI is somewhat adaptable to the parameters of the current game.

(Back to culture flip.) Since culture flip is confusing and (at best) random, human players can't plan on it. If human players can't plan for it, then neither can the AI.

You're not saying you don't plan on culture flipping, are you? I always plan my border cities to have a larger culture than normal, just to try to take an enemy city if possible, and to prevent them from deflecting. Further, all of my newly acquired cities are heavily garrisoned, to improve my chances of keeping them. Cities with wonders are even more heavily garrisoned, as the culture in those was far greater, and the risk is higher.
This is a kind of planning that I'm sure is also included in the AI, and it does not depend that much on parameters. Therefore, it is much more basic to the AI than editing attack values is.

Of course, a quick fix could be to change priorities: If culture flipping is disabled, building libraries in border cities is not more important than in in-land cities. However, this would unbalance the AI, it would not function as good as it does now. In order to do it right, you have to change the entire system, which takes nearly as long as designing it from scratch...

DeepO

[Stuie]

Originally posted by DeepO
erhm... I was asking for knowing the formula, not asking to be able to change it.

This is the closest Firaxis has come to revealing the formula:

http://apolyton.net/forums/showthread.php?s=&threadid=39825

By the way, that's from January, so definitely pre-1.21f.

[DeepO]

Thanks, Stuie. I knew that thread, but that has some hints in it, not a real formula. Besides, it seems like Dan was confused himself on the complexity, as it simply doesn't work out...

Oh where are the times where you got the exact formula for corruption / waste / scoring right in the civilopedia....

DeepO

[Stuie]

Still, it helped me strategize for protecting against and agressively pursuing culture flips. I don't find the concept to be confusing or random, unlike darthx86. I can gauge with reasonable accuracy those cities which will be easy to flip, those which will be difficult, and those which are impossible. Likewise, I usually have a pretty good idea as to which of my cities are susceptible to a flip. Part of that is from Dan's post, part from many hours playing Civ3. Either way, I find it to be a lot less random than combat....

[Geekinstein]

quote:

Originally posted by DeepO
I love culture flipping (just voicing my support), but I can understand people want to turn it off. However, wouldn't it be a solution if you just get an indication how likely it is that a certain city flips? E.g. if it is more than 50%, display a number saying 'flipping is possible', if it is 80% displaying 'flipping is imminent(sp?)'. That should mean you can still do something about it, or take precautions to not loose too many troops if it does.

Besides, is there an exact formula known for the flipping chance? I'd love to see it!

OK, I think I agree with this. How about one of those stupid advisors popping up with something inane like, "O great and all knowing one *grovel* the governor of city Flippinbach tells us that the populace is growing restless and is about to flip out. Better build a few bars, or better yet a library." ??

[DeepO]

Originally posted by Stuie
Still, it helped me strategize for protecting against and agressively pursuing culture flips. I don't find the concept to be confusing or random, unlike darthx86. I can gauge with reasonable accuracy those cities which will be easy to flip, those which will be difficult, and those which are impossible. Likewise, I usually have a pretty good idea as to which of my cities are susceptible to a flip. Part of that is from Dan's post, part from many hours playing Civ3. Either way, I find it to be a lot less random than combat....

I totally agree. But still, an exact formula would be nice

Geekenstein, this was what I had in mind. But, if advisors start denouncing all my possibly flipping cities, I'll soon shut them up (provided there's an option for that). Maybe something less intrusive, in the same spirit.

DeepO

[Austin]

It's been brought up in a number of threads about culture flipping that there "has never been a historical example of a major culture flip"

I would just like to point out that a large number of Rome's aquisitions, especially in the east, came about as a result of culture flipping.

For example, Asia province, Bithnia and Cyrene were ALL "culture flips", the rulers of these kingdoms willed them to Rome because they felt that their people would benefit from being affiliated with Rome. Asia province in particular was the "milk cow" of the Roman republic.

I know that Pontus was a military aquisition, but I think that Cappadocia and Cilicia might also have been "flips". That's the equivelant of all of modern day Turkey, and it was one of the most economically powerfull parts of the ancient world.

Did the Seleucids actually fight Pompey for Syria, or did it roll over too? I'm pretty sure all those little guys like Oshroene simply rolled over for Rome without much of a fight.

Austin

[Stuie]

Good points, Austin. Also reminds me of the many tales of ancient cities under siege that were betrayed by small groups in the city that had ties to the besiegers. They would unlock the gates and let the besiegers in, no battle necessary. Flip.

Then there are cases such as Hong Kong (a British possession) being handed over to China. Sure it was a pre-arranged deal, but still.... it flipped from England to China.

[Miznia]

[retarded post]

[lockstep]

Originally posted by DeepO
wouldn't it be a solution if you just get an indication how likely it is that a certain city flips? E.g. if it is more than 50%, display a number saying 'flipping is possible', if it is 80% displaying 'flipping is imminent(sp?)'. That should mean you can still do something about it, or take precautions to not loose too many troops if it does.

Besides, is there an exact formula known for the flipping chance? I'd love to see it!

DeepO

I'm fairly sure there IS an exact formula for city defection - Firaxis just didn't reveal it, which was a bad idea IMO. What I'd like to see (perhaps in the PTW expansion) is a defection info box in the city screen (similar to the pollution box) that shows the probability of a cultural flip within, say, the next ten turns.

EDIT: And Firaxis, please add the exact defection formula to the civilopedia.

[Catt]

Originally posted by Miznia
A compromise:

Just let me ban cities from flipping TO me, and let me keep my score.

Aren't you always able to choose "We don't want XXX. Rebuff the rebels." when the military advisor pops up to tell you that so-and-so wants to join your empire?

[Fitz]

Here is a post-1.21 comment by Soren about the formula. http://apolyton.net/forums/showthrea...hlight=culture

If I understood him correctly the formula works like this:

Chance of flip = ({[(# of forign nationals)+(# of square in radii under forign control)] * [2 if more culture in city by foriegn civ, otherwise 1] * [2 if in civil disorder, 1/2 if in WLTK day, otherwise one]} - {1 per garrisoning military unit}) * ({Foriegn civ total Culture} / {Your civ total culture}) / (Capitol distance factor {500 to 8000, average 2000})

I imagine this chance of flip is determined on a civ by civ basis, so if you are Egypt, and have a city with 2 persians in it, and two city radii squares under Greek control, there would be an equal chance of a flip to either, not a cumalative flip chance with a determination of who it goes to after. Could be wrong though.

Anyhow, the basic concept becomes you need 1 military unit to stop the flip for each forign national or controlled square, 2 if the other civ has ever produced more culture in that city than you.

As to the chance of a flip, with equal culture and equidistant capitols, it should be around .05% per uncountered national/square (1/2000). Of course, that varies greatly depending on your ratio of total culture and distance to capitols.

[lockstep]

Thanks for the link, Fitz.

[Cyclotron]

Originally posted by darthx86
(Back to culture flip.) Since culture flip is confusing and (at best) random, human players can't plan on it. If human players can't plan for it, then neither can the AI. I agree that culture is an important part of the game, but it still has an impact without the culture flip (namely all diplomatic relations).

Combat is random. Everything in a game must be random, or you would know the outcome ahead of time. Culture Flipping isn't confusing either, because we know the general formulas that go along with it. In terms of what we know about it, flipping is much less "confusing" than diplomacy or espionage.

Human players can plan for it. Human players should plan around it. If by "plan" you mean "know the outcome beforehand," then I could say with equal correctness that combat in any Civ game ever can't be planned for.

As for the impact of culture, borders are important only with city culture, not national culture. In addition, your point about the AI's diplo relations merely strengthens the argument that when flipping is disabled the AI will continue to act as if it had not been.

But back to my original point. I believe that the more options to customize a game, the better. Since the game is supposed to be fun, and since this version of Civ 3 is single player only, who cares if enabling an option unbalances the game?

When the expansion pack comes out, we'll talk again.

There has to be a logical limit. You can't please everyone with a game, and you can't try to by making every feature in the game capable of being turned off. There is virtually no chance that Firaxis will spend time adding a toggle to remove a critical part of the game.

[DeepO]

Fitz, thanks a lot for the link! Somehow, I missed that thread... And even if your formula isn't completely exact yet -- it is as exact as the given info allows, but the precise calculation of the capital disctance factor remains unknown -- I've put it on a post-it and stuck it to my monitor until I know it by heart. Great info!

DeepO