Civ Spit

[The Andy-Man]

Does ANYONE know what causes a civ to split?

Some one should do an experiment, looking into military power, population etc...... No body seems to know!

[Caesar the Great]

How dare you spit on civ! HANG HIM!! HANG HIM!!

[The Andy-Man]

ok so i cant speeel.

i ment like, some one find an old save game of theirs, (someone with plenty of free time) and use the cheat mode to take over civ capitals. perhaps if this one done with losts of save games, and then look for something in common....

[Hoek]

I think it primarily has to do with the size of the empire...if there is a large overall population and many cities, there is a decent (though not definite) chance of it happening.

[The Andy-Man]

i dont think so. alot of people have said that they didn't split the empire with 50 cities, but on the next turn the civ with 10 cities did split....

perhaps it is the ration of people to cities, ie, a 10 citry nation with 10million people will be more likly to split then a 20 city nation with 15 million.

[Blaupanzer]

I think we just discussed this in another thread. (See "Prevent them to escape" for a much more in depth analysis of this.) Anyway, the consensus seems to have been:

1) The splitting candidate's capitol must fall.
2) The capturer must have a smaller number of cities (and possibly be lower on the powergraph than the capitol-loser???). How much smaller has been subject to debate.
3) There must be less than seven civs in the game at the time of the fall. (This is definite, no split if no room on the civ list.)

Recent anecdotal experience as discussed in that other thread refutes some of this thinking.

[Caligastia]

I heard somewhere (here?) that it happens when you take over the capital of a large empire. Dont know what conditions are required though...

[The Andy-Man]

i dont think the conquerer has to be smaller then the splitting civ...

[Caesar the Great]

Perhaps it can only occur under certain governments say monarchy or something?

[The Andy-Man]

that seems unlikley.

[finbar]

There was no consensus in the discussion in the other thread - when is there ever consensus? - but the most experienced players seemed to agree that, for a Civ to split, it required:

1. The Civ to be leading the Powergraph. That doesn't necessarily mean it's the biggest Civ in population terms;

2. That Civ to lose its capital;

3. Fewer than 7 Civs alive in the game in order to provide room for the new Civ - created in the split - to enter the game.

[Smash]

Yes thats it.I've posted it before and stand by it until proven different

Keep in mind the PG reflects several factors.You could be #1 in population but not #1 on the PG.Its pop,gold and tech.

[Xin Yu]

maybe only occur in an oedo year. who knows.

[Alex 14]

Originally posted by Caesar the Great
How dare you spit on civ! HANG HIM!! HANG HIM!!

[Ribannah]

Or in a Barb year. But Finbar's summation seems adequate.
In all those years I have only been able to split a tribe once, but then I'm usually very peaceloving.

[Blaupanzer]

The civ being split does NOT have to be number 1 on the graph. Apparently, it does have to have more cities than the one taking its capitol. Otherwise, Finbar's summary seems adequate. Note the emphasis this puts on taking the capitol to break the back of a civ. This fits historical emphasis of conflicts in the 17th through 20th centuries.

[shade]

Did some serious testing this afternoon and here are some conclusions:
taker=country that takes the capital
haver=country that has the capital
1)The haver does not have to be NR1 on the power graph but he has to be placed higher than the taker
NR2 only splits NR1
NR3 splits NR1 and NR2
...
2)The absoluet number of cities of the haver and taker don't seem to be important.
A 45 city empr DOES NOT split the 50 city empr (*)
A 50 city empr splits a 45 city empr
A 6 city empr splits a 50 city empr and a 45 city empr
A 2 city empr splits a 6 city empr, 45 city empr and 50 city empr

my best gues is that it has something to do with the score
45 city empr=> 1497 pt (*)
50 city empr=> 635 pt (*)
6 city empr=> 96 pt
2 city empr=> 36 pt
Although I couldn't verify this using the cheat mode to increase the score of some empr's=> is this caused by the cheat i don't know.

any ideas???

shade

[Edward]

Shade,

Thanks for testing. You seem to be confirming what Smash, fittstim, and Finbar suspected. It's very interesting to note that someone with as few cities as 6 can still be split.

I don't think you and the Mongols had the same power rating. Unfortunately, the powergraph feature draws power lines based on absolute power ratings. It's easy to exceed it's upper limit and get an uninformative horizontal line at the top. It would be a great improvement for the powergraph's top to equal the most power any civ has ever had in the particular game. The lines would then be drawn based on percentages of this upper limit. Then all powerlines would always "fit" onto the graph. (It would, however, lead to some crazy-looking powergraphs in the first few turns.)

It seems the requirements are
1) capitol is taken
2) fewer than 7 civs alive
3) the taker must be ranked LOWER on the POWERGRAPH than the loser

La Fayette's double splitting of the Axis in his third example (in the "Prevent them to escape" thread) implies that the powergraph ratings the respective civs had at the BEGINNING of their turns are the ratings that are compared.

However, La Fayette, has a counter example (example 2). With the following conditions:
Russians: number 1 on the powergraph
Axis (17 cities): number 2 on the powergraph
Turks (15 cities): somewhere beneath them on the powergraph

The Turks did NOT split the Axis upon taking their capitol.

Are the above conditions necessary but insufficient? Is the third requirement wrong? Or do the above requirements only allow for a percentage chance of splitting?

[Smash]

Maybe that occ bloodlust succession game will provide an answer or two.

civ score is worth a little more looking.

[The Andy-Man]

perhaps all the requirements just bake a 9/10 chance of a split.

perhaps, if you tried taking over the capital of a nation that you could probabl;y split, and then repeating the turn over an pver a few times to see if they always split.

[Paul L]

Another requirement for splitting seems to be that the splitting civ must have at least four cities after losing the capital.

Also, the question if a civ transfers his capitol after the split doens't make much sense, because after a split the new as well as the old civ are always getting a new capitol. May be, if the old civ has >1000 money, you get the message of the transfer and the dumb civ just waste it's money

[Marquis de Sodaq]

Originally posted by Edward
It seems the requirements are
1) capitol is taken
2) fewer than 7 civs alive
3) the taker must be ranked LOWER on the POWERGRAPH than the loser

#3 has been disproven. I've seen the overwhelmingly supreme Mongols split the weak Spanish. In another game, the weak English were ripped asunder by the supreme Dutch (me). In the former case, the Mongols had reduced the Spanish from mighty, sacking city after city. Would the powergraph maybe come into play as the civs stand at the beginning of the war? It seems a bit too much for a game, but we've been surprized before. In the latter example, I just happened to be in the neighborhood of London when the English sneak attacked me. My roving horde sacked the city, and friendly Zulus materialized in the outlying english kingdom.

[Blaupanzer]

Well, Marquis de Sodaq puts a real crimp in our theorizing (guessing?!) with the counter-examples regarding rule three. Who can split whom is shot to pieces if the supreme can split a lesser power. I haven't seen a split in quite awhile, and I always go after the capitol first, if I can reach it. I guessed this was because I am normally supreme. Now, I'm not sure what to think. The earlier examples of who split whom seemed very much to depend on the PG rating. Marquis, any theory of your own as to what your experience indicates about this?

[Marquis de Sodaq]

Well, maybe not. Both games were some time ago. I've only seen civil war a few times in years of gaming.

The Mongol example I remember most clearly. On a big europe-to-india map, the Mongols had started in india, the Spanish around the nile, and me on the Danube. Other civs were spread willy-nilly elsewhere. While I grew slowly against the others, the Mongols grew absolutely huge, absorbing a civ or two on their march westward. Much of their later modern era expansion (iran westward?) came at the expense of the Spanish. When they finally sacked Madrid, the Spanish empire was down from about 30 cities to no more than a dozen. I was firmly in second standing. Anyway, the remaining Spaniards split.

When I split the English, it was on a different large map of eurasia. They were from the central plains, but had cities from siberia to the indian ocean. Probably fewer than 20 total - their geography looked like the result of slow growth from an advanced tribe or two. I was marching a group of bad guys to find some other civ, and had no trouble sacking London. All the outlying cities (far north and far south) became Zulus. The core cities around London remained english. I think this was early modern era - dragoons and such.

About the only thing that seems consistent are your first two conditions and some random elements we haven't hit on yet. The english example suggests distance to the capitol might factor in, but I'd expect somebody could provide an example to counter that idea.

Does anybody have anecdotes to test ideas such as:

• size of map
• sacking of capital in Oedo years
• sacking civ color/powergraph standing (such as the beaker discoveries by samson - you never know!)
• civ restart setting and game year