When do Barbarian units attack?

[OTAKUjbski]

Barbarian animals will not enter cultural borders and have a %-to-attack chance based on iAnimalAttackProb from CIV4HandicapInfo.XML.

Barbarian units first appear in the game after a certain number of turns based on iBarbarianCreationTurnsElapsed from CIV4HandicapInfo.XML.

But when will Barbarian units attack?

I've seen Barbarian units come and go right past my units & cities without even blinking, so there's got to be some element I'm missing.

Anybody know?

 

[dante alighieri]

I've seen Barbarian units come and go right past my units & cities without even blinking, so there's got to be some element I'm missing.

I'm not sure of how its calculated, but the barbs won't attack a unless they have a certain chance of winning, I believe. In my experiencethough, the barbs will ignore the AI units if they can attack me instead. I've seen barb warriors target my archer instead of an AI scout.

 

[Infantry#14]

also, barb units are quite relucatant going into cultural borders in the 3000BC era, I have seen them just exploring while they could have the oppounity to pillage area around my early city

 

[Aka]

Based on all my experience, barbarians attack rather on a whim.

 

[omnimutant]

The great wall with Raging barbarians is your friend.

 

[jason77024]

Answer appears to be as follows: Assuming 'normal' barbs (not raging), no city attacks until the number of cities has doubled. I believe that's based on the original # of civs, but am not certain.

http://forums.civfanatics.com/showpost.php?p=3983344&postcount=33

 

[IronCrown]

In my games, Barb stacks spawned by random events always ignore me, march all the way through my land and go directly to a neighbor I like it though

 

[OTAKUjbski]

Answer appears to be as follows: Assuming 'normal' barbs (not raging), no city attacks until the number of cities has doubled. I believe that's based on the original # of civs, but am not certain.

http://forums.civfanatics.com/showpost.php?p=3983344&postcount=33

Hmm ... I read something similar to that somewhere (might even have been that post), so that must either be true or very close to it.

Thanks.

 

[Aquatic]

As a rule of thumb, I have noticed that, at least on Noble, barbs never attack before the year 1000 BC. After that they seem to get the idea and things can become quite lively for a while!

 

[KMadCandy]

Answer appears to be as follows: Assuming 'normal' barbs (not raging), no city attacks until the number of cities on that continent = number of civs + 1 has doubled. I believe that's based on the original # of civs, but am not certain.

mostly true but a key part is missing IIRC. i may be wrong about "number of civs on continent +1", it might actually be "number of cities on continent * 2". same difference for an isolated start. anyway, the trigger for entering borders isn't time or dependent on total civs on the map at all. the specifics i put in bold above make isolated starts potentially painful, since your second city = barbs can now have a party inside your borders. see the ALC with Mehmed as leader for an example of this. the barb troubles started on page 8, post 146. i like the great wall for isolated starts, frees me up from worrying about troops/powergraph until contact with other cities. the downside is that great spies aren't generally the first you want to generate in an isolated start where you're not earning EPs at all. a settled GSpy is pretty useful tho, so i do try to get GWall if i'm isolated.

 

[jason77024]

mostly true but a key part is missing IIRC. i may be wrong about "number of civs on continent +1", it might actually be "number of cities on continent * 2". same difference for an isolated start. anyway, the trigger for entering borders isn't time or dependent on total civs on the map at all.

But the trigger number should be based on the -original- number of civs, no? Start with 7 civs, then barbs attack once there are 14 cities... regardless of how many civs exist at that point.

Otherwise, barbs could attack cities on different continents starting at different dates. Could be true, but that didn't sound like it squared with the SDK-based info to which I linked above.

the specifics i put in bold above make isolated starts potentially painful, since your second city = barbs can now have a party inside your borders.

Can you put some specifics in bold please?

see the ALC with Mehmed as leader for an example of this. the barb troubles started on page 8, post 146.

Hmm. That second city is founded in 1900BC, by which time every other civ (certainly in Emperor) has at least one extra city. So that's kinda inconclusive, esp. given Sisiutil doesn't remember precisely either.

I will say that I do lots of early rushes, often before any AI sends out a settler. So after such a rush, the number of cities > number of (current) civs. But I haven't seen any barb rushes following closely behind. That's some strictly anecdotal info, fwiw...

Either way, we'll get this figured out. The test would be whether anyone with an isolated OCC start gets attacked. (Now, to find some masochist who'd play that!)