AI sneak-attack probability

Belisar

Defender of Byzanz
Joined
Apr 24, 2003
Messages
1,244
I win 95% of my games the military way, but since I lack the time to play at huge or large maps, most of my games end with cavalry (Emperor) or tanks (Deity).
Because I wanted to try some modern age war-strats, I tried a variant in a recent game: Don 't declare war until the modern age.

I started near the Aztecs and had no problems with them. They were "gracious" towards me and my mil. advisor kept telling me "Compared to them we have a strong mil."
Suddenly, they declared war (no extortion, intact reputation...) with a few jags enetring my territory.
I conquered their lands, GL-rushed my FP their and tried to stay within my variant, no war declaration against any civ until MA.

But in the middle ages, the "polite" Mao suddenly declared war
(My army was "average" compared to him), again without any provocations (close cities or something)
Now it hit me: The Actecs declaring in BC, Mao in the middle ages, are AI civs more likely to declare when they have their UU available? Do they indeed "go" for their GA? Even if their armies are equal or inferior to the human player?
If so, one could 'predict' periods where attacks will have a high possibility and other periods where the player would be relatively secure.

I haven't observed this in vanilla Civ3 and I have played mostly heavy war games with PTW, so it might be something already discovered by others or it could be a coincidence.
Has anybody else observed this, probably in recent patch-versions?

Belisar
 
yes, AIs want their GAs - and the UU is one sure way to get it. Beware of the Spanish, btw, they use that conwuistador for sudden, deep strikes - especially deadly when they ROP rape you!
 
I have noticed this too. The Romans sneak attacked me with a warrior, legionarry, and spearman. Even though I had a warrior, spearman, and about 10 horseman in that city...on a hill.
 
Back
Top Bottom