I haven't actually tried messing around with these, but I'm fairly certain that the following seven constants defined in GalCiv3GlobalDefs control how the ideology tree costs scale.
- IdeologyTraitBaseCost
- CulturalAlignmentBiasSamePyramidGrowthExponent
- CulturalAlignmentBiasSamePyramidGrowthExponentialMult
- CulturalAlignmentBiasSamePyramidGrowthMultiplier
- CulturalAlignmentBiasOtherPyramidGrowthExponent
- CulturalAlignmentBiasOtherPyramidGrowthExponentialMult
- CulturalAlignmentBiasOtherPyramidGrowthMultiplier
If the IdeologyTraitBaseCost is B, the other six constants are c, d, e, f, g, and k respectively, the number of ideology perks acquired in one tree is n and the number of ideology perks acquired in the other two trees is m, then the cost y(n, m) of getting the nth trait in the one tree should be
y(n, m) = d * n^c + e * B * n + g * m^f + k * B * m
if I'm reading the comments in the XML file correctly and haven't introduced an error in transcribing the equation. Note that d and g, the two *ExponentialMult constants, are in the unmodded game defined to be zero, and so you do not see the exponential terms affecting the in-game costs.
