This is another of those questions about the combat system...
Anyone know if mismatched defense is computed as:
0 to sqrt( MaxDefense )
or
sqrt( 0 to MaxDefense )
where "0 to X" means a uniformly distributed random variable from 0 to x, inclusive.
To see the difference, let MaxDefense be 9. The first formula--0 to sqrt( MaxDefense )--would predict that the effective defense for a given round would be:
0, 25% of the time
1, 25%
2, 25%
3, 25%
The second formula--sqrt( 0 to MaxDefense )--predicts
0, 10% of the time
1, 30%
2, 50%
3, 10%
I assumed it was always the top formula, but got suspicious when seeing 8 attack fighters do much worse than expected against a battleship with 22 mismatched defense. Now I'm not sure, and there's no easy way of telling.
Stardock?
