I recently played in a resources game for the first time. I didn't learn much, since although I saw many "Drill built" messages, the Drills were apparently both invisible and intangible, as I never sighted or tripped on any. So I decided to learn more by reading on the forum about resource generation, and I noticed a minor mathematical issue.
In Viech's thread explaining the resource system, it was stated that
(1) the "ideal" is to have the total amount of resource generation be proportional the volume of the union of all the spheres.
(2) the pairwise approximation always results in a total rate of generation that is less than or equal to the ideal rate.
Statement (2) seems to be incorrect...
Let q[SUB]xy[/SUB] for two RGS x and y be (the area of the intersection of x and y) / (the area of a sphere).
The rate for an RGS i is calculated as some base rate times the product of (1 - q[SUB]ij[/SUB]/2) for all j != i.
Now suppose that three RGS a, b, and c are arranged in a line such that q[SUB]ab[/SUB] = q[SUB]bc[/SUB] = 1/10; q[SUB]ac[/SUB] = 0.
So for the interfering spheres the factor (1-q/2) is 19/20.
Suppose the base rate is 1. Then the approximation yields 2 * 19/20 + (19/20)2 = 1121/400.
The ideal rate is 3 - 2 * 1/10 = 28/10 = 1120/400. This means that the builders get 1/400 in absolutely undeserved resource generation!
The formula underestimates the effect of interference since the two areas of intersection in sphere b are non-overlapping.
No doubt the calculation can work fine as is, but nitpicking of dubious benefit is what the open source community lives for, isn't it?