I followed that link and read that paper, and my God, it is AWESOME! All that the guy had in front of him was 5 obscure lines of code and a befuddling 'magic' constant, and he sat on it, worked it out, and using simple calculus and error analysis that anyone can understand, got back the constant that was used, and whats more, he suggested a better value for the constant with proof. Do read the paper !!!!
Whoa, what a morning it has been !