A handy tip to reverse engineer complexity from empirical examples is plot on a log-log plot and measure the gradient (if its a straight line). The gradient is

A handy tip to reverse engineer complexity from empirical examples is plot on a log-log plot and measure the gradient (if its a straight line). The gradient is the exponent, ie. O(n^gradient). Great if your algorithm is so complex you can't analyze it. http://www.dgp.toronto.edu/people/JamesStewart/378notes/05logLog/ http://www.dgp.toronto.edu/people/JamesStewart/378notes/05lo...