python interpreter (evaluate)              3820 ns per element   (measured on 20,000 elements, scaled to 200,000)
compiled C, one ctypes call each            584 ns per element   (200,000 elements)
compiled C, loop inside C                  0.96 ns per element   (20,000,000 elements)
ratio interpreter / C loop: 3,966x   ctypes call / C loop: 606x
