I found this sufficiently uninformative that I went back to http://www.willamette.edu/~fruehr/haskell/evolution.html http://www.willamette.edu/~fruehr/haskell/e

I found this sufficiently uninformative that I went back to http://www.willamette.edu/~fruehr/haskell/evolution.html http://www.willamette.edu/~fruehr/haskell/evolution.html. Incidentally, if your language has big integer arithmetic (like Python does), the following is substantially faster than the usual iterative or recursive approaches: def multiply_range(n, m): if m < n: return 1 elif n == m: return n else: mid = (n+m) // 2 return multiply_range(n, mid) * multiply_range(mid+1, m) def factorial (n): return multiply_range(1, n) Try factorial(100000) to see the performance difference. Why this makes a difference is left as an exercise for the reader. (Note, in some languages, like Ruby, converting to a string representation may be the biggest bottleneck.)