Performance improvements.


R. Clayton (rclayton@monmouth.edu)
(no date)


  When you say keep pace with your fast solution, does it have to be faster
  than your solution?

The criterion hasn't changed: your solution should be be no more than twice as
slow as my solution, whichever one you decide to work with.

  I get results which are typically 1.5 times your faster solution and this
  ratio is same or less (1.3--1.5 times) for range >10 points. Does it mean
  that I still have to radically improve the timing?

If you can radically improve your timing, you should do it (and by "radically"
I mean "asymptotically"). I have in mind an O(n^2) average-case algorithm, but
I haven't looked into it yet.

Significant improvements in performance come from lowering an algorithm's
asymptotic bounds. Diddling around with implementation details is important,
but doesn't provide the payoff that improving the asymptotic behavior does.



This archive was generated by hypermail 2.0b3 on Fri May 10 2002 - 12:45:04 EDT