|
Boost : |
From: David Abrahams (dave_at_[hidden])
Date: 2004-03-03 12:12:43
Joel Young <jdy_at_[hidden]> writes:
> --------
> From: Brian McNamara <lorgon_at_[hidden]>
>> Like any quicksort, it's O(N^2), but the constant-factor costs here are
>
> nlog(n) ?
Average case: nlog(n)
Worst case: n^2
http://www.cs.virginia.edu/~luebke/cs332.fall00/lecture6/tsld009.htm
introsort is reliably nlog(n)
-- Dave Abrahams Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk