randomShuffle

Joseph Rushton Wakeling joseph.wakeling at webdrake.net
Mon Jun 3 14:24:39 PDT 2013


On 06/03/2013 08:28 PM, Diggory wrote:
> I'd guess that the heavy use of floating point arithmetic to calculate the step
> sizes means that algorithm has a fairly large constant overhead even though the
> complexity is smaller.

Yes, I agree.  There might be some optimizations that could be done there.


More information about the Digitalmars-d-learn mailing list