On Tue, May 3, 2011 at 8:32 PM, Dave Angel <da...@ieee.org> wrote: > What I'm surprised at is that nobody has pointed out that the logn version > is also generally more accurate, given traditional floats. Usually getting > the answer accurate (given the constraints of finite precision > intermediates) is more important than performance, but in this case they go > hand in hand.
And that, Your Honour, is why I prefer bignums (especially for integers) to floating point. Precision rather than performance. Chris Angelico -- http://mail.python.org/mailman/listinfo/python-list
