All of the numbers I've reported are with a left argument to timex. The argument I used was 10 for the 500s, and 1000 for the 50s, except my phone where I had to use 10000 to get stable answers for the 50s.
Paul On Thu, Sep 13, 2012 at 12:09 AM, Roger Hui <rogerhui.can...@gmail.com>wrote: > 1. The %. implementation does not take different paths that are dependent > on the values in a non-singular matrix. (Part of what makes it > algorithmically interesting :-). Therefore the time required should be the > same for different random matrices. Of course, unless you have ripped out > most of the stuff from your machine, that time would be impacted by e-mail > arriving, your moving the mouse, the browser doing whatever, your > anti-virus acting paranoid, whatever, whatever, ... > > On Wed, Sep 12, 2012 at 11:09 PM, Joey K Tuttle <j...@qued.com> wrote: > > > I agree with your point, but the "benchmark" has always included > > generating the matrix and that is typically a very small part of the time > > and should be relatively stable (although I suppose inverting the same > > "random" matrix over and over would remove some variation). Your > suggestion > > of using a left argument for 6!:2 is the best way to reduce (or at lease > > smooth out) variability. ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
