On Wed, 2008-08-27 at 13:13 -0400, tedd wrote: > At 12:17 PM -0400 8/27/08, Robert Cummings wrote: > >What does random have to do with > > Oh no! > > Someone mentioned the R word in front of Rob. > > We should put this in a list of things not to do on this list. *heheh* well in the above I'm not so concerned about the issue of random itself, but more about how it doesn't benefit the particular issue being benchmarked. Random (or pretend random depending on your randomly religious views ;) is certainly useful in some benchmarks... for instance testing the speed of a new sorting algorithm, or tree algorithm, etc when a broad data sample is necessary. Cheers, Rob. -- http://www.interjinn.com Application and Templating Framework for PHP -- PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
