* Thus wrote Ford, Mike:
> To view the terms under which this email is distributed, please go to
> http://disclaimer.leedsmet.ac.uk/email.htm
>
>
>
> > -----Original Message-----
> > From: Jason Wong
> > Sent: 27/12/04 10:16
> >
> > On Monday 27 December 2004 12:40, Richard Lynch wrote:
> >
> > > If you want to mimic the behaviour of abs (allowing for positive
> > numbers)
> > > and performance was an issue, that:
> > > $x = ($x < 0) ? - $x : $x;
> > >
> > > is most likely faster than abs()
> >
> > Having nothing better to do I decided to benchmark this:
> >
> > ternary:
> >
> > $doo = -20;
> > for ($i = 1; $i < 10000000; $i++) {
> > $dah = ($doo < 0) ? - $doo : $doo;
> > }
> >
> > abs():
> >
> > $doo = -20;
> > for ($i = 1; $i < 10000000; $i++) {
> > $dah = abs($doo);
> > }
>
> That's not a valid benchmark, since only on the first pass through the loop
> is $doo negative. Personally, I'd want to test it with equal numbers of
> positive and negative values, and I'd want to know the contribution of the
> loop and value-setup overhead, so I'd write it like this:
This really was my first concern about the benchmark (unfair
negative usage). In what ever case one uses to get the absolute
value from a value be sure to document things like:
/* Get the absolute value */
$dah = ($doo < 0) ? - $doo : $doo;
where:
$dah = abs($doo);
is self relevant.
Curt
--
Quoth the Raven, "Nevermore."
--
PHP General Mailing List (http://www.php.net/)
To unsubscribe, visit: http://www.php.net/unsub.php
