Christian Smith wrote:

> I've had times ranging from 25 to 164 seconds on my single IPX. Is the
> example flawed, given that I've had such ranging results on the same
> machine?

I suspect that this is making use of random data to generate a key and
thus isn't a very useful benchmark at all:

Here's a range of runs on a swarm (mips board) sorted by CPU time:

	0.7u 0.0s 0:00.75 96.0% 0+0k 0+0io 0pf+0w
	0.8u 0.0s 0:00.89 97.7% 0+0k 0+0io 0pf+0w
	1.0u 0.0s 0:01.10 90.9% 0+0k 0+0io 0pf+0w
	1.8u 0.0s 0:01.85 98.3% 0+0k 0+0io 0pf+0w
	2.8u 0.0s 0:02.82 99.6% 0+0k 0+0io 0pf+0w
	5.2u 0.0s 0:05.31 99.0% 0+0k 0+0io 0pf+0w

That's quite some variation :-)

Simon.
--
Simon Burge                            <simonb@wasabisystems.com>
NetBSD Support and Service:         http://www.wasabisystems.com/