Re: Clearsgts

[cld@xxxxxxxx (Carl Distefano)]

Reply to note from "Robert Holmgren"  Wed, 1
Dec 2004 07:00:49 -0500

Robert:

> Just out of curiosity, how long does 100 iterations of this
> take on the same machine?

It's almost under the radar: 0.05 seconds (vs. 0.28 secs. for the
CountUp method). Big difference, eh? A factor of five. We've seen
this before: verbose but straight-forward XPL running faster --
sometimes much faster -- than more compact but "fancier" code. No
surprise there; "fancy" generally requires more interpretation (more
processor cycles), and that equals more time.

But, to put this in perspective, ONE iteration of the CountUp method
to null out 44 S/Gs takes about 3 milliseconds, 6 millseconds to
uninitialize them -- acceptable even by New York standards, I'd say.

--
Carl Distefano
cld@xxxxxxxx