Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!watdcsu!kishore2
From: kishore2@watdcsu.UUCP
Newsgroups: comp.lang.pascal
Subject: Re: Perfect number program
Message-ID: <3936@watdcsu.waterloo.edu>
Date: Thu, 8-Oct-87 16:33:16 EDT
Article-I.D.: watdcsu.3936
Posted: Thu Oct  8 16:33:16 1987
Date-Received: Sun, 11-Oct-87 05:36:28 EDT
References: <9662@brl-adm.ARPA>
Reply-To: kishore2@watdcsu.waterloo.edu (Sherman Lang - Systems Design)
Organization: U. of Waterloo, Ontario
Lines: 15

In article <9662@brl-adm.ARPA> silva@tcgould.TN.CORNELL.EDU (Victor M. Silva) writes:
>You can reduce your processing time in half by just having your inner
>loop (FOR K=2 TO I...) only go to I/2 since anything larger cannot be
>a divisor (ie. 2*(I+1) > I ).
>                                                 -VMS-


And you can reduce it further by going to trunc(sqrt(i)) instead.  
The other thing to do is to keep track of factors and to go backwards
in your testing of k.
-- 
 ============================================================================= 
|  Sherman Lang, Systems Design Engineering, University of Waterloo.          |
|					"A screaming comes across the sky..." |
 =============================================================================