On Tuesday 17 August 2004 19:43, Jason Clements wrote: > I'm running Prime95 on a 600 MHz PIII tower PC (double-checking). > > When the machine is otherwise idle, the 'per iteration time' is fairly > consistent on any one day, but can vary quite a bit from week to week. > > Recently, it's been around .47 sec, but at times it's down to .39 or so. > > I've read here that laptops can be slowed by getting hot, but I can't > see that this would be the problem here - there's plenty of air space in > the box.
A lot of HOT air won't cool the CPU any better than a bit of hot air. Dunno about where you are but round here (Northern Ireland) it's been unusually warm for the last two or three weeks, especially overnight. Check the case fan(s). And the CPU fan; some PIIIs seem to work (more or less) with a failed CPU fan, though they show signs of stress as soon as steady load is applied. Or, switch the system off, let it cool down for a couple of hours, switch on & start Prime95 straight away. If you get a good iteration time which tails off over a few minutes to an hour or so, it's definitely a heat-related problem. Regards Brian Beesley _______________________________________________ Prime mailing list [EMAIL PROTECTED] http://hogranch.com/mailman/listinfo/prime
