Author: Robert Hyatt
Date: 17:58:32 01/14/03
Go up one level in this thread
On January 14, 2003 at 12:56:33, Brian Kostick wrote: > >>On January 13, 2003 at 17:57:44, Jorge Pichard wrote: >> >>http://www.pcworld.com/howto/article/0,aid,106700,pg,2,00.asp >> >> > >>On January 14, 2003 at 00:45:56, Robert Hyatt wrote: >> >>There they be dragons... >> >>if you finagle, you had _better_ test. And by testing I mean test for hours >>with a program that fries memory with sliding 1 bit patterns, sliding 2 bit >>patterns, etc. > > >I thought they were going to overclock ram but I did not see it. Dr. Hyatt, is >there a particular part of the article your comments refer too? > >Thanks, BK No. I originally could not access the web site and assumed that they were going to tweak bios settings to either ramp up the bus speed, remove wait cycles, or something similar. That is the way of the overclockers. The problem is, once you overclock, it can be very difficult to detect the rare cases where it breaks something. But most of the time, there will be that one instruction that can't quite get everything to settle before the clock falls, and it blows up. But so rarely that it doesn't seem to be there. I'm _very_ wary of overclocking. Those that like it can do it all they want. But I have enough hardware background to realize just how risky it really is. Sometimes a single instruction will fail. Sometimes just a particular bit pattern for a particular instruction will fail. Way too unpredictable for me to waste the time...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.