I've also had memtest86 give false positives, so an error doesn't necessarily indicate any problem at all, but could be an incompatibility with your chipset...or perhaps some other sort of hardware fault. I still view it as a valuable tool, however. I just try to take the results with a grain of salt when running on chipsets that aren't mainstream or aren't officially supported.
Greg > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:hardware- > [EMAIL PROTECTED] On Behalf Of Thane Sherrington > Sent: Friday, October 31, 2008 3:42 AM > To: [email protected] > Subject: Re: [H] Memtest86 v2 > > At 05:59 PM 30/10/2008, maccrawj wrote: > >IIRC there are two flavors of MemTest86, one with a Plus (+) at the > >end that was newer? > > > >http://www.memtest.org/ > > > >When I was testing issues with VPU recover on my new system as was > >being told to blame my RAM I tried both and think MemTest86+ was the > >one that ran correctly for me. > > None of the software RAM testers are very good. I've had several > situations recently where I ran 50-70 passes with no errors, but the > DIMM turned out to be bad anyway. So if it gives you an error, then > you have bad RAm, but if it doesn't, that doesn't mean the RAM is fine. > > T >
