Memtest stable but not windows stable…

I have been using the 1.51 version of memtest to test memory stability. I ran a full scan (all tests) for 14 hours and looped test 5 then test 8 for a few more hours each. I got 0 errors. But when I boot to windows I get SuperPI errors usually on the first loop. I tried using a low multi which gave me an underclock of my CPU so I would know if it’s a RAM problem and I still get SuperPI errors. Anyone have any idea of what may be the problem?

this is something that has gone on for some time now..... basically it seems that windows (or whatever os you're using) puts additional stress on the memory that causes instability beyond memtest. I'm sure someone else will post greater info on the subject.....

Bah, whats the good of memtest then

I have the same situation here...

I'm fighting to get my GSkill LA and LE stable at 2.5-3-3-7 1T at 275MHz. I can run Memtest86+ 1.51 for how many hours I want it but memory fails very quickly at Prime95 or SP2004.

Not sure what problem is but I saw many others in with the same problem.