Question about memtest for windows

[Kaxx]

Does using this program "guarantee" my CPU isn't causing my errors? What I'm trying to say is that I run a couple of instances of SP2004 (one for each core) and got an error almost an hour into is. So I rebooted into windows and run memtest for windows and got an error about the same amount of time into it. Does this mean it is for sure a RAM error and readjusting my RAM timings correctly will allow it to run flawlessly, or could it still possibly be a CPU error?

[Praz]

It's still possible the error could be processor related. The only way to be relatively certain the error is not processor induced is to use a lower multiplier. Following the Overclocking Guide, it would already be known if the processor is stable at the speed the error is being produced.

[burntcircuit]

memtest for windows is for testing the memory, right! So if you have errors in memtest, its a pretty good bet you have some work to do on your memory timings. Any memory errors will also show up as faults in SP2004, but until you fix the source of the memory errors, you cannot know if the CPU is causing SP2004 to fail.

[Kaxx]

I'm kinda in a catch 2-2 right now.

 

My HTT is 322,

CPU mult is 9x and 1.45 vcore,

RAM is 133 divider and 2.6 vdimm, 1T-3-4-4-8-7-13-2-2-2-3 (perhaps I should try 2T if I get another error???).

 

 

From what I've read on here Opty 165's don't like low dividers but mine has done fine so far with a 133 divider up to 320 HTT, 9 hours SP2004 stable. So now I'm trying to find out whether it's my CPU or my RAM that is the culprit. I just rebooted with a 140 divider on my RAM and memtest error came up much quicker, like around 10 minutes into it. It could be that 140 divider doesen't agree with either my CPU or my RAM. Could be my CPU doesen;t wanna get any higher than 2.88 GHz. Right now I'm back at 133 divider and 1.475 vcore, gonna try that with SP2004 and see what I get. Then I'll try a lower multiplier like you suggested Praz.