Memory testing - How many passes is enough?

[wildbill001]

Well, smallest value it would set was .87500 and every attempt to boot was met with "Over voltage failed. Run setup"

Is that pointing the finger at Asus and the mb? I am running bios 2104 which is the latest.

[wildbill001]

Gskill Tech,

Running one module at a time for multiple passes and 4-9 hours did not produce any errors. It is only when I plug in all 4 that I get this intermittent erroring. Earlier today I ran the box for hours with no issue. Rebooted to check values and BAM! errors all over the memory.

sigh. My baby is really sick I'm afraid.

[Tradesman]

I'm wondering if it might be the advanced timings, if you could check and list what they are loading, you show them above as AUTO, but if GSkill Tech can see what is actually there, that could be the problem (sort of guessing the tRFC and maybe tRC,,,most AMD mobos are really programmed BIOS wise well for 4 sticks...so, I know it's a pain but if you can go in and see what the AUTO is actually setting it might help, if hitting the AUTO doesn't show the actual setting, you ought to be able to type in a number and it will show that and to the side should show what the auto setting is.....they might be like

tWR 16
tRFC 180
tRRD 6
tWTR 7
tRTP 7

Somewhere in those areas, if they don't show what you type and another setting (the AUTO setting) then leave them on AUTO and I'll drop GSkill a note to get back to your thread Fri

[Britton30]

You might try HT Over Voltage at 1.2V

[wildbill001]

So I decided to not let the system run overnight last night. I had shut it down with mem overvoltage at 1.53. couple of hours later I couldn't stand it--I had to boot it up to see what would happen. Sigh, memory errors all over the range.

Ok, we know it won't work at 1.515. We know it sort-of worked at 1.53 but not after a reboot, so let's bump up to 1.545. Runs just fine for several hours. Shut it down for the night.

Boot up this morning hoping that I had finally zero'd in on the problem. @#$!%$%! nope, won't even POST. Power-up/down several times and finally it POSTs and loads memtest (memory overvolt=1.545). Double @#$%@#$! Errors all over.

Drop memory overvolt back to 1.53 and all seems right with the world-for now. sigh.

Checked the other settings to see if they would show what [AUTO] meant but that was a no-go. All it does is give me a list to choose from.

I'm gonna try running my apps this weekend and see what happens. If all is well, fine. If not, I think I'm gonna try RMAing the mb AND the memory.

Bill W

[GSKILL TECH]

No problem, try a new kit to see if it performs differently.

This memory should be plug and play no problem with the latest BIOS.

Thank you
GSKILL TECH

[wildbill001]

So I have hesitated posting these last couple of days for fear of jinxing myself. But it has been almost 5 days now so I think I may be safe. We'll see.

So last time, I said I was going to load up my ESXi and run my server as if nothing had happened. Work it hard until it broke and then start the RMA process. So I set my voltage to 1.53, which seemed to give me the best results. Loaded ESXi 5.0, then brought up a single virtual machine running RH 5.4. 4 virtual CPUs, 8g RAM. It came up fiine and run for a full day with no problems.

Hmmm, ok, brought up some other virtual machines running Solaris x86, Win2k8 (all 64 bit, BTW). No problems. Got all the virtual machines working hard interacting with one another, testing software and applications. No Problem.

Been almost 5 days! No Problems! Thank you Tradesman and GskilTech!!!!

Bill W