new ram - memtest single error

rebelit

rebelit

rebelit

Associate
Joined
12 Feb 2011
Posts
13
Hi all,
I built my latest sytem 3 months ago..

crosshair IV
4gb corsair XMS3 (2x2gb) (slots 1 & 3)

Not overclocked (default everything)

As I make extensive use of VM's on a daily basis I always intended to take the RAM to 8gb in the future.

I bought another 2x2gb set of XMS3 and fitted it this morning (slots 2 & 4). I booted into xp64 and as soon as the desktop appeared I got a BSOD. I went into BIOS and checked I wasn't oc'ing anything. Everything is set to auto. (not sure if the MB attempts any auto OCing though?)

I rebooted again and got a BSOD after about 3 minutes at the desktop

So I booted to memtest+ and ran the default test. 20 minutes through the 1st pass on test 7 I get one error message.

I am still running through a 2nd pass at the moment.

Should I RTM these new modules immediately or is there anything else I need to test / check first?

Also if I do need to RTM do I return them to overclockers or to corsair?

Thanks,

J
 
not yet. I don't much like messing with the bios settings too much. But I'll give it a go after this 2nd pass completes.

I know very little about this sort of stuff but I am assuming if the same single address fails on the 2nd pass that it's probably not a voltage error? ie I would expect multiple errors at multiple addresses if the voltage was too low? Just an uneducated assumption though!
 
If it's failing at the same address each and every time then the RAM may indeed be faulty in which case try it in another slot and see if the problem follows the suspect module. You would RMA through OcUK by the way.
 
Update:

So on the 2nd pass of Memtest, no errors were found. So I tried manually setting the dram voltage (it was at ~1.52 on auto, I raised it to 1.60)

Booted into windows. Still got BSOD after short time.

So I removed the new sticks & rebooted - stable for 15 minutes.

Refitted the new memory, taking care to ensure they were seated properly. (This automatically resets the voltage to auto - I checked)

No BSOD for 10 minutes so far. So maybe one of the sticks just wasn't seated correctly?

The machine is always on so I guess it's just a matter of time now to see if it is stable.
 
rebelit said:
Update:

So on the 2nd pass of Memtest, no errors were found. So I tried manually setting the dram voltage (it was at ~1.52 on auto, I raised it to 1.60)

Booted into windows. Still got BSOD after short time.

So I removed the new sticks & rebooted - stable for 15 minutes.

Refitted the new memory, taking care to ensure they were seated properly. (This automatically resets the voltage to auto - I checked)

No BSOD for 10 minutes so far. So maybe one of the sticks just wasn't seated correctly?

The machine is always on so I guess it's just a matter of time now to see if it is stable.
Click to expand...


rebelit said:
Alas after running happily for a while I started loading it up - VMWare server, apache, mysql, google earth etc... then BSOD :(
Click to expand...

Run a 24/48 hour test, see how your RAM performs, if 100% maybe something else :rolleyes:
 
rebelit said:
So on the 2nd pass of Memtest, no errors were found. So I tried manually setting the dram voltage (it was at ~1.52 on auto, I raised it to 1.60
Click to expand...

With the corsair xms3 you can go up to 1.65V - i would definitely try running them at this as this is their tested voltage. It could be that they’re just not getting enough volts...

If the above doesn't work, as your running 4 sticks, try increasing your NB voltage by an increment or two – as the NB sometimes requires more juice, when using 4 sticks, to keep things stable.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom