• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Memtest Errors

D

dpk33

Senior member
So lately, I've been getting random BSODs. I've narrowed it down to either the memory or the graphics card. I decided to give Memtest a run. As soon as memtest started running, I went to go out to eat. As I came back, there were over 9000 errors (literally!). Right now, it's still running. There are currently around 200,000 errors, and rising as we speak at a very very fast pace. How do I narrow down which specific stick or sticks of RAM that the errors are caused by, if that is possible? And also, what could cause the memory to be this way? Could overclocking the memory or changing the timings/voltages have caused this?
 
Last edited:
Pull out all but one stick, run memtest. Replace with other stick in same slot, run memtest.

Try one of the sticks in another slot and repeat to see if you get the same result to pin down whether it's the stick or the slot. More reliable if you use a stick that produced no errors in the rare case you have two slots bad at the same time.

Probably gets much much more annoying with more than 2 sticks.

Other folks probably have much better ways of pinpointing the error when you have more than 2 sticks running, as I think that could be a very laborious process. I'm curious to see their process so let's see what they say.
 
Last edited:
Basically test each dimm in each slot? My heatsink is gigantic and covers the RAM slots so I have to unmount it and remount it while doing this... Fun. There's currently over a million errors, if that matters.
 
If a dimm shows no errors in a slot, then you can just test that one dimm in one slot and call it done. You know that dimm and that slot are OK.

Problem is when there are errors, then you can't be sure if it's the dimm or the slot, therefore you have to move the dimm that had errors to another slot to test, preferably to a slot you know is previously had no errors with another dimm and thus is good. If the bad dimm suddenly tests ok in another slot, then the slot that gave errors is defective. If the bad dimm keeps showing errors when moved around, it could still be multiple bad slots (but much less likely, because it I think it's usually the RAM itself that fails in most cases, and the odds for two slots to both be bad is much less than for just one).

Does your heatsink cover all of the dimm slots? I would think you can still get at one or two without having to unmount, right? If so, I'd just unmount the sink, remove all the DIMMs, and use one of the uncovered slots to test with and just call it a day. If you can get at two slots then you might be able to test whether your slots are bad. Even if you can't, you'll still find out that your slots are bad when your (hopefully working) warranty replacement sticks come in 😛
 
Well it turns out that the fan of the heatsink was covering the slots, so I just removed that and let it run passively.

Anyways, I think I've narrowed it down to one stick of ram. The thing is, I don't know at all how to RMA anything. Is it basically just shipping out the RAM sticks to their manufacturer? Do I need the original packaging and such?
 
dpk33 said:
Well it turns out that the fan of the heatsink was covering the slots, so I just removed that and let it run passively.

Anyways, I think I've narrowed it down to one stick of ram. The thing is, I don't know at all how to RMA anything. Is it basically just shipping out the RAM sticks to their manufacturer? Do I need the original packaging and such?
Click to expand...

If you bought them as a matched dual channel set I would expect the manufacturer to replace both dimms, of course this will leave you without any RAM (if you dont have any spare lying around)
 
Puppies04 said:
If you bought them as a matched dual channel set I would expect the manufacturer to replace both dimms, of course this will leave you without any RAM (if you dont have any spare lying around)
Click to expand...

I have 4 sticks of ram (2 sets of 2 sticks), so not having the ones that I'm about to RMA isn't an issue. The thing is, I have never RMAed an item before. Is it as simple as shipping it to the manufacturer and expecting it back within a few weeks?
 
dpk33 said:
I have 4 sticks of ram (2 sets of 2 sticks), so not having the ones that I'm about to RMA isn't an issue. The thing is, I have never RMAed an item before. Is it as simple as shipping it to the manufacturer and expecting it back within a few weeks?
Click to expand...

depends, sometimes its as easy as that. Sometimes its as hard as pulling teeth from a velociraptor.

Depends on who made it, i have had nothing but good experiance with corsair. Had some issues with a Gskill RMA in the past where they would not warranty 2 sticks for me when they would not run at the rated speed and voltage, i had to raise the voltage above there listed specs to get it stable and they said thats not a defect, yeah whatever lol.
 
You normally don't need the original packaging. Just contact the vendor's support line, explain the issue and what you have done - they may have you try a few more things. If they approve the RMA, they will provide all of the instructions for it.

Some vendors will cover shipping to them and provide a label, others you need to pay for shipping to them but they will ship back...so each vendor will be slightly different.

Give yours a call and see how they handle things.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top