• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

1 error = bad memory?

M

moussaka

Limp Gawd
Joined
Feb 7, 2006
Messages
253


ever since ive put this computer together about a year ago ive got blue screens about once every few months. it always used to happen when i played WoW, but since i quit that it has happened on the desktop as well. i finally decided to do a memtest last night and it ran for about 11 hours. as soon as it started it found that one error and then nothing else. is this reason for a RMA or should i just live with it?
 
i just rma'd my ram from g skill 2x2gb kit because one dimm had 1 error on it
i think it was test #5 when i had both sticks in i would get random BSOD
but i could run the one stick with no errors fine
and that stick ran memtest with no errors completing all the tests
so yea its bad memory !
unless you like random bsods and a unstable computer
 
well its not like it does it every single day. its maybe once a month, if that. its still frustrating when it happens though! :p
 
mine was bsod's like once a day maybe more sometimes

iam running vista x64bit
 
Without knowing any specs, I'd say pump up your DRAM voltage just a single increment and retest.
 
I'd replace the memory because even one bad bit is one too many, especially when the testing is done at far below the 95 Celcius maximum rating for RAM chips.

You may also want to test with Gold Memory ver. 5.07 and MemTest86 ver. 3.3 or 3.4.
 
i was snooping around in BIOS today and noticed that the votage was set to 1.8 when teh ram calls for 1.9... i swear i changed the voltage when i first built this thing. im going to test again tonight to see if it makes any difference.
 
I wouldn't worry about it. You did 24 passes and only got one error, which means that it's only happening like 4% of the time with that byte. It might bother some people, but I've run computers for months with similar situations and never had a problem. But it might just be me. And like someone said, bumping the voltage couldn't hurt.
 
Desdichado said:
I wouldn't worry about it. You did 24 passes and only got one error, which means that it's only happening like 4% of the time with that byte. It might bother some people, but I've run computers for months with similar situations and never had a problem. But it might just be me. And like someone said, bumping the voltage couldn't hurt.
Click to expand...

Very true, I have run memory with some errors for months before with out major problems (Ballstix). But it's just an unnecessary thing to live with when you have a warranty. I would bump the volts up one notch and if you still get the error, I would send them back. I've had ram that has one error eventually have ten or more errors down the road. I would just get it taken care of sooner rather then later if you can. MOA.
 
Desdichado said:
I wouldn't worry about it. You did 24 passes and only got one error, which means that it's only happening like 4% of the time with that byte. It might bother some people, but I've run computers for months with similar situations and never had a problem.
Click to expand...
That's eerily like what No Condom Dave said back in the 1980s.
 
in all seriousness update your bios. i tested my ram a while ago and it spewed out a disgusting amount errors. after pulling and testing the sticks and then testing some other ram i came to the conclusion that it was not the ram - it was the mobo. a bios flash took care of that one right quick;)
 
anyways, after changing the voltage from 1.8 to 1.9 (what it is supposed to be) this is what occurred last night!



hows corsair's RMA process lol
 
Mike Clements the Corsair rep is a top notch guy. I figure you'll have no trouble with an RMA. You ought to send him a PM, if he doesn't pop in the thread here.
 
moussaka said:
anyways, after changing the voltage from 1.8 to 1.9 (what it is supposed to be) this is what occurred last night!



hows corsair's RMA process lol
Click to expand...

Error's on test 5, can usually be fixed by a small bump in ram volatage. But error's on test 6 and or 7, usuall always means bad ram as long as the ram speed and timings are with in spec.

I would RMA both sticks of Ram as you are now getting errors on both test 6 & 7.
 
Way back when, who knows when, I was have some random errors and blue screens on an older comp, so I decided to run memtest86 for like 12 hours.

I came back and it had only listed 3 errors.

But guess what, I threw new memory in and all the problems went away...

So, yes, I believe one error in memtest could indicate bad memory and mean u should replace them.
 
I can't believe there are actually people in this thread saying "don't worry about it." There should not be any errors in your ram, period. You should not get BSOD's for hardware failure, ever. You can't feel comfortable doing important work on a system with known errors in the memory, even if it's just a few. RMA them and get some that are good.

EDIT: this is assuming, of course, that you are running your ram within their rated speeds. If you're overclocking them and getting errors then you need to test them at complete stock speeds.
 
I have some Crucial Ballistic DDR2 800 (running at 667 4-4-4-12 at the moment, the mobo chipset is 667 limited) and yesterday for no reason that seemed logical I started having the weirdest BSODs I've ever seen. Errors related to SYSTEM_SERVICE and others I simply have never seen before in decades of working with Windows. I decided to do a memtest86+ run off the bootable CD and wham, almost instantly I started having one RAM error after another, starting as low as 60MB (I have 2x1GB sticks). Boy my heart fell through the floor, really, thinking I just put all this together in a low cost machine about 1.5 months ago and was worried it was about to go belly up.

So I pulled 1 stick and re-ran memtest86+, no errors. Tried the same stick in the second slot, no errors. Swapped out for the other stick, first slot, no errors, then the second slot, no errors. Put both sticks back in, no errors.

It's just one of those things, and ever since I reseated the RAM I haven't had any issues at all, running stable once more.

Try reseating the RAM, just for shits and giggles. I know it seems highly illogical that the RAM would mysteriously unseat itself or lose contact (I rarely if ever move the desktop machine, it's on the floor underneath the case, nothing moves it and it doesn't get kicked of even vibrate much at all), and I can't explain why it happens. As I said reseating the RAM cleanly (till I hear the "click" of the DIMM when it seats fully) seems to have given this RAM a kick in the pants, so it can't hurt.

Pull the RAM, retest each stick individually and in each slot, if they both come up completely clean, reseat both sticks or whatever sticks you have back in the slots until you hear that faint "click" when they're fully reseated, and retest again. Chances are you'll find there's no problem at all, I bet. But then again... ;)

One last tip:
Stop overclocking the RAM when you're testing it. It's obvious by the first shot above it's running above spec (standard RAM speeds, that is). Always go back to stock settings (SPD) when testing RAM as that's the only surefire way to get reliable results.
 
To the OP, can you list your complete hardware specs? Also, list ALL bios settings you have made for RAM, CPU, and voltages.
 
- Abit IB9 QuadGT
- Intel E6600
- TWIN2X2048-5400c4 CORSAIR XMS2
- 8800gts (640mb)
- corsair 620hx PSU

cpu options:


ram timings:


uGURU options:


voltage:



is that everything?
 
After testing each stick individually one was good and one was bad. I will start the RMA when I find a temporary, backup set of ram so my computer isn't down for however long it takes.





Thanks for your help.
 
Joe Average said:
try reseating the RAM, just for shits and giggles. I know it seems highly illogical that the RAM would mysteriously unseat itself or lose contact (I rarely if ever move the desktop machine, it's on the floor underneath the case, nothing moves it and it doesn't get kicked of even vibrate much at all), and I can't explain why it happens.
Click to expand...

It happens quite a bit actually. I've got a classroom full of Gateway E4300s that have that problem, usually it's not so much the RAM unseating itself as it is minute amounts of corrosion basically block some of the pins on the DIMM. In the case of the E4300s it seems to be excess humidity in the room accelerates the process, usually removing and reseating the module a couple of times gets rid of the junk, then I blow the slot out with canned air. If that fails, I lightly use a pencil eraser on the RAM module's contacts.
 
I would check your o/c. I don't trust memtest that much. Run prime or your favorate games for a few hours. Thats the best test.
 
You must log in or register to reply here.
Back
Top