Jump to content
Corsair Community

Memory errors go away if power disconnected


Recommended Posts

Can anyone suggest what might be going on?



PC runs for several days, then reboots.

Memtest86+ shows a lot of errors.

Errors remain after restart either by Reset button or Power button (configured for 'off')

System boots OK into Memtest86+. Many errors shown.


However if I disconnect the main lead for 5 seconds, the problems are cleared.

Memtest86+ will not detect errors even after many hours.


Errors return after several days running Windows (2000)


PS is only 370W but as you can see below this system has modest power needs.


System is well cooled - Processor 45, System 34, GPU 53


Any suggestions??



Bios set 5 5 5 15 2.1v

Spread Spectrum - Disabled

Link to comment
Share on other sites

My system will become unstable after its been running for a few days to 2 weeks.... If I reboot and run memtest86 at that point, I get thousands of errors/pass. If I shut down the computer for 5-10 seconds, and then run memtest again..... I get no errors, even if I let memtest run for 24 hrs (20+ passes).


I need to find a solution to this problem!


I have a 625W PS, so I don't think my power supply is an issue.....


edit: I have 4 sticks of memory, and I've been using only 2 of them at a time..... Despite rotating them, the problem seems to keep occurring. I've also tried default BIOS settings, manually setting the timings, using slower memory speeds, and using different memory slots without success. I don't think its the memory at this point, but I'm not sure what else it could be. Any advice on how to further diagnose this problem would be greatly appreciated!!!!

Link to comment
Share on other sites

  • 2 weeks later...

As Indica mentioned by linking to my thread, I have the exact same problem.

And this problem sounds very similar too:




The "Clock Skew" tweaking I mentioned increased the time before the system fails from ~8 hours to ~19 hours but that's it so far.

Like Mr. G I also have 8 GB of RAM. And running with only 4GB didn't help me either.


We all have different Mainboards, CPUs and Power Supplies.

The only thing we all have in common is the same RAM so I think it's a good guess to believe that's the cause of the problems.

Maybe there's a bad batch?


What revision of the RAMs do you all have?

Mine are are all Rev 2.1.


One more thought: The 1x revisions needed 2.2 Volts which has been reduced to 2.1 Volts for the later revisions. Could that be the problem? Is it possible that this was a little too optimistic? That the later revs still need higher Voltage to work stable?



Anyway, I'm glad that I'm no longer the only one with this problem. Maybe more input finally helps solving this issue.

Link to comment
Share on other sites

Mine are all version 2.1 also, says right on the stick 2.1 volts


I also am tryin to run 8gb running with 4 dosent help. Although one of my sticks now has permanent errors that arent wiped when i power off, im in the process of RMA at the moment when i get my 2 fresh sticks i will try just those, maybe i will get lucky and its just all bad luck with faulty ram.

Link to comment
Share on other sites

  • Corsair Employees
With 4 modules I would suggest setting the memory frequency at DDR667 and set the memory Voltage to 2.1 Volts and set the NB/MCH/SPP Voltage to +.2 Volts as well and test the system with http://www.memtest.org. In addition, with some MB's (Mostly ASUS) you have to disable legacy USB in the bios when running any memory test.
Link to comment
Share on other sites

  • 1 month later...

This is an update to the problem I raised at the head of this thread:


I have TWO identical PCs, both have MSI P43 Neo F and TwinX 4GB TWIN2X4096-8500C5DF


I have been testing both PCs with a single memory strip in slot 1,

rotating the four strips through both PCs.


Test have been:

Spread spectrum always disabled

Latest BIOS (V1.10) on both boards

5-5-5-15 1066 2.1v

5-5-5-18 1066 2.1v

5-5-5-18 800 2.1v

5-7-7-24 1066 2.1v

5-7-7-24 800 1.8v

5-7-7-24 800 2.1v


I have obtained alternative memory (DDR2 800) running at the auto setting of

6-6-6-18 800 1.8v


This is stable in both systems both singly and as a pair.

Falure returns with Corsair strips.


Now there seems to be a pattern about the failures:

- two strips BSOD in hours, the other two take several days

- after failure memtest86+ often (always??) reports the same byte in error

for same strip, but different for byte for each strip

(1-000000f0, 2-0000ff00, 3-00f00000, 4-00ff0000)

(failures may be 1 to 4 bits within each 'f')



Where do I go from here?

Link to comment
Share on other sites


This topic is now archived and is closed to further replies.

  • Create New...