TWIN2X4096-8500C5D Produces errors after 2-4 weeks of use.

Hello,

 

I recently purchased parts to build 2 systems from Newegg, one for myself, and one for my wife. The builds went great, and everything seemed to work perfectly.

 

The build went great, there were absolutely no problems with either system, I proceeded to load openSUSE 11.1 X86_64 on to my system, and Windows Vista Ultimate 64bit on my wife's system. Both computers operated perfectly for about 3 weeks. I started noticing crashes on my linux system, firefox would terminate, flash wouldn't play, apps would crash with CRC errors. My first thought was that it was a problem with linux, I reinstalled, and began noticing the problems right away during the install, MD5 sums were incorrect on the same disk I had previously installed from, packages were corrupted and would not install, etc etc.

 

My next step was a hardware failure, I tested memory first, being the easiest to test. I fired up Memtest x86, and found right away that I had massive errors. Over 300 errors were detected in the first 2 minutes alone. At this point, since I had identical memory in the other system, I decided to do a test to rule out the other hardware, I placed the suspect memory in my wifes system, and carried out memtest, the same errors were produced on an entirely different system. I placed the good memory from her system into my system, and the test showed no errors.

 

So I thought I had found my problem, simple easy, no mess, I replaced the memory. Everything was running fine again, until about 3 weeks again. The errors started producing in my wife's system this time, her memory modules failed the EXACT same way mine had. This time, I thought maybe the BIOS settings were wrong, causing errors of some kind, So I lurked on these forums, and found the recommended settings for these memory modules, and my main board, no luck. I next suspected a heat problem, even though both systems are running incredibly cool. I tried to use a dedicated fan for the memory modules, this didn't change anything. I am now in a situation, where I have 2 sets of bad memory, both sets pop numerous errors in memtest, both mine and my wifes memory have gone bad over a short period of time.

 

The only difference in our systems, is the operating system we run, She runs Vista Ultimate 64bit, I run SUSE11.1 X86_64. I am simply at the end of my knowledge in how to fix this issue. Never in my 17 year experience of self building my computers, have I experienced a problem of this sort, where the memory starts off good, and goes bad over a short period of time.

 

 

Please help, My system is more stable due to the fact that it's linux, So my apps crash, however my OS will stay stable. However, my wifes windows system is BSOD every 10 minutes. The one thing I would like to mention, is once the errors in the memory happen, If I power down the system, and wait approx. 5 minutes, The system will run error free for a period of 4-18 hours. However, at that time, the system will once again produce the errors.

 

Thanks in advance for your help

 

Sincerely

Mark Brown

Ram Guy,

 

I have already tested the modules with memtest 86, and they fail. I have retested these modules about 30 times, at diffrerent intervals. They fail at inversion 1's and 0's. and start producing errors at different memory addresses. I usually end up with over 300 errors in the first few minutes. If there is any more specific information you need, please let me know.

 

 

Sincerely

CynicalPride

Are you disabling Legacy USB support when testing the memory? It can cause false positives if not disabled.

Ram Guy & Wired,

 

Sorry it took so long to get back to you on this issue, however, it is an intermittent problem. I was copying a file this evening and my system hard locked(Linux doesn't do that often). Upon trying to reboot, I received CRC errors, which prompted me to pop in memtest86+. I disabled legacy usb, as suggested by wired, and ran the test. I am including a copy of the photo I took of the errors.

 

My system would NOT boot into the OS until I powered down, a simple reset would not do it. Once I powered down, and turned it back on, the system booted with no problems.

Something's wrong somewhere, as memtest is reporting DDR1 @ CAS 1????

Wired & Ram Guy,

 

 

I agree, something is wrong somewhere, however, I can't seem to find where that somewhere is. I have tried setting the timings manually, I have tried upping the voltages, I have tried everything I can think of. I have 2 identical systems that report the exact same thing on memtest.

 

 

I will try a bios upgrade, and report back to you the results.

 

Sincerely

Mark Brown

Ram Guy,

 

Ok, Updated BIOS to latest version, loaded setup defaults, and set the timings manually as posted in the link in your signature for the dominator modules.

 

 

Upon running memtest86+, I was greeted with 32 errors, within 30 seconds. Perhaps you can suggest the timing values for the ASUS M4A79?, setting to 5-5-5-15-2T, seemed to cause me even more issue, and strangely enough, memtest still has the weird CAS reading...Picture attached.

 

Sincerely

Mark Brown

Test one stick at a time in each slot.

Wired,

 

I started to do this, and had errors pop up right away, Something interesting though:

 

Slot 1 : 1067

Slot 2: 800

Slot 3/4 :667

 

I simply cannot continue this test, due to the random factor of this problem. The first 2 test, showed errors, after that, no test show errors, because I had powered down the system long enough, to cause this issue to no longer exist, I will probably have to wait a few hours before I am able to retest.

 

 

I have 2 very similar systems, causing 2 very similar problems. The weird thing here, is that when I replaced the memory the first time, it took nearly 2-3 weeks for the problem to manifest itself again. When I take the modules that were corrupted in my system, and swap them out to my wifes system, The problem carries over. So it has SOMETHING to do with the actual memory modules.

 

Ram Guy, if you could please tell me the exact settings for my mainboard, for these memory modules, It would help me tremendously, I am trying to eliminate user settings in this equation, however I can't seem to find the specific timing for these modules in the M4A79 board.

 

 

Sincerely,

Mark Brown

Ram Guy,

 

As Stated in my original post, I have 2 systems, with the exact same MB, Processor, Memory, and Graphics Cards. Here is my plight:

 

Initial System Build: Everything OK

 

2-3 Weeks Later: Crashes in Linux, Format, upon re-install receive checksum errors and MD5 errors. Memtest throws errors, indicating bad memory. Remove memory modules from my system, and place them in my wifes system. Run memtest, receive exact same errors. Place Good? Memory modules from my wifes system into my system, no errors.

 

2 Days Later: Exchange Bad? memory modules at local retail outlet, for new memory modules, install them in my system, Everything is OK.

 

2-3 Weeks Later: My wifes system fails and starts producing BSOD, with CRC errors, run memtest, and now her memory modules have developed the EXACT same problem mine had, hers just took 4-6 weeks.

 

1-2 Weeks Later: Memory modules from my system(Ones that have already been replaced) Develop the exact same problem as the other sets have.

 

Present: I am beating my head against my desk, trying to solve a problem, which I have never seen the likes of , in my 17 years of computer building career.

 

 

To sum this up, I am not convinced that the memory modules are bad, however, I am also not convinced that anything is "bad", other than maybe some "settings". However, being that this issue seems to produce errors in memtest, and I can physically remove the memory, transfer it to another system, and the problem carries over, I am starting with memory. If I can't get this issue to pan out here, my next stop is going to be with AMD, and after that, with ASUS.

 

 

The interesting thing about this problem, is that perfectly good memory modules, will start out working GREAT, and degrade over a period of time, and cause MASSIVE issues, everything from program errors, to full on system crashes. I have not overclocked anything at all, This system was built out of all matching parts, in fact the Motherboard manual, suggest these memory modules as well as the Phenom II 940. So if you know the settings, or can find the settings for the M4A79 board, that I should be running, I can eliminate "Settings" as a cause of this issue.

 

I will snap a few pictures of my current settings for your review, and you let me know if you think they are suspect.

 

 

Sincerely,

Mark Brown

Bios Photos Attached:

Ram Guy,

 

Yea, but two motherboards, causing the exact same issues, in 2 systems? Did you take a look at my settings, and confirm that they are within spec? Once I rule out the settings I intend to begin a new at ASUS.

My experience tells me to check the Power Supply. Often PS problems show up in weird MB/Mem situations. Considering both systems do the same thing I would suspect the type of PS. I would change brands and up the wattage. A small cost if it resolves a lot of work stoppage.

Was this problem ever solved? Was it ram, MB or PSU ?

I have the same issue...i have already replaced my RAM and after one WEEK with 1067 2.1v and timings in auto starts again with bluescreens! I can play sometimes half an Hour or more without problems and after that in IDLE system restart/crashed(Event monitor error...looking for solution etc.) without reason!

Now i changed settings in auto mean ddr2-800 and seems to work but i am not so sure because i didn't play or test since the problem....will see. Iam so disappointed all brand new but i cant have a stable system!

I decided to change the CPU after replacing RAMs from corsair and it works....Now i have a stable system with DDR2 1067 2.1V. I changed the CPU because of the weird crashes in IDLE and the thought that M4A79 is one of the best MB in the Market! So it remains to change the CPU(Memory controller in DIE) i risked it and seems to be fine right now!