Memtest Fail Questions & BSOD's

[BLACKKETYL]

Here's my Sys specs:

GIGABYTE GA-890FXA-UD5 AM3 AMD 890FX SATA 6Gb/s USB 3.0 ATX AMD Motherboard
SAPPHIRE 100290SR Radeon HD 5870 (Cypress XT) 2GB 256-bit GDDR5 PCI Express 2.1 x16 HDCP Ready CrossFireX Support Eyefinity 6 Edition Video Card
Rosewill Libertas Series LIB-650 650W Continuous @40°C,80 PLUS Certified, Full Modular, ATX12V v2.3/EPS12V v2.92,SLI Ready, CrossFire Ready, Active PFC, Compatible with Core i7, i5 Power Supply
AMD Phenom II X6 1090T Black Edition Thuban 3.2GHz Socket AM3 125W Six-Core Desktop Processor HDT90ZFBGRBOX
GeIL Black Dragon 8GB (4 x 2GB) 240-Pin DDR3 SDRAM DDR3 1600 (PC3 12800) Desktop Memory Model GB38GB1600C9QC
HITACHI Deskstar H3IK10003272SP (0S02860) 1TB 7200 RPM SATA 3.0Gb/s 3.5" Internal Hard Drive
ASUS DRW-24B1ST/BLK/B/AS Black SATA 24X DVD Burner - Bulk - OEM
Rosewill RK-7310 Black USB Wired Super Slim Multimedia Keyboard
Microsoft Windows 7 Home Premium Full
Rosewill RNX-N180PCe PCI Express Revision 1.1 Wireless Adapter

Hello all:

My system is unstable and I've decided to finally do something about it. I get BSOD's fairly regularly. I'd attach the dumps but I'm writing this from the office computer so it's going have to wait. Checking the 50+ dumps, the bulk of the crashes come from ntoskrnl.exe/ntkrnlpa.exe/atikmdag.sys. The "nt" are apart of the kernel so I really doubt that's the root of the problem, it's something else I'm sure. The ATI, is from the GPU but not real sure what's causing it. Catalyst and the associated drivers are up to date it would appear.



Last night I ran:

• Dskchk.ext --> No problems on C:. I ran again on my other drive pulled from my previous comp. A few errors but it appears they were fixed. Nothing major on this HD. Just WoW and some misc files.
• CCleaner ---> Found some lost registry files, cleaned it up and no issues to report.
• Prime95 ---> I only ran for about 30 mins before bed, no issues. I know it needs to run for a lot longer but my hunch is that the CPU is not the issue here.
• Memtest86+ ---> Here's where I got some issues.

Test#1 -----> Ran for about <5mins, screen went blank no response did a soft reset...
Test#2------> Ran again errors started popping up somewhere in between the 5-10min mark. The errors just kept piling up in a few mins I was at like 350k.
Test#3-6---> Ran each stick individually in the first slot failed all failing at around the 1:45 mark or while beginning memtest's test #7. The error just kept piling up like the above test.

So what do you think at this point? Is it the ram stick or the MOBO? It's sort of strange that all of the sticks would fail individually at the exact same time. One thing to note, I checked Gigabyte's website and my particular ramstick is not listed on their compatability chart. They also said that they can't check every stick of RAM on the market so it could be fine just not checked.

The majority of the BSOD's happen while gaming. WoW, Rift, The Witcher, Sims 4, NWN2. Pretty much all of my games I play except EVE for some reason. That game's never crashed on me.

Any help with this be great. If you have any questions please let me know.

Thanks,

 

[SledgeDG]

Did you try the sticks one by one in the other slots?
(That's how you determin if it's the RAM itself or the slot that is bad. It's rather uncommon that all RAMs give up the same time (unless a surge ran through the comp caused by a thunderstorm or by false handling (static charges)

-DG

 

[BLACKKETYL]

I'll give that a whirl after work here and report back.

I really hope it's just faulty RAM.

 

[BLACKKETYL]

Hello,

I went ahead and attached the dumps and perfmon here. There's 51 BSODS I believe since I built the computer.



A few updates:

• I learned that the RAM clock speed was not at the RAM's intended speed. Changed to 1600mhz from 1333mhz.
• The timing was also off. Changed to RAM specs. I thought this might have been the problem but Memtest still reported errors at about the same time during the test.
• I got about half way through individually testing each RAM stick in each DIMM slot. Each failed. I'm pretty sure the other two sticks will yield the same result.
• The ATi GPU is updated
• The Realtek driver has been updated as of this morning.

Please let me know if you have any questions.

Thanks,

 

[Dave76]

There is a known issue with AMD CPUs integrated memory controllers having problems running RAM at 1600 MHz and above.
VERY IMPORTANT INFORMATION REGARDING AMD AM3 CPU's and RAM SPEEDS
Q&A for the topic VERY IMPORTANT INFORMATION REGARDING AMD AM3 CPU's & RAM....

For starters change the RAM back to 1333MHz.

It is unusual to have four individual RAM cards fail when tested one at a time.
Test all the RAM cards in more than one motherboard slot, preferably in all four slots.
Unless, as mentioned, there has been a voltage spike, ground fault, or static discharge.

Lets take a look at your setup.

Information on the memtest86+ RAM testing.
When you run memtest86+ you are looking for bad RAM cards, bad motherboard slots, so you need to follow the testing procedure in the tutorial.
Sometimes errors from these tests can be corrected by adjusting the settings.
RAM - Test with Memtest86+

D/L CPUZ and post a snip of the CPU, Mainboard, Memory, and SPD tabs.
For posting the information , enter 'Snipping tool' in the Start button > Search box, hit enter. Select the area you want to post and save to a convenient place.
How to Post a Screenshot in Seven Forums

Have you been checking your temperatures?
UseCore Temp , to check the CPU temps when at idle, under load and/or testing.
Let us know the minimum and maximum temps you get.


Go to your BIOS and tell us what the CPU, RAM and the CPU integrated memory controller voltages are.


Anytime you have more than 4GB RAM, 2GB RAM cards, more than 2 RAM cards, you will likely need to adjust some settings to get it to run stable.


With the above info we can see if anything needs adjusted.


Running the debugger........
Will add info as I get it.

As suspected

*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
Use !analyze -v to get detailed debugging information.
BugCheck 109, {a3a039d89ca8a4e9, b3b7465eef2576df, fffff800033589f0, 1}
*** WARNING: Unable to verify timestamp for win32k.sys
*** ERROR: Module load completed but symbols could not be loaded for win32k.sys
[COLOR=red]Probably caused by : memory_corruption[/COLOR]
Followup: memory_corruption
---------
 
*******************************************************************************
*                                                                             *
*                        Bugcheck Analysis                                    *
*                                                                             *
*******************************************************************************
CRITICAL_STRUCTURE_CORRUPTION (109)
This bugcheck is generated when the kernel detects that critical kernel code or
data have been corrupted. There are generally three causes for a corruption:
1) A driver has inadvertently or deliberately modified critical kernel code
 or data. See [URL]http://www.microsoft.com/whdc/driver/kernel/64bitPatching.mspx[/URL]
2) A developer attempted to set a normal kernel breakpoint using a kernel
 debugger that was not attached when the system was booted. Normal breakpoints,
 "bp", can only be set if the debugger is attached at boot time. Hardware
 breakpoints, "ba", can be set at any time.
3) A hardware corruption occurred, e.g. failing RAM holding kernel code or data.
Arguments:
Arg1: a3a039d89ca8a4e9, Reserved
Arg2: b3b7465eef2576df, Reserved
Arg3: fffff800033589f0, Failure type dependent information
Arg4: 0000000000000001, Type of corrupted region, can be
 0 : A generic data region
 1 : Modification of a function or .pdata
 2 : A processor IDT
 3 : A processor GDT
 4 : Type 1 process list corruption
 5 : Type 2 process list corruption
 6 : Debug routine modification
 7 : Critical MSR modification
Debugging Details:
------------------

BUGCHECK_STR:  0x109
CUSTOMER_CRASH_COUNT:  1
DEFAULT_BUCKET_ID:  CODE_CORRUPTION
PROCESS_NAME:  System
CURRENT_IRQL:  0
LAST_CONTROL_TRANSFER:  from 0000000000000000 to fffff800030bd700
STACK_TEXT:  
fffff880`031925d8 00000000`00000000 : 00000000`00000109 a3a039d8`9ca8a4e9 b3b7465e`ef2576df fffff800`033589f0 : nt!KeBugCheckEx

STACK_COMMAND:  kb
CHKIMG_EXTENSION: !chkimg -lo 50 -d !nt
    fffff80003358bbf - nt!CmpSetValueKeyExisting+1cf
 [ 30:70 ]
1 error : !nt (fffff80003358bbf)
MODULE_NAME: memory_corruption
IMAGE_NAME:  memory_corruption
FOLLOWUP_NAME:  memory_corruption
DEBUG_FLR_IMAGE_TIMESTAMP:  0
MEMORY_CORRUPTOR:  ONE_BIT
FAILURE_BUCKET_ID:  X64_MEMORY_CORRUPTION_ONE_BIT
BUCKET_ID:  X64_MEMORY_CORRUPTION_ONE_BIT
Followup: memory_corruption
---------

 

[BLACKKETYL]

Thanks for the reply.

I'll go back and change it back to 1333mhz. The memory corruption dump last weekend, that's the first time ever I've seen that BSOD on this rig. I've had it since 11/10. I was playing a new game, not sure it has anything to do with it or not.

After work here I'll test the remaining sticks in each slot and post results. My setup is the default unganged FYI.

I have CPUz installed, I post the details here.

I ran prime95, and under max load highest temp was 60c. I'll have to double check on the idle temp. I'll get the remaining answers to your questions after work.

Thanks for your help.

 

[Dave76]

You're welcome.

Good information.

Have you had 8GB RAM since 11/10?
All matched, same brand etc.?

Unganged is the preferred setting for your RAM.

The Prime95 temp is good.

 

[BLACKKETYL]

You're welcome.

Good information.

Have you had 8GB RAM since 11/10?
All matched, same brand etc.?

Unganged is the preferred setting for your RAM.

The Prime95 temp is good.

Yes sir, all matched and same brand too.

 

[Dave76]

If you post the info requested above, we can check the RAM settings.

More than 4GB RAM, 2GB RAM cards and more than two RAM cards usually need a voltage bump for the integrated memory controller.

Can you enter your system specs in the 'My System Specs' drop box in the lower left corner of any of your posts?
This will keep everyone from having to go to the first post to see your specs.

 

[BLACKKETYL]

Hello,

I changed the RAMM back to 1333mhz. I tested all of the RAM sticks in each DIMM and they all failed at about the same time:



Here's the specs you've requested, I hope:

• Sys Temp 30c Idle
• CPU Temp 41c Idle
• Vcore 1.472v
• DDR3 1.5 running at 1.52vish
• +3.3v running at 3.4vish
• 12v running at 12.1v
• System Voltage 2.5
• DRAM Voltage at 1.51v which is running lower than the given RAM specs of 1.6v
• DRAM VTT voltage .750v
• HT link Voltage 1.2v
• NB/PCIe/PLL Voltage 1.8v

Random question, Turbo CPB is disabled, should I turn that on?

Here are the screenshots of CPUz:
View attachment 156578
View attachment 156579
View attachment 156580

Please let me know if you have any questions.

Thanks,

 

[Dave76]

I need the CPUZ SPD tab.


From your BIOS the 'CPU NB VID Control' voltage, this is the CPU integrated memory controller voltage and is one that will need adjusted.

 

[BLACKKETYL]

Happy Friday Bump,

Let me go check the BIOS...

Ok I found the CPU Northbridge Vid Control Voltage however it just says "normal", it does give me the ability to adjust the voltage in small increments however. But I don't know what the default voltage is. CPU Voltage Control gives me the same option.

Ok here's the SPD screenshot:

View attachment 156662

 

[BLACKKETYL]

Also FWIW I played a game last night, one that gave me all sorts of problems, didn't crash the entire time I played. So I believe what whatever we've been updating or tweaking the last few days may be helping.

I ran a torture test this morning with some different settings. It brought core temps up to 62 and peaks at 63c. Should I be concerned? AMD says 62c is the max it should get up to.

 

[Dave76]

Go to your BIOS and change the 'CPU NB VID Control' to manual, if you still can't see a voltage number then increase it by the smallest amount.
The default voltage is 1.10v, change it to 1.25v.

Change the DRAM Voltage to 1.55v.

Run memtest86+ again, for 7 passes, best to run overnight.
If you get any errors you can stop the test.

Your max CPU temp is 62°C
AMD Phenom II X6 1090T Black Edition

That is pretty warm for an AMD CPU, they usually run cool.
It's safe at that temp, wouldn't go any higher though.
Check that your CPU cooler is not loose, it should be firmly attached.

 

[BLACKKETYL]

Ok thanks again for your help.

I'll change the voltages and check the CPU cooler after work. I'll let you know if there's any problems. If not, I'll run the tests again and report back.

Dave I missed your request from last page about updating "my system specs" I went ahead and updated it so no one has to go back-and-forth.

 

[BLACKKETYL]

Hello again,

Adjusted the RAM/CPU Controller unfortunately it failed again.

Any other ideas? I'm going to find a friend with some DDR3 and hope it's just the RAM sticks. At this point I'm not ready to do that Driver Verifier thing. It seems like trouble. The system is 95% stable, just when I run games but even then it doesn't crash that much.

Thanks again.

 

[Dave76]

Change the 'CPU NB VID Control' to 1.30v.

Change the DRAM Voltage to 1.60v.

Run memtest86+ again, for 7 passes, best to run overnight.
If you get any errors you can stop the test.

These voltages are still within the safe range. It is best to move them up slowly, you want to get a stable system at the lowest voltage possible.
Less voltage, less heat.

If this doesn't pass memtest86+, then borrow your friends RAM to test with.

The other possibilities are a problem with your motherboard or PSU.

Let us know the test results.