Multiple BSOD's on new machine

[Danny72170]

Ran Furmark again, and it started to level off at 85c. It peaked at 90c at which point the program started an alarm. It ran nearly a minute in the 85-90c range and the fan sounded like it was working fine. The video didn't appear to be glitchy and I was getting around 22 fps.

The resting temp for the GPU appears to be around 55c

 

[Danny72170]

Got another BSOD today and apparently one last night though I never saw the BSOD. I was shutting down the computer and it appeared to shutdown normally but it restarted instead and when it came back to windows there was a message saying there was an error. I've enclosed the dump files with this message and I'll be running memtest86+ all day. Haven't had a chance to run it for any length until now.

 

[Danny72170]

OK, ran memtest86+ again today for a little more than 7 hours. It ran through 7 passes and produced 2 errors:

Error #1 was on test 5, pass 3 at an address at 629.8MB
Error #2 was on test 5, pass 6 at an address at 628.5MB

Memtest seems to find errors when all 6 gb is in the computer but when I ran the test for each individual memory stick and slot no errors were found.

 

[usasma]

Motherboard, CPU, and RAM are involved in handling memory (most commonly). This doesn't consider video RAM AFAIK - but we can test that also.

Please run these free stress tests. The first is for video RAM while the second will test the CPU and mobo (to a certain extent). Please run the Blend, Small FFT's, and Large FFT's tests in Prime 95 (the second test).

FurMark download site: FurMark: VGA Stress Test, Graphics Card and GPU Stability Test, Burn-in Test, OpenGL Benchmark and GPU Temperature | oZone3D.Net
FurMark Setup:
- If you have more than one GPU, select Multi-GPU during setup
- In the Run mode box, select "Stability Test" and "Log GPU Temperature"
Click "Go" to start the test
- Run the test until the GPU temperature maxes out - or until you start having problems (whichever comes first).
NOTE: Set the alarm to go off at 90ºC. Then watch the system from that point on. If the system doesn't display a temperature, watch it constantly and turn it off at the first sign of video problems. DO NOT leave it it unmonitored, it can DAMAGE your video card!!!
- Click "Quit" to exit

Prime95 download site: Free Software - GIMPS
Prime95 Setup:
- extract the contents of the zip file to a location of your choice
- double click on the executable file
- select "Just stress testing"
- select the "Blend" test. If you've already run MemTest overnight please run the "Small FFTs" test instead. (run all 3 if you find a problem and note how long it takes to error out with each)
- "Number of torture test threads to run" should equal the number of CPU's times 2 (if you're using hyperthreading).
The easiest way to figure this out is to go to Task Manager...Performance tab - and see the number of boxes under CPU Usage History
Then run the test for 6 to 24 hours - or until you get errors (whichever comes first).
This won't necessarily crash the system - but check the output in the test window for errors.
The Test selection box and the stress.txt file describes what components that the program stresses.
More details on the use of this test: Torture test your CPU with Prime95

Two other video stress tests (may be more stressful than FurMark):
Video Memory stress Test - МИРNVIDIA / Утилиты / VMT
Artifact Locator - МИРNVIDIA / Утилиты / Artifact Locator
Sorry, but I don't read the language that this website is made in.

Another interesting test that just came to my attention: Download - OCCT Website english
USE AT YOUR OWN RISK - program doesn't have a whole bunch of safety features to protect you from yourself!

 

[Danny72170]

I've run Furmark and Prime95 several times now but will run them again.

 

[usasma]

Just FYI - I have eye problems and have had to skip re-reading the posts before I reply. In this case I didn't remember your previous efforts. My apologies.

If they (Furmark and Prime95) didn't show anything before, they're not likely to show anything now.
Have you checked your BIOS to see if there are any updates that apply to the memory?

Where do we stand for the overheating stuff? Have you fiddled around inside the case (after unplugging it) to see if anything seems to be very hot? BE VERY, VERY CAREFUL!!!

How close together are the RAM sticks when all are in the system? Try running MemTest with all sticks in - AND pointing a house fan at the sticks to ensure that they remain cool.

 

[Danny72170]

In previous tests, Furmark got to 90c within 30 seconds at which time I shut it down.

Prime 95 results the last time I did them was as follows:
Blend - first error at 1 minute and last at 1 hour and 15 minutes
Large FFT - first error at 2 minutes and last at 34 minutes
Small FFT - No errors after 9 hours at which time I shut it down

I'll check on Bios updates

Edit to add: As far as the memory sticks go, there are 6 slots and the 3 memory sticks are in the 1st, 3rd and 5th slots

 

[usasma]

If rooting around inside the case doesn't reveal any very hot areas, then I'd try the MemTest with a fan that I described above.

If the MemTest still fails, then I'm guessing it's a motherboard problem. So the BIOS update is our best bet at this point. (if the MemTest doesn't fail, then it's most likely overheating memory that is the cause).

If I'm wrong, then it's gotta be the CPU - which is rather expensive to replace (especially since I'm guessing).

Don't buy anything that you can't return - this is all guesswork on my part. At one point you may have to choose between mobo and CPU - but don't know if now is the time.

 

[Danny72170]

You wouldn't happen to know how to check which bios version I have through Windows? I'm on the motherboard website and I can't find out if the latest version is what I have.

 

[usasma]

From the BSOD summary in post #3:

BiosVersion = V8.14
BiosReleaseDate = 11/02/2010

FWIW - sometimes it's listed in MSINFO32.exe (runs in Windows)
or in systeminfo.exe (runs in a command prompt).

 

[Danny72170]

Ok thanks, there's been an update since then. The newest BIOS is dated 3/19/11 and the invoice date when I bought the computer was 3/23/11. The update says improved memory compatibility but so does the update from 11/2/10.

http://www.msi.com/product/mb/X58-Pro-E.html#?div=BIOS

 

[Danny72170]

BIOS is updated. I'll run another memtest tomorrow and let you know the results.

 

[usasma]

BIOS updates aren't a certain fix - especially when the documentation doesn't spell it out in detail.
Good luck!

 

[Danny72170]

I ran a memtest overnight. 7 passes with 1 error.

Error #1 - Test 5, Pass 6 in an address at 2029.8MB

 

[usasma]

If that's with a fan blowing on the RAM - then it's gotta be either the mobo or the CPU

It's most likely the mobo (from the Prime95 results)
- no errors in Small FFT's - this tests the CPU cache and very little RAM
- errors in the blend test - lot's of RAM tested, but also tests a bit of CPU and mobo
- errors in Large FFT's - this tests a static area of the RAM (less intense than the Blend) and stresses the CPU/mobo interface more than Blend.

But, further confusing things is the architecture of your CPU ( Intel Core - Wikipedia, the free encyclopedia ) - this means that the "mobo" stuff we're looking at with Prime95 is handled in the CPU.

Still, if I were a betting man, I'd bet on this being a motherboard issue rather than a CPU problem. All bets are off if the system has been subjected to damaging overheating or voltage surges (these will indiscriminately mess up components at random).

In short, it's your choice (and your money). Try to replace the component from a place that will allow you to return it for a full refund - next best is those that'll let you return it for a restocking fee.

Good luck - and please let us know how it turns out!

 

[Danny72170]

Ok thanks for all the help