Solved having extremely hard bluescreen problems

[joe06]

Dear Jonathan,
great work but i got several question ...
running your test with or without xtreme burn-in @all the time @1FPS
furthermore i dont see any settings in the top left corner CORE MHZ : 0, MEM: 0, Temp 0°C
tried version 1.9.0 and 1.9.1 ...

really got no idea ... having extremely hard bluescreen problems ...

Greez

 

[Jonathan_King]

Hello Joe,

Would you mind creating a BSOD report and posting it in a new thread HERE? Post a link to your new thread here, and I'll check it out. It sounds as if there may be a serious hardware problem with one of the core components or video card (RAM, motherboard, CPU). While you wait for me to look at your BSOD report, double-check your hardware. Make sure the video card is seated properly in the PCI-e slot, the power connectors are hooked up right, remove the RAM and reseat it, etc. Also check for signs of possible overheating, such as dust clogging heat sinks, vents, and fans.

To create a BSOD report, follow the instructions here: http://www.sevenforums.com/crashes-debugging/96879-blue-screen-death-bsod-posting-instructions.html

 

[joe06]

Hey Jonathan your re still alive nice

Firstly, thx alot for your fast reply. The last mb 3-5 hours I tested nearly everything (still not sure about graphic card)...

Memtest for Ram (~1hour)--> No problem
HD Test with HD Tune --> No problem

Now the interesting thing: Prime 95
Using the Small FFT (as you suggested) I get a fatal error @cpu2 (sometimes @both cores)

"FATAL ERROR: Rounding was 0.46484375, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Tue Mar 27 19:44:07 2012]
FATAL ERROR: Rounding was 0.5, expected less than 0.4
Hardware failure detected, consult stress.txt file.
[Tue Mar 27 19:57:31 2012]
and so on...."

These Errors occur after only some mins (not more than 10)

The strange thing is, if I stop 1 worker, e.g. on core 1..the programms runs without problems (and the other way around). But running prime on both cores gives me instantly (as mentioned above) a fatal error and my PC freezes or directly "bluescreens".....

Thus I didnt OC my CPU, I am nearly sure that it is dead....I hope I get a used one by my friend, and I ll test it again tomorrow...im crossing my fingers, hope the graphic card isnt dead aswell....


Thx in advance and best wishes joe

PS: I m trying to create a BSOD (if my pc doenst crash 24/7)...^^

 

[Jonathan_King]

Sounds almost like RAM, in that case. If the CPU itself was the problem, I'd expect the problem to show up on only the faulty core. There is no way to isolate which RAM sticks are being tested with Prime95, however.

I'd encourage you to keep pursuing the RAM route. Try running Memtest86 overnight. A faster, but more labor-intensive method would be to remove all but one stick of RAM and then run the Blend Prime95 test. If you get errors, replace that stick with another one, and test again.

 

[joe06]

I hope i created a the BSOD file the right way ?!

Concerning the RAM (2 x 2GB) maybe 2 month old. I started Memtest 3 times for mb 1hour -->0 Errors.

If I try the Blend test my PC (AIDA64 Extreme Edition) says me 100% CPU usage and thus a fatal error occurs after the mentioned 2-10 mins :///

Temperatures are all checked and fine!

Ahh I forgot to mention....This d.... Pc crashes nearly all the time ...
using some tabs @firefox or IE-->bluescreen
Youtube videos -->crash in ~30-60 secs
Online stream (legal)-->sometimes it works up to mb 5mins but normally ~30-60 secs

BUT: Playing warcraft 3 tft or even Diablo 3 beta works fine at all .... I really got no idea ://

 

[Jonathan_King]

Thanks for the report. I believe we're looking clearly at a RAM (possibly CPU) problem, so go ahead and monkey around with removing one stick and running Prime95 on it. We'll see if we can't trace it down!

FYI - I'm asking the mods to move these posts to a separate thread as we have determined it is not related to the video card or Furmark itself. The old links should redirect to the new thread.

 

[joe06]

Ok, sry if I posted this wrong ://

gonna remove 1 stick, reboot and prime..BRB

 

[joe06]

Ok, here are my results for Stick1:

Stick 1:

First Blend:
Test 1 at 512K Worker 2 = CPU2 stopped at Test1 „Fatal error....“ (1st test of this series) But I didnt had a BSOD

Second Blend:
Test 1 at 512K passed on both workers.
Test 2 at 8K „Fatal error...“ at Worker 2 = CPU2 (3rd test of this series)

Just switched the ram sticks gonna start blend test for Stick 2 now

EDIT:
Tested ram stick 2:
1st try Blend test: like stick 1 (fatal error at cpu2)
2nd try Blend test: again Worker2 = CPU2 got a fatal error @test 1 512k (3rd test of this series)

So, all the time these errors (but all without a BSOD) occur @CPU2 ?!

My first impression, CPU is damaged is right?
(I am a real PC newbie, reading posts for about 10 hours now and still got now idea at all ^^...)

 

[joe06]

Bought a new CPU - Problem solved!

Thx a lot jonathan !

Best wishes Joe

 

[Jonathan_King]

Glad to hear it!