Instability with new C2D E6850 and EVGA 680i SLI (all stock clocks)

Frazas

Member
Jun 26, 2001
124
0
0
Before anything else, here are the computer specs:

Core2Duo E6850 3.0Ghz
EVGA Nforce 680i SLI
OCZ 2GB DDR2 PC2-8500 SLI XTC
Seasonic S12-500W
2 x GEFORCE 7800GTX (SLI)
4 HDs (1 - 10000rpm and 3 - 7200 rpm)
1 DVD-RW
1 Creative X-FI
6 120mm fans (NZXT tower case)

I am having problems in my computer that was upgraded a couple of days ago with a new CPU, mobo and memory, and I haven't even started messing with overclocks - instability is always with stock clocks.

The instability happens whenever I run any CPU stability test like Prime95, Orthos or Everest (both Vista and XP give the same errors with fresh installs on both of them). After a couple of seconds of any of the tests (usually doesn't even make it to 1 minute) it gives a hardware error and stops the test. At first I thought that it might be a memory problem, so I tested with Memtest and it passed with no errors. Secondly I thought that it might be a power supply not giving enough "juice" to all the hardware, so I stripped the system to only one graphic card, 2 hard drives (with Vista and XP) and one memory DIMM, but still the errors occurred.

The registered voltages on idle/load in the full system and the stripped down are:

+3,3v - 3,18v
+5v - 4,92v
+12v - 11,93v

They are all below their normal values. Might the problem be here?

All the temp reading are within the normal parameters:
CPU - 35º
Northbridge - 45º

Lastly, just for curiosity I tested every stability test with one of the CPU cores and with both... and here is what happened:

- with only core#0 all the tests pass with flying colors;

- with only core#1 all the tests fail;

- with both cores all the tests fail;

Since I can now replicate the error every time, it is just a matter of finding which component is failing, which is why I am posting this - to get your expertise on this issues.

I don't have any spare parts that could help me separating the problem, so from the facts stated above, which component might be the culprit: CPU, mobo or power supply?
 

robisbell

Banned
Oct 27, 2007
3,621
0
0
well, did you update the BIOS to the last non beta release? you did do a re-install since you replaced major parts of the PC right?
 

Frazas

Member
Jun 26, 2001
124
0
0
Yes, did the latest BIOS update, and had to reinstall both the SOs (XP and Vista) due to the driver conflicts.

I just did not reset the CMOS and removed the battery after the processor update.

Have to try that when I get home.
 

Frazas

Member
Jun 26, 2001
124
0
0
I've done everything that I could think off...

I reset the CMOS and took the battery off, I messed with the CPU BIOS settings, I downgraded to P30 BIOS, messed with the BIOS settings, re-upgraded do P30, messed with BIOS settings.

I rechecked the CPU physical location, reseated the cooler...

I am loosing my mind here...
 

pol II

Member
Oct 4, 2004
173
0
0
Originally posted by: Frazas
Before anything else, here are the computer specs:

Core2Duo E6850 3.0Ghz
EVGA Nforce 680i SLI
OCZ 2GB DDR2 PC2-8500 SLI XTC
Seasonic S12-500W
2 x GEFORCE 7800GTX (SLI)
4 HDs (1 - 10000rpm and 3 - 7200 rpm)
1 DVD-RW
1 Creative X-FI
6 120mm fans (NZXT tower case)

I am having problems in my computer that was upgraded a couple of days ago with a new CPU, mobo and memory, and I haven't even started messing with overclocks - instability is always with stock clocks.

The instability happens whenever I run any CPU stability test like Prime95, Orthos or Everest (both Vista and XP give the same errors with fresh installs on both of them). After a couple of seconds of any of the tests (usually doesn't even make it to 1 minute) it gives a hardware error and stops the test. At first I thought that it might be a memory problem, so I tested with Memtest and it passed with no errors. Secondly I thought that it might be a power supply not giving enough "juice" to all the hardware, so I stripped the system to only one graphic card, 2 hard drives (with Vista and XP) and one memory DIMM, but still the errors occurred.

The registered voltages on idle/load in the full system and the stripped down are:

+3,3v - 3,18v
+5v - 4,92v
+12v - 11,93v

They are all below their normal values. Might the problem be here?

A fluctuation of 5% is usually okay; did you take those measurements via software or multimeter? Things look okay though from PSU side of things.

Originally posted by: Frazas
All the temp reading are within the normal parameters:
CPU - 35º
Northbridge - 45º

Lastly, just for curiosity I tested every stability test with one of the CPU cores and with both... and here is what happened:

- with only core#0 all the tests pass with flying colors;

- with only core#1 all the tests fail;

- with both cores all the tests fail;

Since I can now replicate the error every time, it is just a matter of finding which component is failing, which is why I am posting this - to get your expertise on this issues.

I don't have any spare parts that could help me separating the problem, so from the facts stated above, which component might be the culprit: CPU, mobo or power supply?

Seems to me like core #1 is the problem, so I would focus attention on the processor.

 

Frazas

Member
Jun 26, 2001
124
0
0
Thank for the replies.

The voltages are from values in the BIOS monitor.

Both the CPU and motherboard are new (bought them on Tuesday) and the 24pin/8 pin/4 pin molex are all connected to the motherbard.

 

robisbell

Banned
Oct 27, 2007
3,621
0
0
it's a toss up, I am not seeing anything saying it's definitely one or the other. I'd swap out the processor and see, if it was me.
 

TestedAcorn

Golden Member
Mar 14, 2007
1,228
1
0
I'm not exactly sure what the problem is, but I have heard that a lot of mobos have trouble with the e6*50 processors. Might be the problem, might not.
 

Jaguar07

Junior Member
Nov 14, 2007
2
0
0
After you flashed the BIOS to the latest non beta, did you reset the CMOS to defaults, then reboot, then modify them to suit?
 

Frazas

Member
Jun 26, 2001
124
0
0
The problem is now detected. The bad component was the CPU. After testing a new motherboard the problem replicated. When testing the CPU in a Asus P35 board, the problem replicated.

How unlucky can a guy be to get a bad CPU with an error like this?

My only problem now is that the E6850 is out of stock, and I'll have to wait till next week to get a new one.

Thanks for all your support, and keep in mind this error. It might come up again for another user.