• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Good Overclocking Gone Bad; What Happened?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
So it doesn't appear to be the RAM, since it fails in pretty much the same way with both sets.

I took the Zalman CPU fan/heatsink off, and while I wouldn't say that the thermal meterial had "burned off", I would say that it didn't look like right. It might have happened when I removed the unit, but the material wasn't covering the entire surface of the CPU.

So I removed the material, cleaned it, purified it, and reapplied Arctic Silver 5. It's now running Memtest86+ with my original (FAST) RAM, again at stock BIOS settings. So far so good (for 4 minutes)...
 
When I boot to the BIOS, the Hardware Monitor page says that the CPU temperature is 31 degrees, and the MB temperature is 38 degrees.

It also reports the following voltages:

VCORE: 1.232V
3.3V: 3.296V
5V: 5.171V
12V: 12.355V
 
Originally posted by: InterMurph
When I boot to the BIOS, the Hardware Monitor page says that the CPU temperature is 31 degrees, and the MB temperature is 38 degrees.

It also reports the following voltages:

VCORE: 1.232V
3.3V: 3.296V
5V: 5.171V
12V: 12.355V
Click to expand...


Isnt that vcore low?
 
The motherboard/BIOS picked that value, since everything is set to Auto.

I'll try manually setting it to 1.425, which is what I was using when I was running at 3.65 GHz.
 
No, it didn't stabilize it.

I am now going to try lowering the mutiplier from 11 (2.93 GHz) to 6 (1.60 GHz).

I also ordered an E6300 for $190. I just don't have enough time to spend tinkering with everything else. I need to know if it's the CPU!
 
I lowered the FSB speed from 266 to 200, and it ran for 14 minutes, then locked up.

My E6300 should arrive today; that's my next test.

This is quite frustrating!!!
 
HHhmmm
Do you have a friend with a similar system?
Where you can test the CPU and memory?
Thought about the motherboard going bad?
Since you made so many changes, and still getting lockups.

Rafael
 
The trend I see is this:
1. You run a round of Memtest cold, it lasts for let's say 10 minutes.
2. You reboot and it lasts for 4 min.
3. You reboot again and it lasts for 1 min.

The times are just approximations of what you posted but it looks like a trend. I say it's a heat issue somewhere, a failing component that gets worse as it gets hot, or a bad PSU. It is heat or power related for sure. And, it still could be the MCH on the MOBO getting hot then failing more quickly with each test round.

Just to eliminate it as a suspect, do you have a spare video card to test with? Also, do you have a spare PSU? If so, try running with the other PSU.
 
I popped in the E6300, and it's running at 1.86 GHz. That's faster than my 1.60 GHz tests with the X6800. So if this works fine, it's a clue that the CPU is on the fritz.

Yellowbeard, I think I agree with your analysis. It definitely lasted longer after I left it powered off for a long time, then a much shorter time after a simple power cycle with no cool-down.

I do have an extra video card I can test with. I notice that the video card's heat sink is right next to the Northbridge chip fan. Although that was the case during the 4 months that it worked just fine at 3.65 GHz.

 
Originally posted by: InterMurph
Acanthus, I was at 1.425V with my stable configuration. What is electromigration? It doesn't sound reversible.
Click to expand...

It would not be, however, thats a safe voltage.

That isnt whats wrong.

Its caused by shoving too much voltage into any cpu, the transistors themselves are actually damaged slowly over time. (they actually slowly degrade anyway, but unsafe voltages greatly accelerate the process)
 
First freeze-up after 13 minutes.

Power cycle, and second freeze-up at 2 minutes. Just like with the X6800. So it doesn't look like the CPU.

Anybody want to buy a slightly-used E6300?
 
During testing, put the biggest fan you have blowing across the MOBO so it will pass cool air over the NB and the video card. Also, I forgot earlier, when these freezes occur in the OS, did you look at the event viewer? It's probably no use since you are seeing the same type of issue in DOS with Memtest.

Were you ever able to test another PSU?
 
I haven't yet tested my older power supply; that's next after the video card. When I swapped video cards, I also moved it to the other PCI Express slot, which is farther away from the Northbridge chip.
 
I didn't replace the video card; I just moved it to the other PCI-Express slot, since it was in contact with the Northbridge fan.

30 minutes on Memtest86+ and no errors; my best run in a long time. So I'm upping the speed to 2.93 GHz, restarting the test, and going to bed.

If it's the motherboard, I wouldn't be too said; that would only cost me $200 to replace. I'm just glad that it doesn't appear to be the X6800, which would cost me $1,000...
 
I think I'm on to something. It ran at the default speeds for FSB (266 MHz), CPU (2.93 GHz) and memory for over 10 hours.

Now I'm going to start up Windows and run some real code.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top