Peculiarity while Stress Testing Intel G3258

yoshijefferson

Commendable
May 20, 2016
1
0
1,510
Hello all,

I ran into a strange problem while stress testing my current configuration with the Intel G3258. Before I begin, I am quite new to PC building and very new to overclocking, so please forgive the limitations of my current understanding.

I have been trying to slowly increase the clock speed of my Pentium over the last couple of days. I am using the stock cooler and the voltage setting is currently set to 'auto'. I was able to achieve a stable 3.8 GHz yesterday, and decided to try for 4.0 GHz today. Despite the increase, the voltage only went up from 1.176 V to 1.177 V. Nevertheless, I decided to stress test it using CPU-Z and HWMonitor to measure temperatures. Now, at first everything seemed to be going great, even though I was suspicious that the voltage might not be sufficient. Basically, the stress test went on stably for about a half hour, with temperatures never exceeding 65C. From what I understand, that's pretty good. I left the computer for a bit and returned shortly thereafter. My monitor had gone into sleep mode and I shook my mouse to 'wake it up'. Everything came back up, and looked normal, and then a couple seconds later, I got a WHEA message that Windows had encountered and unknown error and had to shut down.

I found this curious because the temperatures seemed fine. I've been running another stress test as I write this, and it has been going on for about 20 minutes. Again, everything seems fine right now. Temps below 65, and task manager shows steady maximal CPU usage. Basically, what I'm wondering is the following:

1. Could the crash have been from something totally unrelated, given that my temps are fairly low and stable?
2. If the crash was from the CPU, could it be due to the 'auto' voltage setting on my BIOS not giving enough voltage to the CPU? And if so, what do you guys think I should set it at to stabilize the configuration?

Thanks a ton for any help in advance.
 
Solution
Well, the only thing I can think of is the voltage, I can't tell you for certain what it should be, just experiment with it for a little while. Make sure that your 4.0GHz is locked in and the system can't "turbo boost" it when it's under load. So you might want to go for the manual voltage control. 65C is very good for a stock cooler, I wouldn't of expected that.

Just go up in the voltage by .001 increments. You'd be surprised how far you can go with just going up .005.

LogicalProcessing

Honorable
May 22, 2014
266
0
10,960
Well, the only thing I can think of is the voltage, I can't tell you for certain what it should be, just experiment with it for a little while. Make sure that your 4.0GHz is locked in and the system can't "turbo boost" it when it's under load. So you might want to go for the manual voltage control. 65C is very good for a stock cooler, I wouldn't of expected that.

Just go up in the voltage by .001 increments. You'd be surprised how far you can go with just going up .005.
 
Solution