fcdwn777

Distinguished
Sep 4, 2010
245
0
18,680
Hello. I am currently running a phenom ii x4 stable @ 3.9mhz. I am pretty new to OCing and thus far have done so by increasing the cpu multiplier by intervals of .5 to test stability. When I go to 4.1 I get that "windows is shutting down to prevent damage"
screen after I try logging into windows, and at 4.0mhz I got the same screen after playing a game for a bit, I'm assuming this is because my cpu temp hit 50c and it is a safety thing.. My question is two fold. First, is there a way to remove or increase the temperature at which windows will self shut down? Even if there is, should I do it? From what I understand, 50c is still ok for a cpu to run at, albeit a little warm. Second, I know I can change the voltages on my cpu to overlcock, but I don't know how this works. Might I be able to achieve a stable OC higher than 3.9 if I were to bump up my voltage some? Thanks for your time.
 

kelthic

Honorable
Feb 28, 2012
1,078
0
11,460
increasing voltages generally increases temperature, so doing so would likely defeat your purpose. There is no way i know of to increase the threshold of the safety mechanism that makes the system shut down when a certain temperature is reached. The thing you should consider here is your cooling. Providing better cooling to your cpu might allow you to go higher in your overclock. What CPU cooler are you using?
 

Z1NONLY

Distinguished
Computer shutting itself down is a big red flag.

I wouldn't push your CPU any harder until you check some things first.

Download prime95 and HWmonitor.

Run prime95 blend test and HWmonitor for at least 30 minutes. (Starts to get the cpu its hottest at about the 20min mark)

The core temps shown on HWmonitor are sometimes 10 degrees lower than the real temps, so look for the highest temp on the list shown. (other than the video card temp)

TMPIN2 is the CPU temp on my motherboard but your may be different.

If it goes above 55C, shut down the test and lower your Vcore or get a better HSF.
 

fcdwn777

Distinguished
Sep 4, 2010
245
0
18,680


I'm using a hyper 212+ for cooling. So it's on air.

Computer shutting itself down is a big red flag.

I wouldn't push your CPU any harder until you check some things first.

Download prime95 and HWmonitor.

Run prime95 blend test and HWmonitor for at least 30 minutes. (Starts to get the cpu its hottest at about the 20min mark)

The core temps shown on HWmonitor are sometimes 10 degrees lower than the real temps, so look for the highest temp on the list shown. (other than the video card temp)

TMPIN2 is the CPU temp on my motherboard but your may be different.

If it goes above 55C, shut down the test and lower your Vcore or get a better HSF.

I don't run those unstable temps anymore. I have tested 3.9 in the past and it is fine. Also, I don't know what HSF is.
 

kelthic

Honorable
Feb 28, 2012
1,078
0
11,460
HSF is the heat sink fan, or your hyper 212+ Could be you've hit the safe limit of what your specific chip can handle. You could add a second fan for push/pull and get about 2-3 cooler temps but for that little of a gain, why bother? Could change out your cooler for something that cools more, but again, that's a preference thing.
 

Z1NONLY

Distinguished


Hyper 212+ is a good air cooler so I would experiment with lower voltages.

If your computer was shutting down from heat, it's possible that your CPU was getting more voltage than it needed to remain stable.

I'm running a negative offset of -0.075 volts.

I forgot to mention that you should also download CPUZ to and run it while you are stress testing your rig. It will show you the voltage you are running in real-time.
 

TRENDING THREADS