Originally Posted by billy66bare
OCCT pushes your CPU to determine your max temperature and stability by throwing a lot of code at it at once. If it's crashing, you're are probably unstable. The program itself *is* causing the crash, it's telling your overclock is probably unstable. Or, that there is the possibility that you could end up with corrupt data if your room where to warm up a few degrees, or you took on a particularly demanding task. Phenom II's like lower temperatures over higher voltage. So you could be stable all day long if you kept it below 45C, but once your temperature starts to get higher, your stability starts to suffer. If your room got a little warmer, or if you played or left a large video encoding, your overclock could become unstable due to the heat build up in the room. If you've run OCCT to determine your absolute max temperature, and you know you're stable at that temperature, then you have nothing to worry about.
Now, 100% stability is almost impossible when overclocked and is a hot debate. Personally, I've never had a crash or corruption from an overclock that passed 1 hour of OCCT on Large Data sets. That's a personal preference from a few years of experience. And to avoid a 4 page "100% stability" war, I'll leave it at that.
Does that clear it up a little for you?
Hi, thanks for your reply, in this case it needed more voltage and additional heat was no problem. I upped the vcore to around 1.475 and 3600 passed OCCT. Thanks for your input on OCCT! max temp during test was 64C, a bit warm. This chip is sick for what it cost me. Lets see "cough" Intel do this for $75.00 US
- Clock Speed 3600.11
- RAM Speed 1333
- Vcore 1.48
- HT Link 2000
- NB Speed 2000
- Motherboard ASRock Fatality990FX Pro
- Chipset 990FX
- CPU Cooling AC Freezer64 Pro
- Number of Core's 4
- CPUZ"]Cpuz Validation
Here are the goods
Edited by pshootr - 10/18/11 at 3:15pm