OC voltage and stability

[moonbogg]

Something cought my attention. My CPU @ 4.5 with 1.35v offset was mostly stable, but BF3 crashed once. So I upped the voltage one notch to 1.36v, and now nothing crashes and its Intel burn test stable for at least 10 runs as well? one little voltage bump like that can make a difference from unstable to basically totally stable? I haven't done a 12 day run of Prime or anything, but its pretty stable with such a small bump in voltage. I would think it would take a bit more voltage to get stable. What say you? NORMAL?

 

[Kenmitch]

When using offset vcore it's the single/dual core load that can sometimes cause instability. Upping the vcore slightly can remedy this issue. Kinda normal I guess you just need to find the fully stable vcore.

Using something like Intel Burn Test and testing with only 1 thread and 2 threads will help for long term stability testing.

 

[Idontcare]

What say you? NORMAL?

Absolutely normal. If you want something more meaty than opinion, check around the web for Shmoo Plots, you'll learn more about Voltage and stability than you ever wanted 😀

 

[cmdrdredd]

Yeah that's how it is for me. 1.256v is no go but 1.26v is ok.

 

[Ken g6]

Imagine a computation circuit on your CPU like a roller coaster. Under normal conditions it works fine, but if you don't give it enough energy (kinetic/voltage) to get over every single hill, it will get stuck someplace.

 

[moonbogg]

Imagine a computation circuit on your CPU like a roller coaster. Under normal conditions it works fine, but if you don't give it enough energy (kinetic/voltage) to get over every single hill, it will get stuck someplace.

I like that analogy. Thanks guys for replies. Appreciated indeed.