What actually kills a CPU when OCing?

Well bumped into a problem when incresing my vcore. Goes up in tiny increments (0.0125 I think) until it gets to 1.6V, then it just jumps up to 1.8!! Far from ideal but booted it up anyway, ran prime95 for a bit and doesnt get over 68 so its not too bad, plus when gaming the highest I saw was 53. Got it upto my target of 3.5GHz easily at the voltage and just left it there, might see how much further it will go later :p

Theres still the electromigration thing which might be reducing the life of it quite a bit, but its a £40 processer, its no big deal if it dies really I guess haha, excuse for an upgrade!
 
I wouldn't go above 1.5v/1.55v with air cooling. After that, it'll need watercooling. If you are on water, and are getting good temps, even then I wouldn't jump up to 1.8v, 1.6v should get it to quite a high speed.
 
I believe the biggest issue is heat. To much juice will cause too much heat anyway. However most chips now will have safety limits in the chip thanks to the motherboard.

I personally am not worried about overclocking a CPU and "killing" it. With fire.
 
Voltages (both the core voltage, and the fsb voltage) are the real killers, heat is covered more or less by intels thermal throttle, and thermal shutdown systems.

The fsb voltage can reduce the stability of high overclocks if pushed too far.

Current chips are pretty sturdy though, and it kinda depends how long you expect to keep the chip. Many overclockers change their processors so often that they never get close to their "expected lifetime". CPU's are pretty much designed to last 10+ years running 24/365 at full load with a modest cooler.

Overclocking may reduce that to 5years, or 1 year (or less if your a bit nuts).

Heat can cause instability and crashes, but at the end of the day its the volts that kill.
 
You must log in or register to reply here.
Back
Top Bottom