[micah_jones]

Ok so basically, my e5200 is at around 1.38 volts right now in bios. Intel says 1.365 is the limit(I think?) I just want to know, how high can the voltage go before I really start sacrificing lifespan? I wanted to go for 4ghz, but I dont want it to only last a week at that. Should I just go by temps and make sure I dont go over, say 65C under full load? Or can volts still kill the CPU even though it is not getting very hot? I hope this makes sense, thanks guys.

 

[mothergoose729]

Voltage will kill the CPU over time. However, you have to volt pretty high to kill it very quickly. What usually happens is the CPU slowly degrades until it cannot perform at the speeds it used to. The thing I find it at a certain point, raising the voltage does little to help you get a higher overclock. You might have to add .05 v to get 100 more mhz of speed, and in real world apps that doesn't mean much. I would keep it at 3.8ghz, because you will probably need at least .04 more volts to get it stable at 4.0ghz.

 

[t_russell]

1.4v is the widely considered safe limit for 45nm CPU's.

 

[grossebeaver]

Quote:

Originally Posted by t_russell 1.5v is the widely considered safe limit for 65nm CPU's.

The E5200 is a 45nm CPU, so 1.5v is a bit too much. The max most people are applying to 45nm's is around 1.4-1.45v, but it really depends on how conservative you are. I've OC'ed a couple E5200s and really didn't see a huge improvement in stability over 1.38v-1.4v.

 

[NuclearCrap]

Quote:

Originally Posted by t_russell 1.5v is the widely considered safe limit for 65nm CPU's.

E5200 is 45nm.

And for 24/7 usage, I wouldn't go much over what Intel recommends. 1.38V would be fine if they say 1.36V.

 

[t_russell]

My mistake. Edited original post.

 

[micah_jones]

Ok guys, thanks a lot for all your help. I think I would need about 1.43-1.44 volts to keep it stable at 4.0ghz. It really isnt about the performance increase, its about how much cooler 4.0 sounds than 3.8

I will just stay at 3.8 for now.