Cpu voltages and cpu life

[z0de]

I know the "safe" ranges for overclocking but im still unsure of by how much im shorting the life of my CPU as i would like it to last at least a few years. im currently

at

10x250=2.5ghz @1.4x110%=1.54v

stock is

11x200=2.2ghz @1.4v

 

how much is this reducing the life and whats the stock life?( i couldnt find it on the amd website)

also i want to overclock to 2.7ghz but from looking around it looks like ill need 1.6v+ will this reduce the life alot? a table would be nice

 

assume temps will be below 50 C

currently at 35-43C under heavy load for a long time.

[Torch]

Depends on Cooling.

I think nobody kwows how long a Cpu will work it Rip's.

My XP-M3000+ Socket 754 runs @1,82v almost 18 month 24/7 without a glitch.Aircooled

[noahsark1126]

These CPUs are too new to really know how much voltage shortens lifespan. I would suggest staying below 1.6V. The extra OC you will get from going over that won't be worth it.

[aspevacek]

with my Sandiego 3700 I can achieve 2.88 @ 1.55 volts (that is 1.375 + 13%) thry that and see what you chip will do. the Idle temp is 34 with a load temp righ around 40 deg cel.

[supershanks]

Agree with noahsark1126 re 1.6v looking round some of the forums u see some ridiculous figures.

 

I think it's a very valid thought, but is difficult 2 quantify.

[n_w95482]

with my Sandiego 3700 I can achieve 2.88 @ 1.55 volts (that is 1.375 + 13%) thry that and see what you chip will do. the Idle temp is 34 with a load temp righ around 40 deg cel.

What would be a better way of hitting that voltage, 1.375+13% or 1.55+0%? I'm currently running mine at 1.55+4% but I might try lowering the normal voltage part and raising the percent, if it makes a difference in overclocking.

[z0de]

thanks for the replys although they didn't put my mind at rest thou. i would also like to know about the high % or high vcore, but im also on the 704-2bta bios (for ram reasons) so i can only go to 113%. ill give 2.88 ago when im at home again.

[Thracian]

All this talk about cpu voltage shortening a cpu's life is getting a bit rediculous.

 

I have 2 socket A systems both with a vcore of 2v running folding@home 24/7 for over 1 1/2 years.

 

It's more a combination of cpu temps and vcore. Voltage increases temps..the hotter the cpu the less stable at a givin frequency. Running your cpu hot while pumping it full of voltage will shorten it's life, how long? Who knows. But I see ALOT of new people, new to overclocking, new to DFI scared as hell about vcore.

 

Worry more about the overclock, as by the time your cpu gives out, it will be a distant memory in the computer world....

 

Cheers,

Thrac

[z0de]

But in my world it will be a very recent painful memory. and if a 300mhz extra overclock will cut the life in half id like to know about it.

[Guest scott]

All this talk about cpu voltage shortening a cpu's life is getting a bit rediculous.

 

I have 2 socket A systems both with a vcore of 2v running folding@home 24/7 for over 1 1/2 years.

 

It's more a combination of cpu temps and vcore. Voltage increases temps..the hotter the cpu the less stable at a givin frequency. Running your cpu hot while pumping it full of voltage will shorten it's life, how long? Who knows. But I see ALOT of new people, new to overclocking, new to DFI scared as hell about vcore.

 

Worry more about the overclock, as by the time your cpu gives out, it will be a distant memory in the computer world....

 

Cheers,

Thrac

 

Well I don't understand it either, as this is my 1st OC learning experience.

 

Will it kill my cpu to run all the time at my highest Prime95 stable (25.5hrs) of 2570 MHz & 45 deg C?

To do that I have to feed it (vcore=1.475 + vid special=110%) = 1.6225 volts (actual draw is 1.584 v).

 

AMD says the max volts for my cpu is only 1.400 v.

 

If I'm killing the cpu, I'd like to be told that so I would then de-tune it down to it's other rock-stable point at 2500 MHz & 38 deg C, with vcore=1.375 +vid special=110 totaling 1.5125 v (actual draw is 1.456 v, pretty close to AMD's recommendation).

 

What do you think?

[Thasp]

It doesn't matter what we think. We can't tell you, because we don't know, because 99% of us have NO IDEA how these cpus are engineered, what they can handle, etc. 1.6v could mean a 7 yr life, or a 2 yr life, who knows?

 

If you are that nervous over a higher vcore, then don't bother with it. It seems the higher speeds aren't worth the paranoia over the future loss of your processor.

[EdLSmith]

Assume stock voltages and stock CPU speed.

If normal load produces 32C therefore we expect 3 year life at that level.

The rule of thumb for guessing the effects of heat on CPU life is life is reduce by 1/2

for ea additional 10C. Then a good guess if running at 42C would be 1.5 years life.

Now look around and tell me how many CPUs you have running at 42C and how long they have been running. I have one that I have been playing with for 5 months but it has run at 68C, 55C and now running at 37C. Lets see it ran at 68C one evening and at 55C a couple of days; how long will it last? Damn, I'm stumped. I guess I'll just have to worry about it. But How long. Untill the FX-60 is released and then I will have to worry about that.

 

Whoot. IMO