overclocking and CPU life?

[fallstoofast]

My friend told my that overclocking a CPU will decrease its life by about 3 months (of course I think it'll really depend on how much you overclock it)

What's the initial life? If it's reasonably long like 3 years, I could think of doing it.

If there are any other pro's or con's anyone would like to share, I'd be grateful

thanks,

 

[CTho9305]

Overclocking and overvolting both lower the average lifespan of CPUs. There are lots of numbers floating around, but they're all based on miniscule sample sizes and anecdotes. Only the CPU manufacturers (and possibly large OEMs) have the real information (because they conduct experiments on large numbers of CPUs), but they don't generally make that information available. One of the large problems with end-user reports is that end users don't experience accelerated wear-out for many months or even a few years; by that time, manufacturing processes may have changed enough that the end-user experience no longer applies (e.g. if CPUs historically failed due to electromigration, but start failing more due to another effect, OCers will have drawn conclusions that can't be applied to the current CPUs).

 

[OCGuy]

google "The truth about CPU degradation". Interesting read.....many factors involved.

 

[boomhower]

CPU's will typically function way past there useful liefspan at stock speeds, i.e. they will be to slow for day to day useage before they will die. Overclocking can decrease the lifespan due to added heat and voltage. As long as you are properly cooled and use reasonable voltage you will replace it long before it will die. Since you are one this site you will likely change out your hardware much more often than a joe sixpack.

This is just assuming averages. Obviously some hardware will die sooner than others.

 

[aigomorla]

okey on stock settings intel guarentee's 3 yrs. Per say there warrenty period.

So you can assume the rock bottom lifeline on your cpu will be 3yrs @ stock settings, however i gaurentee you its more likely to be more.

Now heat, voltage, and overclock play a hugh deal on cpu life. Without knowing those settings it would be impossible to tell you.

Also intel has never full released a fault sheet on the processor, so we dont even know how long they really last on stock settings. All we know is intel gaurentee's them to run at stock settings for 3 yrs.

So no, i dont care who tells you what, unless he was on the direct board for the C2D design board, no one can tell you how long your chip will last.

Even then, without knowing those 3 variables, even that intel specalist wouldnt be able to tell you.

So any answer you get, is false.

And your question has no real answer.

 

[VirtualLarry]

I'm willing to estimate that my E2140s will last at least 3 years, even running at 1.425v (BIOS) and at 9C to TJMax. Just my guess, but I've never had a CPU "degrade". I don't push the voltage either.

 

[aigomorla]

Originally posted by: VirtualLarry
I'm willing to estimate that my E2140s will last at least 3 years, even running at 1.425v (BIOS) and at 9C to TJMax. Just my guess, but I've never had a CPU "degrade". I don't push the voltage either.

and it probably will. As i said, no one will find out until that day comes.

By then i highly doubt you'll still be using that processor.

 

[BTRY B 529th FA BN]

Originally posted by: fallstoofast
My friend told my that overclocking a CPU will decrease its life by about 3 months (of course I think it'll really depend on how much you overclock it)

What's the initial life? If it's reasonably long like 3 years, I could think of doing it.

If there are any other pro's or con's anyone would like to share, I'd be grateful

thanks,

my P4 Northwood 478 3.06 HT 533 Rambus cpu still works, i bought it over 4yrs ago and its been overclocked the whole time

 

[DSF]

Your friend pulled a number out of his butt.

As others have said, it's impossible to place any firm number on these sorts of things, but unless you're doing so-called "suicide overclocking" you can expect your chip to last as long as it will be useful.

 

[ShawnD1]

Personal experience:
-stock Athlon 2200+ lasted 6 years before it died
-Athlon 2500+ overclocked by 20% died after 3 years

It has a lot to do with heat. That overclocked 2500+ was running at a constant 60C because of Folding. I bet Folding is responsible for quite a few of the deaths in my house.

Anandtech article about CPU life

 

[myocardia]

Originally posted by: VirtualLarry
I'm willing to estimate that my E2140s will last at least 3 years, even running at 1.425v (BIOS) and at 9C to TJMax. Just my guess, but I've never had a CPU "degrade". I don't push the voltage either.

I'm willing to bet you're totally wrong about that. I can assure you those chips are throttling, which we all know wears them out at an exaggerated pace. Duvie did a bit of research awhile back concerning throttling that might interest you.

 

[unr3al]

Very interesting read. I find it intreaguing as I am an avid overclocker myself and wondered about the topic. I have had my CPU (Brisbane A64 X2 4000 2.1GHz) since 09/2007 and have been overclocking and undervolting and various other things.

An interesting question would be: If you run your CPU at stock frequencies, but lowered the CPU voltage substantially, would that extend its lifetime? I'm guessing it would.

I can run mine at 2.1GHz and 1V. 0.25V down from stock, but I prefer running it overclocked. Gets everything (video and audio editing/encoding/decoding etc) done so much faster lol

 

[fallstoofast]

I guess CPU lives are long enough to last till it becomes obsolete. I don't think I'll be using the same processor if a Hex Core 5.0GHz is out at a few hundred bucks... if that happens within the next few years

 

[StinkyPinky]

I have a P3-650 overclocked by about 20%. It's been like that since I got it....in I guess 2001 or so.

 

[Serradifalco]

Intel and AMD know how hard we push these things. I think these chips are designed to last in extreme conditions. I had my e4300 oc'd to 3Ghz for one year with no issues. My average use for a chip is about 1 1/2 to 2 years. This last upgrade was quicker than normal because of my P35 board. I bought it when the P35 boards first came out so it still supports the latest cpus, and very well if I may say so!

 

[unr3al]

We would hear a lot more complaints online about prematurely dying chips if they were going to pop like crazy under a little overclocking. Who was it that spoke of suicidal overclocking? Well, that might get it to pop. Hence the appropriate name

 

[aigomorla]

Originally posted by: unr3al
We would hear a lot more complaints online about prematurely dying chips if they were going to pop like crazy under a little overclocking. Who was it that spoke of suicidal overclocking? Well, that might get it to pop. Hence the appropriate name

no im sorry.

today people think 50% overclock is given so they push massive voltage.

sure 1.5V wont kill your chip right away, but lets put this in a person perspective.

Someone is forcing to you sprint all the time taking deeper and heavier breathes.

i thikn you'd colapse sooner then the guy whose walking at a normal pace.

 

[killster1]

wow this thread is hilarious.. all the people that replied.. "i have had my cpu overclocked for 2 years and its fine!".. ive actually killed a cpu by overclocking it.. a preshott 2.6ghz oc'd to 3.4ghz for a year.. was in a box i was letting someone use and they had it running at insane temps.. (very dusty and hot room) my guess 70C+ i think i had it shut down at 75C.. well eventually it just wouldnt boot anymore at any speed.. i thought it was teh asus mobo as ive had soooo many of them die but it wouldnt post in a known working mobo either..

if you keep your temps under 55C LOAD your cpu should last 5 years+ i have 4 opteron165's that ive run overclocked since they came out, all of them run fine but eventually i had to bump the voltage to maintain the overclock on one of them (the only one of the bunch i had super undervolted but stable at 2.8ghz)


also consider that with your cpu overclocked and producing more heat everything else will have higher temps unless you duct the cpu air out.. hard drives hate heat everything else should be fine..

 

[killster1]

Originally posted by: aigomorla

Originally posted by: unr3al
We would hear a lot more complaints online about prematurely dying chips if they were going to pop like crazy under a little overclocking. Who was it that spoke of suicidal overclocking? Well, that might get it to pop. Hence the appropriate name

no im sorry.

today people think 50% overclock is given so they push massive voltage.

sure 1.5V wont kill your chip right away, but lets put this in a person perspective.

Someone is forcing to you sprint all the time taking deeper and heavier breathes.

i thikn you'd colapse sooner then the guy whose walking at a normal pace.

well lets think about that.. 50% overclock at less than 1.5v (shouldnt take 1.5 come on..)


so your person is running for 50% longer when the race is happening.. other guy is resting by the time you finish.. so does this really make the guy running for longer last longer!?


this really only applies to encoding etc.. as games will use the cpu no matter and just run slower instead of taking longer

 

[OCGuy]

today people think 50% overclock is given so they push massive voltage.

sure 1.5V wont kill your chip right away, but lets put this in a person perspective.

The only necessary 50% OCs are the inexpensive chips anyway. People with the E8400+ chips are averaging 3.8-4.1ghz, and around 1.4v max on air that I am seeing. That is very close to the intel spec of 1.3625 max, and only ~33% OC.

The benefits of being able to buy a E7200, and have it perform even better than a stock E8600 will in Q3, far outwiegh any decrease in lifespan. We are being extremely spoiled right now, and I can only hope it carries on to Nehalem and beyond.

And to be honest, I hope AMD gets thier crap together, so Intel cant pull any stunts, i.e. disabling overclocking, delaying future product, etc.