Relations between cooling and the cpu's lifespan

[papermate]

if i use my cpu's stock settings (eg 1GHz) and it lasts 5 years at an average of 50 degrees, would the lifespan be the same if i overclocked my cpu(eg 2GHz), but adding a cooling solution which retains the cpu's average 50 degrees?

 

[Grimmy]

Talking about lifespans on CPU is kinda ridiculous don't you think?

When people talk about lifespans, to me, it may generally mean being able to keep up with technology needs. Like new instruction sets to get better performance in games, servers, office apps that require number crunching to where time = money.

As far as OC'ing system, as long as you keep the voltage specs right, have good cooling, give it good maintenance (keeping dust free) that system should even out live a dog since we are talking 5 years of an average in actual operation.

 

[Crashman]

if i use my cpu's stock settings (eg 1GHz) and it lasts 5 years at an average of 50 degrees, would the lifespan be the same if i overclocked my cpu(eg 2GHz), but adding a cooling solution which retains the cpu's average 50 degrees?

no, every clock cycles has the potential to erode pathways, and increasing voltage (often needed to stabilize an overclock) increases the likelyhood that any given clock cycle will erode any given pathways by some tiny amount.

I've never had a CPU die from overclocking. Then again, I've seen non-overclocked CPU's last 15 years.

 

[Reynod]

There was a paper put out in the late 70's regarding temperature and lifespan ... I can't recall it properly but there was considerable emphasis on electromigration.

Essentially the less voltage you throw at the cpu and the cooler you can keep it ... the longer it sould last.

If I can find it I will post it.

Mind you it's aimed at engineers ... not Joe Public.

I therefore struggled ...

I have a 3000+ XP chip running at 2.4 (instead of 2166) ... since I bought it ... years ago. Overvolted too (1.65 to 1.75 or so). Back then that was a nice jump.

Now you can get a E / Q6600 (9 X 266) and stick it to (9 X 333) without an overvolt if your lucky ... 600Mhz for nada (2.4 to 3 Ghz).

Mind you I killed a few Durons and Celerons along the way ... and more expensive chips ... and cried a few times.

Keeping your rig cool by tidying up the innards and having good fans and a decent cooler are the basics ...

 

[jj463rd]

I have quite a few systems 25 to 30 years old that still work.

 

[papermate]

umm, what do you guys mean by the voltage of the cpu? do you mean the voltage from the psu or the mobo?

 

[ejay]

There was a paper put out in the late 70's regarding temperature and lifespan ... I can't recall it properly but there was considerable emphasis on electromigration.

Essentially the less voltage you throw at the cpu and the cooler you can keep it ... the longer it sould last.

If I can find it I will post it.

Mind you it's aimed at engineers ... not Joe Public.

I therefore struggled ...

I have a 3000+ XP chip running at 2.4 (instead of 2166) ... since I bought it ... years ago. Overvolted too (1.65 to 1.75 or so). Back then that was a nice jump.

Now you can get a E / Q6600 (9 X 266) and stick it to (9 X 333) without an overvolt if your lucky ... 600Mhz for nada (2.4 to 3 Ghz).

Mind you I killed a few Durons and Celerons along the way ... and more expensive chips ... and cried a few times.

Keeping your rig cool by tidying up the innards and having good fans and a decent cooler are the basics ...

Here you go:

http://en.wikipedia.org/wiki/Electromigration

http://en.wikipedia.org/wiki/Pentium_4#Northwood

"Overclocking early stepping Northwood cores yielded a startling phenomenon. When core voltage (Vcore) was increased past 1.7 V, the processor would slowly become more unstable over time, before dying and becoming totally unusable. This became known as Sudden Northwood Death Syndrome, which is caused by electromigration.[12]"

 

[sarwar_r87]

I dont mean to steal the thread, but i actually had the same question (well sort of)and i was actually planing to ask this same question b4 i decide what cpu to buy.......

this is the question i wanted to ask......
i dont like OCing but
- whats the probability, for example, if i buy 100 q6600 how many could hit 3Ghz without overvolting.....and if i have to overvolt, by how much on average.
-if i keep the stock cooler, what will my temp look like?
-will this OC have a huge effect on its life.coz i seen lot of q6600 clocked higher and no1 really complains bout life span

i have done little OCing, but the idea of Increasing volt really creeps me out. also high temp.
so i was wondering if i could do q6600@3GHZ on stock volt and cooler.......i ask about stock cooler coz in my country, only one importer imports aftermarket cooler, and the charge a lot for them. and also i dont think i will get a option of choosing as they are not likely to have more than two or three models of cooler

thnku

 

[abasoufiane]

yes you can overclock your q6600 WITHOUT touching the voltage to 3ghz, i'm talking about the g0 stepping and it's almost guaranteed. don't use the stock cooler though.

 

[eklipz330]

the voltage of the cpu is the voltage needed to keep the processor running stable, any power comes from the psu, not the motherboard

 

[speedbird]

this is the question i wanted to ask......
i dont like OCing but
- whats the probability, for example, if i buy 100 q6600 how many could hit 3Ghz without overvolting.....and if i have to overvolt, by how much on average.
-if i keep the stock cooler, what will my temp look like?
-will this OC have a huge effect on its life.coz i seen lot of q6600 clocked higher and no1 really complains bout life span

Hi,

No two CPU's even of the same model are guaranteed to overclock the same. It's likely that you can hit at least 3Ghz on a Q6600, but you may need to bump the voltage up slightly. My Q6600 was not stable at 3Ghz with the stock Voltage, so I upped the voltage one notch on my MSI motherboad BIOS. The supplied Intel stock cooler is, in one work terrible. My temps were hitting in the seventies with the crappy Intel cooler at overclock speeds, but an Aftermarket cooler soon put those temperatures down to more acceptable levels. I suggest you buy a better cooler, if you have any overclocking intentions.

As for Life span there's no accurate answer for that, but if the Temps go above the manufacturers thermals limits, then this can be bad for life span.

 

[Crashman]

the voltage of the cpu is the voltage needed to keep the processor running stable, any power comes from the psu, not the motherboard

The power comes from the motherboard Seriously though, even though the power supply provides the 12V, the motherboard's voltage regulator makes it 1-2 volts.

 

[bobbknight]

Heat and electronics life span are an inverse curve.

 

[Crashman]

Heat and electronics life span are an inverse curve.

That's true. But there are other factors that affect electromigration, such as frequency and voltage. If you up the voltage but maintain the same temperature, the added force will still cause the pathways to errode faster. And transistors still have a finite number of times they can switch, just as electromagnetic relays, so increasing the clock frequency does have some affect even if heat and voltage remain constant.

So heat's the biggest killer, but it's not the only one.

 

[Merman]

Too many people believe overclocking is "safe" as long as they don't increase their processor core voltage - not true. Frequency increases drive higher load temperatures, which reduces useful life. Conversely, better cooling may be a sound investment for those that are looking for longer, unfailing operation as this should provide more positive margin for an extended period of time.

http://www.anandtech.com/showdoc.aspx?i=3251&p=6

 

[Grimmy]

You forgot to add the paragraph under that graph:

The graph above shows three curves. The middle line models the minimum required voltage needed for a processor to continuously run at 100% load for the period shown along the x-axis. During this time, the processor is subjected to its specified maximum core voltage and is never overclocked. Additionally, all of the worst-case considerations come together and our E8500 operates at its absolute maximum sustained Tcase temperature of 72.4ºC. Three years later, we would expect the CPU to have "degraded" to the point where slightly more core voltage is needed for stable operation - as shown above, a little less than 1.15V, up from 1.125V.

Now doing test to make sure you have a stable clock is not going to even come close in keeping even a stock CPU at 100% load for 3 year straight. Also its talking about the max tcase temp, or thermal temp and vcore.

Based on that, OC a CPU, using speed step which reduces the OC (x6 at rest), and even being at idle reduce the heat and voltage. That graph would not apply for normal usage. That graph would only apply for an insane way of degrading the silicon, even at stock speed, running it at 100% usage on a year 3 scale.

So how many of you guys put your CPU at 100% on a day to day basis?

 

[coret]

So how many of you guys put your CPU at 100% on a day to day basis?

Or more simply, how many people here fold?

 

[Grimmy]

I do... but not in the summer.

Edit:

And in the winter, my load temp will not be even close to the max thermal spec for my CPU.

Edit:

I just brought up 4 different WU on folding. So far my max temp is 51C tcase (ambient 80F or 30C), and my max vcore is 1.34, and the VID is 1.312.

The max spec's for the Q6600 - 71C max thermal spec, and vcore 1.50v. So the graph wouldn't even apply to my quad OC to 3.2ghz, and that its 65nm rather then 45nm.

So with the E8500, you would have to really watch what your doing in OC'ing, since its max vcore is 1.362 with a max thermal spec of 72C, in trying to reach 4ghz.

 

[rockstone1]

if i use my cpu's stock settings (eg 1GHz) and it lasts 5 years at an average of 50 degrees, would the lifespan be the same if i overclocked my cpu(eg 2GHz), but adding a cooling solution which retains the cpu's average 50 degrees?

No, because a one ghz PC would have a lifespan much longer than 5 years. I have a Pentium 3 in my house that still works after 9 years (and I only had to open it once or so!), and I have one whose age I don't know (got it at a flea market because I felt like it), that is also Pentium 3 that works.

Anyway, such a CPU would last a long time, but doubling your CPU power and maintaining stability like that would be almost impossible. Assuming you could, it would not last very long due to the stress of the CPU attempting to due do things that much faster. Also, if you didn't use a CPU multiplier and instead used the FSB then your system would be even MORE unstable then before, because now, everything is faster then it is rated to be.

 

[goonting]

Nice chart....cooler cpu is beter than the hot ones....even my old shuttle has a radiator like cooler keeping things at 53-55Celsius at full load

 

[warezme]

so all this good info tells us, (in my best Italian accent) Fugedaboudit! and enjoy your CPU for however long you have it, at regular speed or OC'ed because if its gonna die it was going to die anyways and you most likely didn't kill it unless you overvolted it to an insane level.

 

[sportsfanboy]

Who cares, in two to three years after a processor is released, it's value is cut in 1/4, maybe more. I say crank it up, and if the little prick dies in two three years, upgrade or buy a new one on the cheap.