Overclock.net banner

Overclocking life span argument

614 Views 12 Replies 10 Participants Last post by  captthunderpnts
Suppose I have the two following systems

1) Q6600 running all at stock, running at 30deg. celcius

2) Q6600 overclocked to 2.8ghz running at the same voltage and all other settings as system 1, as well as running at 30deg. celcius

Will the lifespan of the 2nd CPU be lower than that of the 1st system simply due to increased CPU cycles?
1 - 13 of 13 Posts
I can't say one way or the other, but I can say I've never had a chip reach the end of its life before I needed an upgrade anyway overclocked or not. I think if the lifespan is affected it would still be long enough that it would be obsolete by the time it dies anyway.
There is really no way to tell, comparing only two CPUs.

The only way to have a valid experiment would be to have like 100 CPUs, half overclocked, running for 5 or 6 years straight, and calculating the mean time of failure.
Not really, high temps are usually the killer. Don't forget that these same chips are made and then rated at 2.4 GHz, 2.6 GHz with differing prices ranges since they passed some factory tests and rated at that level.

Even though prices may differ on basically the same chip, a chip is really different when the specs differ in cash capabilities, dyies, etc...
See less See more
Not reaching the life span is not the question. I know it is quite a hypothetical argument, but there has to be some reasonable scientific expectation as to which would die first.
Quote:


Originally Posted by TEntel
View Post

There is really no way to tell, comparing only two CPUs.

The only way to have a valid experiment would be to have like 100 CPUs, half overclocked, running for 5 or 6 years straight, and calculating the mean time of failure.


Personally, I think if you were running any cpu at full bore for that long they would fail long before, but it's impossible to find out since the PSU would probably fail before the cpus running under load for so long.

I wonder if any of these CPUs would run at 100% for even two years straight without failing...

Whats the longest period of time that anyone on here, left their computer fold for without turning it off? (granted thats not quite %100 percent usage)
See less See more
You'd expect the higher clocked CPU to not last as long, but at such a mild OC it's hard to tell. The lower clocked one could fail first. But we're talking about 2012 or later. You could be wiped off the map by a hurricane before either one fails.

And forget 2.8, just clock it to 3.0 and enjoy.
All CPU's have different lifespans, just like people.
Certain factors can affect the lifespan of a chip though

The main 3 being Voltage, Heat, and clockspeed
Clockspeed has a minimal effect on lifespan IIRC. It's voltage and heat that really affect the lifespan
As I said, this is not an argument about if the PSU would die first, if they were running at 100% or if a hurricane would come.

I am asking, ceterius paribus except for the difference in clock speed, which would be expected to fail first?
The higher clocked one due electron migration. It has to deal with more electrical pulses which will eventually wear out the connections.
  • Rep+
Reactions: 2
2
Quote:


Originally Posted by DuckieHo
View Post

The higher clocked one due electron migration. It has to deal with more electrical pulses which will eventually wear out the connections.

Thanks Duckie. I made the same argument to the person I was arguing this with.
See less See more
Quote:


Originally Posted by DuckieHo
View Post

The higher clocked one due electron migration. It has to deal with more electrical pulses which will eventually wear out the connections.

Ducky is absolutely correct on that.

However, the rate of electron migration will not increase too much as to noticeably shorten the lifespan of the chip, providing you are running stock or near stock voltages. With a mild overclock on the same volts, you could expect 5.75 years vs 6 years with the given chip. Of course, somebody else brought up the wonderful point that comparing only two chips isn't really a viable means of measure, due to minute imperfections in the manufacturing process.
See less See more
1 - 13 of 13 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top