OC vs Lifespan

[m25]

I have overclocked my new 3000+ (venice) @ 2150MHz to gain a good 20% in render times. Does anybody know (better by experience) how this affects the lifespan of the PC in terms of years (I often load it with hours of renderings)?

 

[Pain]

The answer is: There is no answer. You can guess. 37% reduction if usuable lifespan. That is my guess.

 

[m25]

The answer is: There is no answer. You can guess. 37% reduction if usuable lifespan. That is my guess.

8O 37%... doesen't look scientifically proven but makes sense. That's 6 years if 100% is 10, guess I won't be at it's sixth birthday, though somewhere I stil have a celeron 333 running from 1999

 

[mad_fitzy]

How long do you plan on using that cpu. Ive been running my cpu at 400mhz over stock at 1.8 volts for 3 years at 100% load and hasnt failed on me yet.

 

[kamel5547]

The answer is: There is no answer. You can guess. 37% reduction if usuable lifespan. That is my guess.

8O 37%... doesen't look scientifically proven but makes sense. That's 6 years if 100% is 10, guess I won't be at it's sixth birthday, though somewhere I stil have a celeron 333 running from 1999

We still have P1 133's running in a close to 24/7/365 environment. This is in a soils lab so the dust factor is huge (and they've never been cleaned out).

 

[Half_h_omega]

I'd say that thermal cycles are your biggest worry. Over clocking will decreases the life span because the CPU will get hotter, therefore thermal fluctuations will be greater but if it remains at a certain temp all the time your OK. It’s a well know fact that if you leave comp on all the time it will last longer (hard drives/fans are the exception those bearings are going to wear out eventually). I know it sounds strange but a CPU that’s loaded all the time will probably last longer than one that gets intermittent loads.

edit: for dumb ass spelling error

 

[ivoryjohn]

Actually, the answer is: 42

A mild overclock, especially at stock voltages won't affect lifespan if temperatures are normal.

The truth is that the die in your CPU was designed to run near the highest speed they sell, then they clock a few back by locking the multipliers, and they clock a few up that pass rigourous testing.

If you take a cpu and clock it up to the fastest in it's family, then you are not affecting its lifespan any more than you would buying the more expensive one to begin with.

And unless you are extreme and blow it up, it's lifespan will likely outlive it's usefulness anyway. (It will be obsolete, slow and incapable of running Windows xxx before it wears out).

 

[m25]

Actually, the answer is: 42

A mild overclock, especially at stock voltages won't affect lifespan if temperatures are normal.

The truth is that the die in your CPU was designed to run near the highest speed they sell, then they clock a few back by locking the multipliers, and they clock a few up that pass rigourous testing.

If you take a cpu and clock it up to the fastest in it's family, then you are not affecting its lifespan any more than you would buying the more expensive one to begin with.

And unless you are extreme and blow it up, it's lifespan will likely outlive it's usefulness anyway. (It will be obsolete, slow and incapable of running Windows xxx before it wears out).

Well, you all mention the CPU.. what about the chipset?
I noticed that it got really hot, even though the motherboard temp was just 34°C. (However, I couldnt make out by touchif it was cooler @ normal FSB200)

 

[m25]

The answer is: There is no answer. You can guess. 37% reduction if usuable lifespan. That is my guess.

8O 37%... doesen't look scientifically proven but makes sense. That's 6 years if 100% is 10, guess I won't be at it's sixth birthday, though somewhere I stil have a celeron 333 running from 1999

We still have P1 133's running in a close to 24/7/365 environment. This is in a soils lab so the dust factor is huge (and they've never been cleaned out).

while the celeron is still used for CAD and works great in 2/3D

 

[clue69less]

The answer is: There is no answer. You can guess. 37% reduction if usuable lifespan. That is my guess.

That's right, no answer. Too many other variables. Will you overload your power supply and have it cook your mobo and CPU? Will lightning get past your isolation? Will a cat pee on your HSF? Will your HSF fan die? Even if you had excellent control of all possible variables, there would be a reasonably large lifetime variation expected for an OCed CPU.

 

[clue69less]

Temperature? Maybe

I still believe it all depends on the voltage used.

I took part in an accelerated aging study on multilayer builds. We found that at the extremes, either temperature or voltage excess would lead to premature death. For sure, migration across layer interfaces is accelerated at higher voltages. Migration is uncool because it changes the work function of the substance. Cool the chip enough and migration is slowed. Heat it enough and migration accelerates even at stock voltage. Hold the temp steady and vary the voltage and migration accelerates above a threshold voltage.