Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

From a business standpoint...

If intel could take a 3.0 GHz CPU, and overclock it to 4.0 GHz - whilst still covering their warranty they would do so. It's absolutely free and makes the chip perceived to be "more powerful" by the general public and could then command more money.

Corporations are always trying to squeeze out an extra buck. The sheer fact that they don't overclock their products from the factory goes to show it must be a bad thing economically. Does it not?

share|improve this question
    
a 4ghz chip from intel isn't an overclocked 3ghz chip, it's a 4ghz chip, period. –  RobM Mar 23 '11 at 10:24
    
It's = it is. You want its. –  houbysoft Mar 23 '11 at 11:10
1  
It shortens your warranty. Instantly. –  Tom O'Connor Mar 23 '11 at 11:45
add comment

closed as off topic by John Gardeniers, SvW, RobM, Chopper3 Mar 23 '11 at 14:43

Questions on Server Fault are expected to relate to professional server, networking, or related infrastructure administration within the scope defined by the community. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about reopening questions here.If this question can be reworded to fit the rules in the help center, please edit the question.

5 Answers

up vote 4 down vote accepted

Intel can't overclock their chips, by definition. Overclocking is when you increase the speed above what it is set to from the factory. Intel is the factory. Whatever speed they set a chip to run at, becomes the official speed that overclocking is relative to.

They simply produce a batch of chips, and then they test them to see if they can run safely at the intended speed. If they produce chips that can work reliably at 4GHz, then they sell them as 4GHz chips. if they produce some that can only work at 2.8GHz, then they sell them at that speed.

A CPU doesn't have some kind of "natural" speed. They design the chip with a clear expectation of how fast it should be able to run, then they manufacture it and hope for the best, and finally, after testing, they know what speed it is actually able to run at, and then they configure it to use that speed by default.

So depending on your perspective, Intel either always overclock their chips, or they never do. But they can't overclock more than they do already.

share|improve this answer
1  
It should be noted that sometimes chips get actually labeled lower than they could run to satisfy the market demands and to close the gap between available chips and a hypothetical high demand for certain lower-spec chips. If you get one of those chips, it could be rather safely overclocked. Now the only remaining issue is to identify those :) –  Holger Just Mar 23 '11 at 10:47
    
@Holger: true, and I thought about including that in my answer, but in the end I decided it would complicate matters for little reason. My main point was that chips don't have an "inherent" speed. Instead they are configured by Intel based on what the chip is capable of in their testing, and on their business needs. –  jalf Mar 23 '11 at 11:27
add comment

Due to the vagaries of chip manufacturing, not all the parts produced are of the same quality.

So, the binning of parts is partly due to the fact that some of the slower clocked parts may have failed certain tests at a higher speed but passed it at a lower speed.

If you overclock some of these slower parts, there is no guarantee that it will work. Increased voltage and current can also cause parts to fail faster.

So, it's not a good idea.

share|improve this answer
add comment

If Intel were to overclock their products, sure, they would probably work during the warranty period. However, that's not all there is to it, since the brand is worth so much more. If Intel got a reputation for processors failing after 5 years, people would stop buying their processors -> bye, bye, Intel.

So, that shortsightedness might pay off for a couple of years, but it's not a long term winning strategy.

And yes, overclocking -> higher temperatures -> more stress -> shorter component life. That's simple physics.

share|improve this answer
add comment

Higher heat is never a good thing. Also many processors are made almost identically on the same product line and then underclocked and sold cheaper than non-underclocked or otherwise uncrippled chips. It's cheaper than designing and manufacturing each model uniquely.

share|improve this answer
add comment

Higher heat and higher voltage make for faster deterioration of semiconductors through shifting of impurities inside SiO2 that are responsible for the semiconductor properties. Here is a nice, readable paper on the topic:

http://www.phmsociety.org/node/373

share|improve this answer
add comment

Not the answer you're looking for? Browse other questions tagged or ask your own question.