• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Why do CPUs never seem to break?

413x

413x

413x

Caporegime
Joined
13 Jan 2010
Posts
33,217
Location
Llaneirwg
Was just thinking about this. And I'm not really sure if I've just been lucky. But it seems like CPUs never seem to die.

Obviously hdd die because they are mechanical. Fans on graphics cards. ButI must admit I don' k much about how a cpu actually works (I know how it's made and what it contains.)

Is the thread title true? And if so, why?
 
Rare that a PC will come in with a CPU faulty but it does happen from time to time. Less than 5 failures last year and that's across all chips / sockets, usually something else has happened to trigger the failure.
 
Why is the information regarding MTBF of a CPU under NDA? :confused:

I think it is very safe to assume that CPUs have MTBF over 1,000,000 hours, maybe even more than 10,000,000 hours.
 
I've never known one fail, although I'm sure it can happen on rare occasions.

Unless you cook it with too much voltage, there's not much that can go wrong with them, assuming they weren't faulty from factory.
 
4K8KW10 said:
Why is the information regarding MTBF of a CPU under NDA? :confused:

I think it is very safe to assume that CPUs have MTBF over 1,000,000 hours, maybe even more than 10,000,000 hours.
Click to expand...

It's an interesting point if that's true, but why bring out chips so frequently? Trying to cover something up maybe?
 
My 1090t that I sold on here is probably still rocking it was on 24/7 @ 4ghz for almost 10 years before I sold it and it was still working like a champ. CPU's these days are like tanks :)
 
because they a single thing, a gfx card is 10, 20 or 30 bits.. glued to a bit of plastic.. Possibley also one company makes the hole cpu so the bits have less moving / shipping etc... A lot less bits to break and less messing about involved in creating the final product
 
CPU lifetime is good due to various reasons like transistors which is on microprocessor have several features like oxide thickness, width and length are scale by constant factor etc ... and hence CPU never seems to be break....
 
Properly made semiconductors are inherently reliable when run inside designed specs.
There's no mechanical wear and when operating voltage and temperature are kept in reasonable areas there's very little wear from those.



Though Flash memory used by SSDs is very different thing.
It has extra insulated floating gate in transistor and basically uses pulse of overvoltage to push charge through insulator into that floating gate.
Which degrades/wears that insulation.
Similarly erasing charge causes wear.
Think Flash memory like some clay/stone tablet:
Scratching those markings to it wears it, just like grinding surface flat for new writing.
And eventually you'll go through it.
 
Memory of all sorts generally tends to be a bit more susceptible to wear due to the density and often running with less tolerances to electromigration from voltage, etc.

Intel and to an extent AMD have generally been fairly conservative, especially at first, with the voltages used in their CPUs as can be seen with the often later VID range increases in the data sheets. Also Intel's design approach tends to result in robust products but with some degree of inflexibility in approach that makes them as a company poorly suited to certain products like gaming GPUs.
 
You must log in or register to reply here.
Back
Top Bottom