is it nearly impossible to burn out a CPU anymore?

[brblx]

a mild curiosity i've been thinking about. personally, i've never had a CPU or GPU fail on me, and i've run some HOT stuff.

i started building computers in the pentium 2 era (what active cooling?). as i've built newer computers with more complex cooling setups; and helped other people with poorly seated GPU coolers, push pin woes, and whatnot, i've still yet to see a microprocessor destroyed by heat. you'd think an i7 with little to no heatsink contact would not withstand multiple bootups and crashes as the user doggedly attempts to get the computer set up. but i've seen it- just correct the issue and it seems to be no worse for wear afterwards.

am i just being silly? or have we just gotten to a level of efficiency where these things just have a hard time lighting themselves on fire?

wonder what an overclocked P4 with no cooler and no failsafes would do...it's weird, those space heaters, in many cases, have been running maintenance free in a hell of a lot of workplaces for ten freakin' years.

 

[Idontcare]

The device physics are immutable, you can and do destroy your IC's a little bit every time you use them. There is simply no way around the reality of the second law of thermodynamics.

Use them at hotter operating temps or at higher operating voltages and you destroy them all the faster.

The degredation that is occurring may be transparent to you though because of all the safeguards built-in, the engineering safety margin, and the levels of ECC involved plus the fact that not every error results in a hard-lock.

You could very well be silently corrupting your data files or infusing mathematical errors into all your calcs without having realized it yet.

I killed my QX6700. It happens.

 

[pantsaregood]

I have a Pentium MMX somewhere that still runs. Loads of Pentium 4s that still run fine. Cedar Mill Celeron still runs (and at 4.32GHz) even though its heatsink wasn't properly seated for a year or so. Had a 450MHz Pentium 3 Katmai whose fan was broken that never failed.

Really though, modern CPUs are ridiculously tenacious. Nehalems ran pretty hot in general. Later revisions of the Pentium 4 were hot, too. CPUs from ~2000 on do a pretty good job of saving themselves if heat is an issue. They'll throttle themselves back if they're too hot, or even shut down if it is required.

http://www.youtube.com/watch?v=A4l8aQ07yvU

AMD obviously didn't have the thermal safety feature down with K7.

 

[brblx]

wow, yeah, that AMD lost the magic smoke for sure.

so the reason the others are always fine is just failsafes? and a P4 in a generic dell with 10 years of shop dust inside doesn't trip it?

 

[Idontcare]

wow, yeah, that AMD lost the magic smoke for sure.

so the reason the others are always fine is just failsafes? and a P4 in a generic dell with 10 years of shop dust inside doesn't trip it?

Yeah they'll throttle themselves, reducing clockspeed and forcing wait-states, unbeknownst to the user outside of the fact that the user will probably notice the computer seems to be running really really slow.

I can't seem to find the review site anymore, but there was a review site that did awesome reviews of cpu's in which they characterized the throttling thresholds of cpu's as a function of temperature and power-consumption. Drawing a total blank on the website name and random googling isn't getting me any closer...

 

[OBLAMA2009]

ive heard that intel expects cpus to last 10-12 years. so the practical answer is "no it is not possible to burn out a cpu because no one ever runs a cpu for longer than anywhere near its lifespan"

 

[ibex333]

You could very well be silently corrupting your data files or infusing mathematical errors into all your calcs without having realized it yet.

I killed my QX6700. It happens.

WHAAAAATTT?! Are you sure? An overclocked CPU can corrupt my data files? where did you hear that? First time I hear something like this.... Lately I have been experiencing a "ghost effect" with my PC. Sometimes some random files just... Disappear! I could have sworn I had em a few weeks ago and then.. poof! They're gone!

Can it really be because I overclocked my CPU?

 

[drizek]

Have you tested it for stability?

Run prime95.

 

[skipsneeky2]

Doubt it ibex333, at best i think a unstable overclock would cause a crash or a blue screen.

Done overclocking on plenty of cpus without such a issue.

Never fried one cause im anal about temps and having the best cooling solution for the processor.

 

[BD231]

The issue is still the same, only now you burn up your chips at a much slower rate via over volting which they call CPU degradation. You still mess up your hardware, just not typically to the point of starting a fire or rendering the CPU useless.

The biggest concern is damaging your chips scalability, at higher voltages you can very easily shave off a good portion of overclocking headroom if volts stay to high over time.

 

[Bill Brasky]

Back in my overclocking days, my motherboards would fail long before the cpus. I have a stack of perfectly functional cpus and no board to run them on. lol.

 

[Idontcare]

WHAAAAATTT?! Are you sure? An overclocked CPU can corrupt my data files? where did you hear that? First time I hear something like this.... Lately I have been experiencing a "ghost effect" with my PC. Sometimes some random files just... Disappear! I could have sworn I had em a few weeks ago and then.. poof! They're gone!

Can it really be because I overclocked my CPU?

Doubt it ibex333, at best i think a unstable overclock would cause a crash or a blue screen.

Done overclocking on plenty of cpus without such a issue.

Never fried one cause im anal about temps and having the best cooling solution for the processor.

Yes its true.

Another risk is silent data corruption by undetected errors. Such failures might never be correctly diagnosed and may instead be incorrectly attributed to software bugs in applications, device drivers, or the operating system. Overclocked use may permanently damage components enough to cause them to misbehave (even under normal operating conditions) without becoming totally unusable.

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

It is the reason for the R in RAS.

Reliability means features that help avoid and detect such faults. A reliable system does not silently continue and deliver results that include uncorrected corrupted data, instead it corrects the corruption when possible or else stops and reports the corruption.

http://en.wikipedia.org/wiki/Reliability,_Availability_and_Serviceability

 

[Bartman39]

The issue is still the same, only now you burn up your chips at a much slower rate via over volting which they call CPU degradation. You still mess up your hardware, just not typically to the point of starting a fire or rendering the CPU useless.

The biggest concern is damaging your chips scalability, at higher voltages you can very easily shave off a good portion of overclocking headroom if volts stay to high over time.

Overvolting causes "electromigration" and is more harmful than degradation which is caused by heat from overclocking... If you overclock without a voltage increase then degradation will still happen due to the increased heat but more advanced cooling can help with this... But when you overvolt a cpu it will not only increase the heat it will increase electromigration which you cannot do anything about... Micro pathways get smaller with every die shrink so this becomes more of a problem until there is a design change of some sort... Kinda hard to insulate a few million tiny wires though...:hmm:

 

[brblx]

shit, did i accidentally create a highly technical thread? heh.

seems like the opinion is fairly unamimous- there are very limited instances where you can directly kill a CPU with heat, but excessive heat over time shortens the life of the part- whether this is actually noticeable to the user would be the more arguable factor.

 

[PlasmaBomb]

Overvolting causes "electromigration" and is more harmful than degradation which is caused by heat from overclocking... If you overclock without a voltage increase then degradation will still happen due to the increased heat but more advanced cooling can help with this... But when you overvolt a cpu it will not only increase the heat it will increase electromigration which you cannot do anything about... Micro pathways get smaller with every die shrink so this becomes more of a problem until there is a design change of some sort... Kinda hard to insulate a few million tiny wires though...:hmm:

The increased frequency will also increase the wear rate... if I have read IDCs contributions correctly.

(So for an overclock the increased heat and increased frequency will both reduce the CPUs life)

 

[Wanescotting]

Well, I have forced thermal shutdown on my 1090t............still overclocks to 4.2ghz, still stable......i'm sure I have cut the life span of my cpu by, oh, say 30%....BUT I will have upgraded by then (if bulldozer EVER comes out so I can compare to SB)

 

[Fayd]

WHAAAAATTT?! Are you sure? An overclocked CPU can corrupt my data files? where did you hear that? First time I hear something like this.... Lately I have been experiencing a "ghost effect" with my PC. Sometimes some random files just... Disappear! I could have sworn I had em a few weeks ago and then.. poof! They're gone!

Can it really be because I overclocked my CPU?

highly doubtful. he's speaking in terms of theories.

 

[Bartman39]

The increased frequency will also increase the wear rate... if I have read IDCs contributions correctly.

(So for an overclock the increased heat and increased frequency will both reduce the CPUs life)

This is true also...

WHAAAAATTT?! Are you sure? An overclocked CPU can corrupt my data files? where did you hear that? First time I hear something like this.... Lately I have been experiencing a "ghost effect" with my PC. Sometimes some random files just... Disappear! I could have sworn I had em a few weeks ago and then.. poof! They're gone!

Can it really be because I overclocked my CPU?

highly doubtful. he's speaking in terms of theories.

OMG...? How long have any of you been in the overclocking game...? From several years ago we learned after overclocking our old single core systems (from Celeron 300a`s to XP`s) along with the old OS`s would corrupt files just about every time we had a crash (was a cross your finger game)... The older 9X series of OS`s were alot worse but with the arrival of NT tech things got better but it still happens... Why do you think systems crash...? A virus or worm can cause havoic but a dying cpu or data corruption caused by and overclocked system can be alot harder to figure out... What do you do when you read a dmp file that has nothing but 0`s...? It happens...

 

[Idontcare]

OMG...? How long have any of you been in the overclocking game...? From several years ago we learned after overclocking our old single core systems (from Celeron 300a`s to XP`s) along with the old OS`s would corrupt files just about every time we had a crash (was a cross your finger game)... The older 9X series of OS`s were alot worse but with the arrival of NT tech things got better but it still happens... Why do you think systems crash...? A virus or worm can cause havoic but a dying cpu or data corruption caused by and overclocked system can be alot harder to figure out... What do you do when you read a dmp file that has nothing but 0`s...? It happens...

I too am a bit baffled by the, uh how to say this in a non-inflammatory way, ignorance confused shown in some of these posts...

What do they think apps like LinX and Prime95 are checking for when they check for computing errors?

Where do they think those computing errors would have shown up if they were using applications which were not checking for errors in parallel?

And the fact that apps like LinX and Prime95 do not check for computation errors in all 800+ instructions means their OC could very well pass Prime95 and LinX but still be generating computing errors that are corrupting their datafiles.

I happen to run an app called Gaussian which would generate random garbage output files instead of the expected output on OC'ed cpus that passed prime95, I'd have to back off on the overclock to get the correct results. F@H is another example of an app that will result in errors.

There is a reason it was common OCing knowledge that you do NOT install windows while your CPU was OC'ed.

I am surprised, as you are, that people don't know this. Oh well, ignorance is bliss as they say! Now these guys have a whole new level of crap to be worried about thanks to this thread. Whoops!

 

[Ben90]

Basically every crash that that happens because of an overclocked CPU happens because a transistor didn't switch correctly. When testing new overclocks, its very easy to see that our lame transistor messed up. Your computer or application needs to restart.

Often times, when run on the brink of stability, that lame transistor messes up on some math not visible to the user. Suddenly your F@H molecule pulls some crazy quantum physics and an atom is moved to a completely wrong spot. Fortunately, F@H has an error correcting algorithm so you don't send them wrong results. The work unit just gets discarded. Not every program checks for computation errors.

Every single file on your hard drive is subject to silent data corruption.

 

[Gryz]

A few years ago I had a completely watercooled system. One day I had blockage in my GPU-fullcover waterblock. Blocking the full loop (cpu, gpu and psu). Both my CPU and GPU heated up to 110C. I was raiding at the time, we were learning a new boss. Framerates were dropping, I wasn't sure it was because of the new content, or some other problem. Framerates dropped to 10 fps and lower.

Powered down the system. Found the problem. Cleaned my loop. The system would boot. But after a minute I would get artifacts in 2D, and 3D wouldn't work anymore. 8800gtx was fried. The E8500 cpu is still working fine today (this text is typed on it).

So I'm wondering. Can CPUs handle more abuse than GPUs ? Was it the memory on my 8800gtx that melted, but not the GPU ?

 

[Tuna-Fish]

So I'm wondering. Can CPUs handle more abuse than GPUs ? Was it the memory on my 8800gtx that melted, but not the GPU ?

The 8000 series used a new solder material that earned a reputation for melting too easily when the chip overheats. The semiconductors themselves can survive much higher temps for a short duration of time, but the material below them is plastic and solder, and most likely, that was what you broke.

 

[pantsaregood]

If overclocking results in undetectable errors so much, what's to say that CPUs aren't throwing around loads of errors at stock speeds?

 

[nyker96]

this is easy to answer, in recent years, all the cpu made from intel/amd has temp diod that shuts down when overheated. so you don't see many cpu gets melted down due to temperature issues anymore.

 

[drizek]

Even my Barton had autoshutdown. I think it might have been the first AMD proc to do it(and it saved my bacon)