IntelBurnTest Ruin Laptop?

Can running IntelBurnTest damage a thin notebook (MacBook Air)? I ran it in BootCamp on "high" setting for 50 passes and it got very hot (I didn't check temps), it failed at pass 45 but didn't give an exact reason, the system never froze or blue screened.

It's a new Air mid-2012. I'm worried I may have shortened it's lifetime. If everything still appears fine, should I be ok?

 

If everything still appears fine, then there is no way of telling if you shortened its lifespan. Usually temps are supposed to shorten the lifespan of ICs but considering you just ran it on one day, it shouldn't have done significant or noticeable damage to your laptop's lifespan.

P.S. For future reference, try not to do anything more than browsing the net, writing Word docs or making ppts on your Air

 

One or two tests will not appreciably damage the lifespan of the notebook. Doing it over and over every day might.

 

It only ran a couple hours. Can't believe it made it that long tbh. I'm SUPER paranoid so I run tests on everything. Turns out people don't recommend running IntelBurnTest on laptops, especially Macbooks. I come from the PC desktop world.

Now I'm freaking out! I hope it doesn't crash a couple months down the road. uhhh so much stress from testing all this crap, frist my ext hdd, now this. I love / hate computers. :-(

 

If it fails at the 45th pass, have apple replace it it until it can pass 50 tests

 

Running 10 passes on my Pentium only took 5 minutes or so. Couple of hours for 50 passes sounds pretty slow...

 

good god why?, IBT is BRUTAL on desktops let alone laptops, and its primary use is to weed out OC instability. The mobile chips are quite good at keeping themselves safe from current but the heat is a big killer, any prolonged amount of time above 90 degrees (some advocate 80 degrees) is unsafe for ICs. You also have to consider the effect of the heat on the nearby ICs as well as while the CPU is nice and cool, the radiant heat may be detrimental to the neighboring parts.

 

Maybe it downclocked or something to prevent higher temps?

 

Thanks. I see now it was super dumb. How would I know if something is damaged? All I know is to test everything by using it normally. I'm still very worried.

 

Is this sarcastic? I hope so.

 

Even with the heat, a modern CPU should easily be able to handle 50 passes of the IBT ("easy" as in complete it). If your Air is having issue trying to calculate a floating-point number, I'd definitely return it / exchange until it can actually pass the test.

 

Hmm.. Other are saying it's insane to run this test, you're saying it should pass... Not sure which is right. I don't think it failed though, cause the stop message said something like there could be something missing from the install or some bull about not sure why it stopped.

 

Stop scaring this guy - it might be dumb to run so many passes, but if the laptop is running fine there's NOTHING to worry about.

Listening to these self-appointed guru geeks is silly, Reventon - just stop running these synthetic tests, and enjoy your notebook!

 

did you force the CPU to run at full speed using Throttlestop? if no then its probably OK, Apple has an excellent throttling scheme to keep the CPU from overheating so it might've actually spent most if its time downclocked on IBT. Its easy to forget that Apple does sell machines to business clientèle so its not correct to say they cannot handle heat, I'd say the aluminium unibody helps with heat better than most Toshiba/Acer laptops I've seen.
I'm a firm advocate for lighter sensitive tests vs brutal load tests like IBT, being stable under heavy stress does not equate correctness/stability. I've seen many Furmark stable GPUs fail under trivial testing conditions like AtiTOOL or running self checking Floating Point calculation Loops. It is very rare for Intel CPUs to be DOA as far as I've experienced, while they do fail under prolonged OC, I've personally never had to RMA an Intel CPU which has been running at stock (even mobile chips).

 

No throttlestop. It's working fine right now, and past the return period. I guess I expected someone to say, if it still works fine, theres no problem. Would you go through the trouble to rma or get it repaired?

 

You were running a stability test for hours, it eventually failed because the computer's temperature exceeded design limits, and after cooling down everything once again functions properly. Unless persistant problems start to show up it highly, highly, unlikely that your burn test did any lasting damage. Certainly not worth RMA'ing.

This thread is starting to remind me of a few times when people asked me about RMA'ing CPUs because they did not overclock as well as they wanted them to. If you want to do a useless RMA and waste time and (their) money go for it.

 

IBT is no where brutal when compare to people prime95 +furmark their machine...

If it run fine for a hourish, then don't worry about it.

 

That program is good for short stability tests, otherwise do not run. Heat is killer and not only for CPU, but for the surrounding modulators, power supply and for everything which is connected with the extra jack.

 

Failing a stability test (i.e. giving a wrong calculation) definitely should not happen at stock settings. Ever. The CPU expected to be stable at the factory settings. The only thing that should happen when the CPU gets hot is to either to throttle (e.g. while still performing the instructions correctly) or, in extreme cases, to shut down.

Though if it doesn't affect OP then I guess there's no problem. But if he wanted an RMA that's definitely a valid reason to.

 

LOLOLOLLOL. All PC OEM's have the same throttling scheme. It's called TJMax. The CPU does it itself. The Air, like any PC, will gladly hit that limit if you push it.

Extreme heat is only bad for the physical CPU when it's ran at those temperatures over long periods of time. You begin to get transistor damage and you'll eventually begin seeing electron migration between transistors... then it'll be pretty much toast. But by long periods, I mean LONG periods of time. Not overnight.

The main thing heat damages in notebooks is the motherboard and other components that aren't rated to handle 90/100C+ temperatures. Ex: RAM/Motherboard components/SSD...


IMHO, your Air is just fine.

 

Should I run the test again for another 50 to verify if the new firmware/efi bios update fixed stability?

 

Agreed, but I really would not be surprised to see that the cause of the failure was a component heating up past its specified limits. IBT is notorious for punishing, and exceeding, thermal cooling capabilities for computers. I agree that a CPU should (and they do) have throttle mechanisms. At the same time there are other periphery components (RAM, PCB components, etc) that do not have defense mechanisms for overheating so you have to rely on the cooling system. Its possible that the CPU did not overheat, but the RAM did, or some other component did. Without (a lot) more data its hard to say.

Its also possible you experienced a failure actually caused by a bad RAM chip or some bad cache (this possibility is remote) rather than overheating. Do some testing with MemTest86+, NOT IBT, to confirm RAM/cache stability.

If you're going to run IBT again, run it for 10-20 minutes or less. Preferably less. Closely monitor temperatures as well so if you DO experience a failed test you can try and determine if it was caused by overheating or not.

 

How could it pass 45 times if it were messed up? I've already run memtest overnight. I thought the way it stopped was weird because it never said it failed. Can't remember exactly but something about could be missing a software component. It's on bootcamp W7 x64 SP1 not updated. Could be a bootcamp driver issue, I don't know.

I just don't want to run into issues when using Parallels, PhotoShop, XLD audio conversion, etc. my normal work flow.

 

This thread failed to deliver. I clicked hoping to hear about a damaged laptop.

Why are you so paranoid about the stability of a stock clocked CPU?

Wait, why aren't you paranoid enough? Srsly, 50 passes? You need to run 500,000 passes!!!! You need to keep running stability tests for a whole year, because what if on day 363 it errors? What if it fails on day 364?

Or, get over it, and just enjoy using your computer.

BITD I used to do system builds for a computer shop. We did those burn-in tests, and advertised it. I don't recall anything ever failing during the 24 hour burn-in if the computer passed the original build and worked fine.

 

If there was a fault (big IF) it might be a fault that only shows up when a part is sufficiently hot. A RAM chip that can't run at the rated speeds if it gets to a certain temperature, etc. It's highly unlikely but I have seen it happen. Again, if you've run the recommended tests and gotten no failures (MemTest86+ for X hours, low number of IBT loops) then just stop and enjoy the computer.