How exactly is a CPU damaged by heat? It's not like the

>>60289879
Metals expansion past 100 C supresses tolerance

>what even is physics
Please just use Google. This is grade school stuff.

>>60289879
>It's not like the fucking metals start melting after 100c
maybe the silicon inside does.

>>60289920
>Si melting at 100C
kek

>>60289920
Silicon melts at 1400c anon...

>>60289920
Please tell me you are not some scandi nigger who thinks silicon is silikon

>>60289879
After a certain tpb it doesn't fit back in place anymore

>>60289879
Electromagnetic migration.

1 nm expansion kinda matters famala

>>60289879
Retarded frogposter.

whats the upper limit for CPU temps? 100 C?

>>60291849
every cpu has operating temperatures. It's like regulated by the government every electronic has this posted.

>>60291849
There's no magic number, but yeah a good rule of thumb is that 100 C means you pull the fucking power cable out of your PSU immediately.

Something's usually wrong if you're sustaining temps in the mid- to high-80s. Yes, plenty of people regularly put their CPUs under that much load and it's okay, but those people need very effective cooling solutions instead of baking their processors all day.

Your chips are soldered at 200C +

It's not that the temperatures are damaging your cpu, but rather that the transistors stop functioning properly past 100C, so there's no point in running the CPU any hotter.

>>60289879
fr4 in pcbs starts to burn at 140C
most coolants can't take more than 120C

>>60289879
Silicon, unlike metals, loses resistance with higher temperature, thus you're risking short-circuiting your CPU. Also don't forget that thermal sensors show only the average temps, some parts can be hotter than that. Also PCB can start melting. Also the dust can start burning. But I'm interested in where the 100 number came from as well, what were the tests for that.

File: maha.jpg (22KB, 235x318px) Image search: [Google]

22KB, 235x318px

>>60292058
After 100c water begins to boil, thus = CPUs are made out of water, they don't want us to know

This whole thread is a disgrace.
Electrons moving through the transistors slowly damage it by knocking atoms around, higher temperatures make it easier to knock atoms out, and higher voltages means the electrons have much more power to do so. Since transistors are like 50 atoms across, if enough atoms are moved to the wrong place the thing doesn't work anymore.

>>60291849
A good rule of thumb is don't go over 80C under full load
70C during normal use
To prevent decreased lifespan
But to outright kill it you need quite a bit more than 100C

Do CPU heat problems pose a risk to the CPU itself only or can the mobo also have issues?

>>60292219
Yes.

>>60292170
>knock atoms out
Holy shit! Does that mean that after 100C a CPU becomes radioactive?

>>60292455
by your retarded logic, entropy is what makes things radioactive

>>60292455
https://en.wikipedia.org/wiki/Electromigration

>>60292170
except that's fucking wrong

The only thing remotely similar is electromigration, except that happens in conductors and is proportional to current density and not voltage

>>60292579
>a CPU isn't made of conductive material

>>60292579
>current density has no relation to size of conductor

>>60292591
>>60292619
Point me to literature where electromigration is a problem in Silicon rather than interconnects

>>60292535
But wiki says exactly that: https://en.wikipedia.org/wiki/Radioactive_decay
>The decay process, like all hindered energy transformations, may be analogized by a snowfield on a mountain. While friction between the ice crystals may be supporting the snow's weight, the system is inherently unstable with regard to a state of lower potential energy. A disturbance would thus facilitate the path to a state of greater entropy.

>>60289879
>intel shill damage control

>>60292666
>>analogized
continue being retarded

>>60292730
>A disturbance would thus facilitate the path to a state of greater entropy
So no entropy - no radiation. Who's being retarded now?

>>60292455
>>60292579
http://www.sciencedirect.com/science/article/pii/0038110188901062

Retards

>>60292886
>bipolar, not CMOS
>aluminum contacts with no tungsten plug
>silicon isn't even electromigrating
>1988

really grasping at straws here huh

You'll never get electromigration in silicon because there are no grain boundaries in monocrystalline silicon

>>60292745
it is likely all matter radiates over time

>>60292172
my old phenom x2 945 got past 113°c and didn't die after it
funniest thing is that he was full throttling at 113°c
and got back to stock frequency at 110°c