if you ran a 3570 at around 100-110C for 12 hours (constant throttle under load)...

[Turbonium]

...would it be permanently damaged?

 

[Centauri]

So you're saying it still functions? Amazing.

 

[Exophase]

It's supposed to shut down if the temperature is dangerously high. This isn't a sophisticated function, some very simple ICs have this feature and CPUs/motherboards have had it for a long time, even before they had internal temperature diodes and had to rely on thermistor probes between the CPU and heatsink.

Naturally, there has to be a fair amount of margin for this to work, or the shutdown logic itself could be compromised. The only real risk is if the thermal diode isn't reporting very good values, but in that case you shouldn't even be able to read the bad temperatures unless your readings are coming from somewhere else. Nor should it be regularly throttling.

So if it isn't shutting down Intel probably thinks it's safe and if Intel thinks it's safe it's probably safe.

 

[Phynaz]

First you most likely were not running that hot as they will throttle to keep themselves at 105. Running at 105 is the design limit so it should still function fine. You have probably cut many years off it's serviceable lifetime and may now be looking at mere decades.

 

[Pilum]

...would it be permanently damaged?

No, IVB throttles at 105°C, stays there and works fine. Tested on my i5-3550 with stock cooler, IBT and non-optimal thermal environment. Didn't run it for hours though, just a few IBT runs. But as this is within specs, no need to worry.

Edit: Oh, and at turbo +4, i.e. 3.9 GHz. Don't know if you can get a stock 3550 to 105°C.

 

[ShintaiDK]

No is the correct answer.

However how you gain that temperature could be another matter.

 

[Puppies04]

You will not have damaged the cpu in the sense that it will still perform as it was originally intended, i.e at stock. That said, heat is one of the factors that degrades cpus over time so you might have knocked a hair off it's maximum OC frequency. I really wouldn't worry about it though.

 

[ShintaiDK]

If it failed or got damaged within 3 years. I am sure Intel would cover the RMA under warranty. Assuming its running stock.

 

[Arkadrel]

So you're saying it still functions? Amazing.

^ this.

If you run a cpu at 110C for 12hours it probably wouldnt be funktional.

You will not have damaged the cpu in the sense that it will still perform as it was originally intended, i.e at stock. That said, heat is one of the factors that degrades cpus over time so you might have knocked a hair off it's maximum OC frequency. I really wouldn't worry about it though.

^ wrong.

The stuff the chip is made of, starts to melt around those temps (I believe).
Also even if its just a tiny bit, its not just "knocking hairs breaths of max oc ing potential off".

Minor damages results in error calculations occureing more often (even perfectly fine CPUs have them), but a CPU thats been damaged is bound to have more.

 

[AtenRa]

...would it be permanently damaged?

This measurements are for the T-Junction (core temperature measurement from the DTS censor) or CPU temperature (Tcase)??

 

[ShintaiDK]

^ this.

If you run a cpu at 110C for 12hours it probably wouldnt be funktional.




^ wrong.

The stuff the chip is made of, starts to melt around those temps (I believe).
Also even if its just a tiny bit, its not just "knocking hairs breaths of max oc ing potential off".

Minor damages results in error calculations occureing more often (even perfectly fine CPUs have them), but a CPU thats been damaged is bound to have more.

The tjmax is 105C. It will throttle at 105C. This is the garantee Intel gives for its 3 year warranty.

 

[amenx]

Based on earlier C2D chips I've had, the system just shuts down when it reaches TJMax. With newer chips, it should also do the same, not just throttle. For example, if your using the stock push pin cooler, the pins can pop out if they're not absolutely secured. When it reaches 105c its likely something like that is the cause, and it should shut down rather than throttle.

 

[Pilum]

Based on earlier C2D chips I've had, the system just shuts down when it reaches TJMax. With newer chips, it should also do the same, not just throttle. For example, if your using the stock push pin cooler, the pins can pop out if they're not absolutely secured. When it reaches 105c its likely something like that is the cause, and it should shut down rather than throttle.

IVB throttles at 105°C until the temps go down again. The system doesn't shut down (i5-3550 on Asrock Z77 Pro4).

Excerpt from the RealTemp log:

12/20/12  14:45:55  3890.13   103   101    98    95    99.0
12/20/12  14:46:00  3860.13   105    99    96    95    99.8
12/20/12  14:46:05  3890.13   103   101    97    96    99.4
12/20/12  14:46:10  3850.13   104   101    98    95    99.4
12/20/12  14:46:15  3820.13   102    98    95    94    99.6
12/20/12  14:46:20  3830.13   105   103    97    95    99.6
12/20/12  14:46:25  3840.13   104   101    97    96    99.9
12/20/12  14:46:30  3820.13   104   102    98    95    99.8
12/20/12  14:46:35  3850.13   105   101    98    96    99.6
12/20/12  14:46:40  3820.13   103   101    97    95    99.8
12/20/12  14:46:45  3760.12   104   101    97    95    99.8
12/20/12  14:46:50  3820.13   103   102    95    95    98.4

 

[ShintaiDK]

It actually shuts down at 135C, should the throttle fail to lower the temperature. But the mobo might do it sooner.

 

[Pilum]

It actually shuts down at 135C, should the throttle fail to lower the temperature. But the mobo might do it sooner.

But this should never happen in real-life, even with a disconnected HSF. The CPU will just throttle to a crawl; Intel has had this down pat since the P4. If you put the system into an oven, OTOH...

 

[tweakboy]

Poor CPU, 105c ....... Intel should have not put the plate crapola,, so block can have contact with the cores...gl

 

[IntelEnthusiast]

It actually shuts down at 135C, should the throttle fail to lower the temperature. But the mobo might do it sooner.

For the 3rd generation Intel® Core™ processors the thermtrip is at 130°C, to shut it down at. This is 5° hotter than what the 2nd generation Intel Core processors were at. http://www.intel.com/content/www/us/en/processors/core/3rd-gen-core-desktop-vol-1-datasheet.html page 71

 

[Vinwiesel]

Quote: You will not have damaged the cpu in the sense that it will still perform as it was originally intended, i.e at stock. That said, heat is one of the factors that degrades cpus over time so you might have knocked a hair off it's maximum OC frequency. I really wouldn't worry about it though.

^ wrong. The stuff the chip is made of, starts to melt around those temps (I believe). Also even if its just a tiny bit, its not just "knocking hairs breaths of max oc ing potential off". Minor damages results in error calculations occureing more often (even perfectly fine CPUs have them), but a CPU thats been damaged is bound to have more.

No, Puppies04 had it right. If you run a chip at high temperatures it will wear at an accelerated rate, reducing its performance, and increasing blue-screens and crashes. Most likely this damage will only be seen when attempting to overclock.

The chip does not "melt" at 110C, in fact many manufacturers do accelerated life testing at well in excess of 110C. Intel decided that 105C was the maximum safe temperature by running their chips at significantly higher temps than 105.

 

[Magic Carpet]

Based on earlier C2D chips I've had, the system just shuts down when it reaches TJMax.

My E2160 doesn't shut down, overclocked to 3.2 it only throttles when the temps reach around the 88-90c mark.

 

[amenx]

An e6400 I had with Intel stock cooler shut down when one of its pins popped out (bumped my knee hard on the case). This is the main reason I think it has the 'Thermtrip', to protect against such events or improper cooler installation. In normal usage (fan intact) if it overheats, throttling should occur.

 

[Turbonium]

Idc? What's your take? I'm guessing not much different than what's already been posted.

 

[sheqangi]

That could mean anything though.

 

[Idontcare]

Idc? What's your take? I'm guessing not much different than what's already been posted.

Its not an issue for the CPU since it was allowed to throttle as needed.

The chip is designed to last for years under load at 105C. Keeping it cooler than 105C will allow it to last even longer as well as use less power (less leakage).

AMD's chips are not designed to operate anywhere near those temperatures. I mention this only to point out that there is no rule of thumb other than knowing what the manufacturer determined to be safe. It varies every chip generation, even within the same node.

Where OC'ers get into significant degradation is when they over-volt and get their processor hot.

Running at TJmax with stock settings is not a concern, but don't do that at 4.5GHz with 1.3V D:

 

[Magic Carpet]

An e6400 I had with Intel stock cooler shut down when one of its pins popped out (bumped my knee hard on the case). This is the main reason I think it has the 'Thermtrip', to protect against such events or improper cooler installation. In normal usage (fan intact) if it overheats, throttling should occur.

I see. I had a Pentium 4 running in the 90c range with the same heat sink issue. It wasnt until I looked inside and noticed the fault. I was surprised it was still running though. Some boards/bioses have different fail-safe systems in place.

 

[Turbonium]

What's a normal max temp for a 3570 under 80%+ load?