4790k oc to 4.7ghz at 100ºC ? o_o

Akonovo

Reputable
Jun 18, 2014
35
0
4,540
Hi,
My processor (4790k oc to 4.7ghz with auto vcore (autoset at 1.362v)) reach 100ºC in less than 10minutes of prime95 blend test.
My cooler is an Noctua NH-D15 on an Asus Hero VII.

The system seems to be fully stable, with iddle temps of 27ºC and gaming temps of <70ºC. (24ºC mobo temp)

Are this temps normal? Or I failed during NH-D15 installation? (I think I applied too much thermal compound) I had bad luck with the cpu lottery?
Also I want to know if there is a way to set in ROG bios to force a shutdown when the CPU reach certain temp.


Im currently testing manual vcores (instead of BIOS auto-set) Ive seen reviews that keep this cpu stable at 4.8ghz 1.327vcore with 80ºC max temp....
Can a 0.035 vcore reduction mark a 20ºC difference?? o.o
 
Solution
Given the low idle temp I'd suggest that the hsf is installed correctly.

I'd suggest that was too much, try lowering the voltage, or an offset of -0.005 or similar as the auto voltage will have ramped the voltage up to match the OC and it might be overvolting for what is needed.
Given the low idle temp I'd suggest that the hsf is installed correctly.

I'd suggest that was too much, try lowering the voltage, or an offset of -0.005 or similar as the auto voltage will have ramped the voltage up to match the OC and it might be overvolting for what is needed.
 
Solution
max temp for Haswell, don't remember if it was 98C or 102C but you don't want to be anywhere near that if you want to keep your cpu for a long time. A good place to be is around 80-75*c for max temps, also remember that doing a prim test will stress your cpu a lot more then gaming so your temps will always be higher when stressing it.
 

Akonovo

Reputable
Jun 18, 2014
35
0
4,540
Yay, I know that 100ºC is too high, it was like 10sec at that temp (I stoped the test on that point). But, as far as I know, 4790k throttle at 90ºC and cpu reset at 100ºc to prevent chip damage. Dunno why mine keept running at 100% stability without any issue/error/warning (its strange)...
The air of my case was being pushed hot, but not burning (like my graphics card at 80ºC do), so maybe its a Temperature load issue (were using RealTemp 3.70)...

Ill try to find where its the option to overheat-shutdown on this bios, and then ill try to set voltage manually (instead of auto) cause seems that my motherboard is overvoltaging my cpu. Some reviewers reach stable 4.9ghz with lower vcore (1.32v) with "only" 82ºC, while my bios set 1.36v for 4.7ghz...

For those who ask, I know that this OC is overkill, specially with single GPU. But its a new computer (2days old) and im stresstesting to know its limits in order to look for bottlenecks and know if return it for upgrades/downgrades.

So, anyone knows why my cpu dont throttle/shutdown? :/


 

Akonovo

Reputable
Jun 18, 2014
35
0
4,540
Definetly the overheat issues was due Asus auto-vCore. Limiting it to a maximum of 1.32v, my overage temps during stresstests are 70ºC with peaks of 75ºC.
Overclock to 4.7ghz isnt 100% stable at 1.32, but its stable enough. I'll test it a bit more, but stress tests took 2hours for BSOD, and gaming seems to be fine (tested on gw2 and StarCitizen).

So, learned rule number one: Never let the bios to autoregulate the vCore if you dont have nitrogen cooler xD
 


good, look for errors in event log. WHEAs are recoverable errors, that will bsod under some loads, no wheas equals stability