i7 8700k Delid Results Temp Comparisons

[changsta2004]

Hi. This is an update from my previous post on overclocking. I finally delidded my i7 8700K.
I am wondering what temperatures people are getting around the 4.9-5.1 GHZ overclock and what cooler you guys are using.
I am using an Air Cooler by Jonsbo (Jonsbo CR-601), found in Korea
At 4.9 GHZ, 1.23 V, I get idle temps around 40-41 and 65-67 under a 20-min Aida64 CPU test.
At 5.0 GHZ, 1.29 V, I get idle temps around 43 and load around 68-70 (same 20 min test)
Are these temps normal after delidding, or still quite high?
Thanks!

 

[CompuTronix]

Read: Intel Temperature Guide - http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html

 

[vMax]

Just delidded my 8700K using conductonaut on the die and on the inside of the Cpu lid...very thin layer. Yes, I was terrified but turned out easier than I thought.

I am using the Corsair H150i Pro AIO and Temps for 4.9GHz at 1.278v Overclock on Aida64 for a 20 minute run are 59 Degrees C on the hottest core average 55 degrees...The temp drop was immense especially on Prime95 small FFT's as again I am in the 60's degrees instead of 80's degrees C...

My idle temps are 27 to 29/30 degrees C max but I do live in the UK, it can get as low as 24 degrees C in the morning..

Have not tried 5GHz, but will get the courage up soon...

 

[urbancamper]

That is pretty darn good cooling for a really bad air cooler for your cpu. Seems like you hit the lottery with those speeds at such low voltages. I would suggest testing with prime 95 version 26.6. Do the small fft test. The blend test is no proof of stability. It could be your actually stable but most likely not.

Post pictures with your proof.

 

[changsta2004]

I've been running the "Blend" test and things were stable. So I guess it's better to run the Small FFT option?
I will try that and post the results for my both 4.8 and 5.0 OC settings.

 

[changsta2004]

Btw, can you explain the difference between the "Blend" test on Prime95 and the Small FFT test? Also, vs the Aida64 CPU test.
My computer actually immediately blue-screened after trying the Small FFT so I had to up the Voltage at 4.8 to 1.24V. The results were it didn't blue screen but the Cores 4 and 6 actually throttled (CPU temps hit yellow (80's) and CPU Load dropped from 100 to 50's).
It didn't do it when I tested Aida CPU testing though.

 

[changsta2004]

Here is with Aida64 (4.8 GHZ on 1.3V). I only ran it for a few min b/c I already knew the results.
Idle 37 ish, Load, 60's

Here is what happened after using your recommendation with Prime95.

The thermals are definitely higher in all cores. 2 of the cores won't stay stable as well.

Can I get a solution to why this is happening on the Smal FFT test? Do I need to up the voltage?

 

[vMax]

Borrowed from elsewhere....OK so basically you would want to use small FTT's to isolate the cpu when trying to figure out what part of your system is giving you stability issues i.e. not enough vcore because it tests mostly the cpu only without stressing the rest of the system as much... When you go to a test like Blend, it will use a lot of RAM and a combination of Large and Smalle FTT's for a more over all stressing of the CPU & RAM. So if you had a failure using blend test you might not know what caused the fault.

 

[urbancamper]

Borrowed from elsewhere....OK so basically you would want to use small FTT's to isolate the cpu when trying to figure out what part of your system is giving you stability issues i.e. not enough vcore because it tests mostly the cpu only without stressing the rest of the system as much... When you go to a test like Blend, it will use a lot of RAM and a combination of Large and Smalle FTT's for a more over all stressing of the CPU & RAM. So if you had a failure using blend test you might not know what caused the fault.

This answer is only partially right. The OP stated he had failure during small fft testing. The blend test only slightly stresses the cpu. it mostly tests Ram. Also the obvious cause of the system fault is there is not enough voltage on the cpu. It is not a matter of a part causing the failure.

There is not enough voltage to the cpu. You need to increase your voltage to the cpu incrementally while also using LLC(load line calibration) to keep the voltage from dropping down to stock levels when the cpu is at load.

Really though just keep it stock speeds for now. That heatsink you have is not suitable for overclocking. Max has the right idea with that Corsair H150i

 

[vMax]

Borrowed from elsewhere....OK so basically you would want to use small FTT's to isolate the cpu when trying to figure out what part of your system is giving you stability issues i.e. not enough vcore because it tests mostly the cpu only without stressing the rest of the system as much... When you go to a test like Blend, it will use a lot of RAM and a combination of Large and Smalle FTT's for a more over all stressing of the CPU & RAM. So if you had a failure using blend test you might not know what caused the fault.

This answer is only partially right. The OP stated he had failure during small fft testing. The blend test only slightly stresses the cpu. it mostly tests Ram. Also the obvious cause of the system fault is there is not enough voltage on the cpu. It is not a matter of a part causing the failure.

There is not enough voltage to the cpu. You need to increase your voltage to the cpu incrementally while also using LLC(load line calibration) to keep the voltage from dropping down to stock levels when the cpu is at load.

Really though just keep it stock speeds for now. That heatsink you have is not suitable for overclocking. Max has the right idea with that Corsair H150i

Thanks urbancamper...gets confusing sometimes..