[SOLVED] Cooling System Temp

Seabass101

Distinguished
Jan 18, 2016
102
0
18,680
Hi, I delidded my i7 8700k @ 1.240v and have Core @ 4.7GHz/Cache @ 4.7GHz with a Corsair H115i 280mm AIO cooler and was wondering if 75c on Borderlands 3 is normal or high for a delid?
P.S. 60Hz/60FPS 4K low settings, Max temp is 75c
 
Solution
Yeah, that's high.

A delid, with that cooler (IF it's configured properly), while only running a game, not even a thermal stress test, should result in a much lower temp. Especially at only 4.7Ghz which is not bad compared to the stock all core turbo boost, but most people delidding these are probably going for more like 4.9 or 5Ghz all core OC. I'm not saying that is what YOU should be going for, I'm just saying your result is maybe not quite where it ought to be, but certainly is better than the factory configuration.

What did you use for TIM? Did you restore the heat spreader or are you using a direct to die cooling solution?

Where is your radiator installed? Is it configured as an intake or an exhaust configuration? What case...
Yeah, that's high.

A delid, with that cooler (IF it's configured properly), while only running a game, not even a thermal stress test, should result in a much lower temp. Especially at only 4.7Ghz which is not bad compared to the stock all core turbo boost, but most people delidding these are probably going for more like 4.9 or 5Ghz all core OC. I'm not saying that is what YOU should be going for, I'm just saying your result is maybe not quite where it ought to be, but certainly is better than the factory configuration.

What did you use for TIM? Did you restore the heat spreader or are you using a direct to die cooling solution?

Where is your radiator installed? Is it configured as an intake or an exhaust configuration? What case? How many case fans other than the radiator and are those, each of them, configured as intake or exhaust?

Have you run Prime95 Small FFT, with AVX/AVX2 disabled, yet? Games are not a suitable metric for determining thermal compliance. You need a steady state load as that is what complies with the Intel data sheets and is a commonly held standard for thermal testing. Any workload that is not steady state is likely to either exceed or remain below the level of 100% TDP, or both, and that's not what you want to see.

You can use either Prime or OCCT small data set for thermal testing.

 
Solution

Seabass101

Distinguished
Jan 18, 2016
102
0
18,680
Well what I did is i have a Copper IHS and i have it secured by the motherboard and AIO cooler inplace so no adhesive has been used. I used Thermal Grizzly Kyronaut paste for the application on the CPU die underneath and on top of thew Copper IHS in a Peas size method so i know it is evenly spread and not overflowing or to much.
 

Seabass101

Distinguished
Jan 18, 2016
102
0
18,680
Could it be the AIO is the problem because the AIO radiator tubes are facing up and that could cause a little air to get in and cause worse temps? Though I had to do it for space and placement availability in my case.
 

Seabass101

Distinguished
Jan 18, 2016
102
0
18,680
not really, what I mean is that the AIO is mounter at the front of the case with the fans as intake, but the placement of the tubing is upwards vs the normal downwards they should be. So the AIO is mounted front and the tubes are coming from the top of the Radiator if that makes sense

and this could cause air to get trapped and was wondering by how much could performance suffer doing this like that?