Hello Everyone!
I am new to this forum and overclocking in particular. I built a new system few months ago and been playing with OC setting. The system is primarily used for gaming and light office work.
I am trying to determine if my overclock considered "safe" in terms of voltage/power for 24/7 use. It seems to me that temperatures i am observing are too high for custom water loop, compared to those temps I saw across the web.
My system specs:
9900K
Gigabyte Aorus Master Z390
Gskill 16GB 3600MHz@16-17-17-38 1.35V
Ek Velocity CPU block
EK 011 Distro plate with ddc 3.1 pump ( swapped the impeller from 3.25 )
EK PE360 rad with EK Vardar 120 in Push
EK XE 360 rad with Vardars in the push and Arctic p120 in pull ( RPM normalized )
GTX 1070 Strix ( leftover from the previous PC)
I am currently running two profiles: stock "undervolt" and 5.0. Both t passes through P95 no avx 4 hrs, avx 4 hrs, Realbench, IBT 2.54, OCCT all setting, CBR20 and CBR15.
For both profiles, I use a custom fan/pump config. Flow rate is at ~0.5GPM at idle and ~1.4GPM at load, fans speed configured to work based on coolant temp: up to 27C - 500RPM, 28C - 600RPM, 29C- 900RPM, 30C - 1100RPM, 31C+ - 1200 RPM ). My usual ambient temp is between 20C-24C. Highest dT between Ambient/water I saw was 9C after 4 hours of P95 at 1200rpm on the fans
Stock undervolt:
Core: 47x
Ring: 43x
AVX offset 0
Voltage: Normal ( adaptive ), -0.020 offset
LLC: Low
AC/DC Loadline: Power saving
PWM switch rate: 300 MHZ
All C states, power limits, and power saving features are disabled
XMP Ram - 3600@16-17-17-38, 1.35V, VCCIO - 1.15V, VCCSA - 1.2V
The temp test was conducted at 23C - 23.5C ambient temp. At idle it sits around 30-32C across cores and at 1.178V at VRVOUT. before running stress tests, i "heat-soaked" the loop by running prime 95 29.8 with AVX for 1 hour.
OCCT AVX Small Data set:
66 max temp, VRVOUT 1.105V, 136W@120A
P95 29.8 AVX:
76 max temp, VRVOUT 1.109-1.115, 178W@162A
CBR20 (loops):
65 max, VRVOUT 1.117-1.141, 130W@114.5A
It seems pretty hot for stock undervolt. OFC, things get worse with 5.0 OC:
5.0 OC
Core: 50x
Ring: 47x
AVX offset 1
Voltage: Normal ( adaptive ), -0.100 offset
LLC: Low
AC/DC Loadline: auto
PWM switch rate: 300 MHZ
All C states, power limits, and power saving features are disabled
XMP Ram - 3600@16-17-17-38, 1.35V, VCCIO - 1.15V, VCCSA - 1.2V
OCCT NO AVX Small Data set:
83 max temp, VRVOUT 1.252V, 190W@152A
P95 29.8 AVX:
88 max temp, VRVOUT 1.207, 224W@190A
CBR20 (loops):
81 max, VRVOUT 1.262, 178W@142A
Same ambient of 23-23.5C, the cores at idle sit at 34-35C. VRVOUT is 1.307-1.315. there is a spike up to 1.35V when the load changes ( ex. when I stop P95 )
I also find odd that with increased load the gap between hottest and coolest core goes up to 10C - Core 1 and 7 being the coolest and Core 2 and 4 the hottest.
I tried to re-seat the CPU block multiple times. The TIM used is Cryonaut. below are the screenshots from HWinfo for all the above-mentioned tests. EC_TEMP2 is the coolant temperature - I have a sensor in the bottom radiator where water exiting back to the res.
Attached screenshots of HWinfo for all the tests I did. ThankS~!
I am new to this forum and overclocking in particular. I built a new system few months ago and been playing with OC setting. The system is primarily used for gaming and light office work.
I am trying to determine if my overclock considered "safe" in terms of voltage/power for 24/7 use. It seems to me that temperatures i am observing are too high for custom water loop, compared to those temps I saw across the web.
My system specs:
9900K
Gigabyte Aorus Master Z390
Gskill 16GB 3600MHz@16-17-17-38 1.35V
Ek Velocity CPU block
EK 011 Distro plate with ddc 3.1 pump ( swapped the impeller from 3.25 )
EK PE360 rad with EK Vardar 120 in Push
EK XE 360 rad with Vardars in the push and Arctic p120 in pull ( RPM normalized )
GTX 1070 Strix ( leftover from the previous PC)
I am currently running two profiles: stock "undervolt" and 5.0. Both t passes through P95 no avx 4 hrs, avx 4 hrs, Realbench, IBT 2.54, OCCT all setting, CBR20 and CBR15.
For both profiles, I use a custom fan/pump config. Flow rate is at ~0.5GPM at idle and ~1.4GPM at load, fans speed configured to work based on coolant temp: up to 27C - 500RPM, 28C - 600RPM, 29C- 900RPM, 30C - 1100RPM, 31C+ - 1200 RPM ). My usual ambient temp is between 20C-24C. Highest dT between Ambient/water I saw was 9C after 4 hours of P95 at 1200rpm on the fans
Stock undervolt:
Core: 47x
Ring: 43x
AVX offset 0
Voltage: Normal ( adaptive ), -0.020 offset
LLC: Low
AC/DC Loadline: Power saving
PWM switch rate: 300 MHZ
All C states, power limits, and power saving features are disabled
XMP Ram - 3600@16-17-17-38, 1.35V, VCCIO - 1.15V, VCCSA - 1.2V
The temp test was conducted at 23C - 23.5C ambient temp. At idle it sits around 30-32C across cores and at 1.178V at VRVOUT. before running stress tests, i "heat-soaked" the loop by running prime 95 29.8 with AVX for 1 hour.
OCCT AVX Small Data set:
66 max temp, VRVOUT 1.105V, 136W@120A
P95 29.8 AVX:
76 max temp, VRVOUT 1.109-1.115, 178W@162A
CBR20 (loops):
65 max, VRVOUT 1.117-1.141, 130W@114.5A
It seems pretty hot for stock undervolt. OFC, things get worse with 5.0 OC:
5.0 OC
Core: 50x
Ring: 47x
AVX offset 1
Voltage: Normal ( adaptive ), -0.100 offset
LLC: Low
AC/DC Loadline: auto
PWM switch rate: 300 MHZ
All C states, power limits, and power saving features are disabled
XMP Ram - 3600@16-17-17-38, 1.35V, VCCIO - 1.15V, VCCSA - 1.2V
OCCT NO AVX Small Data set:
83 max temp, VRVOUT 1.252V, 190W@152A
P95 29.8 AVX:
88 max temp, VRVOUT 1.207, 224W@190A
CBR20 (loops):
81 max, VRVOUT 1.262, 178W@142A
Same ambient of 23-23.5C, the cores at idle sit at 34-35C. VRVOUT is 1.307-1.315. there is a spike up to 1.35V when the load changes ( ex. when I stop P95 )
I also find odd that with increased load the gap between hottest and coolest core goes up to 10C - Core 1 and 7 being the coolest and Core 2 and 4 the hottest.
I tried to re-seat the CPU block multiple times. The TIM used is Cryonaut. below are the screenshots from HWinfo for all the above-mentioned tests. EC_TEMP2 is the coolant temperature - I have a sensor in the bottom radiator where water exiting back to the res.
Attached screenshots of HWinfo for all the tests I did. ThankS~!