F5F Stay Refreshed Power Users Overclocking i5 4670k 4.4ghz

i5 4670k 4.4ghz

i5 4670k 4.4ghz

J
Juan2610
Posting Freak
875
03-14-2016, 05:45 AM
#1
i'm currently overclocking my i5 4670k, reaching 4.4ghz with 75-80c under full load.
the setup includes:
i5 4670k at 4.4ghz + Prolimatech Panther CPU cooler
z87n-wifi motherboard
1060 gtx 6gb
2x4gb 1866mhz RAM
GS600w PSU

my current OC parameters are:
clock ratio: 44
VCore: 1.27v
voltage set manually to stay below 1.27v
i've tested 4.5ghz and experienced BSOD during stress tests.
adjusting from 1.27v to 1.3v with small steps (0.05) didn't cause issues.
boots normally, idle and load temps remain stable between 4.2ghz-4.5ghz and 1.27v-1.3v.
maybe another BIOS setting is missing to boost performance further?

temperatures are around 45-50c at idle and 75-80c under load.
i'm just starting out with OCing, any guidance would be greatly appreciated!
J
Juan2610
03-14-2016, 05:45 AM #1

i'm currently overclocking my i5 4670k, reaching 4.4ghz with 75-80c under full load.
the setup includes:
i5 4670k at 4.4ghz + Prolimatech Panther CPU cooler
z87n-wifi motherboard
1060 gtx 6gb
2x4gb 1866mhz RAM
GS600w PSU

my current OC parameters are:
clock ratio: 44
VCore: 1.27v
voltage set manually to stay below 1.27v
i've tested 4.5ghz and experienced BSOD during stress tests.
adjusting from 1.27v to 1.3v with small steps (0.05) didn't cause issues.
boots normally, idle and load temps remain stable between 4.2ghz-4.5ghz and 1.27v-1.3v.
maybe another BIOS setting is missing to boost performance further?

temperatures are around 45-50c at idle and 75-80c under load.
i'm just starting out with OCing, any guidance would be greatly appreciated!

C
Cherrie24
Member
179
03-14-2016, 06:48 AM
#2
The difference between idle and load temperatures feels quite small to me. How are you measuring your "load" temperatures? With higher voltage, your readings should rise. A 0.03V change should noticeably affect a few degrees. Regarding ways to boost your output, it's limited. More voltage could help, but success depends on the silicon quality and isn't guaranteed. You might need a larger voltage increase to reach higher frequencies, which doesn't seem practical. For your current output of 4.4GHz (+600MHz over Boost), it seems reasonable, especially considering the cooler setup you're using. It appears that achieving higher outputs with standard cooling methods was previously limited.
C
Cherrie24
03-14-2016, 06:48 AM #2

The difference between idle and load temperatures feels quite small to me. How are you measuring your "load" temperatures? With higher voltage, your readings should rise. A 0.03V change should noticeably affect a few degrees. Regarding ways to boost your output, it's limited. More voltage could help, but success depends on the silicon quality and isn't guaranteed. You might need a larger voltage increase to reach higher frequencies, which doesn't seem practical. For your current output of 4.4GHz (+600MHz over Boost), it seems reasonable, especially considering the cooler setup you're using. It appears that achieving higher outputs with standard cooling methods was previously limited.

B
ByrRoZz
Member
175
03-14-2016, 11:21 PM
#3
The difference between idle and load temperatures seems quite small to me. How do you measure your "load" temperatures? With higher voltage, your readings should rise. A 0.03V change usually affects a few degrees. Regarding ways to boost your output, it’s limited. More voltage could help, but success depends on the silicon quality and isn’t guaranteed. You might need a larger jump like +0.1V to reach higher frequencies, which doesn’t seem practical. Concerning your current output at 4.4GHz (+600MHz over Boost), it seems reasonable. With better cooling, you’re likely limited to around 4.4-4.6GHz and 1.25-1.3V. It’s unfortunate you’re on the lower side, but that’s why it’s often described as a lottery. Your temperatures are approaching the upper limit for comfortable long-term operation, and your voltage of 1.3V is near the maximum recommended for air cooling.
B
ByrRoZz
03-14-2016, 11:21 PM #3

The difference between idle and load temperatures seems quite small to me. How do you measure your "load" temperatures? With higher voltage, your readings should rise. A 0.03V change usually affects a few degrees. Regarding ways to boost your output, it’s limited. More voltage could help, but success depends on the silicon quality and isn’t guaranteed. You might need a larger jump like +0.1V to reach higher frequencies, which doesn’t seem practical. Concerning your current output at 4.4GHz (+600MHz over Boost), it seems reasonable. With better cooling, you’re likely limited to around 4.4-4.6GHz and 1.25-1.3V. It’s unfortunate you’re on the lower side, but that’s why it’s often described as a lottery. Your temperatures are approaching the upper limit for comfortable long-term operation, and your voltage of 1.3V is near the maximum recommended for air cooling.

S
SeaOwner
Junior Member
5
03-15-2016, 06:22 AM
#4
Review the BIOS Advanced CPU Core configurations and Voltage Tab in this video.
S
SeaOwner
03-15-2016, 06:22 AM #4

Review the BIOS Advanced CPU Core configurations and Voltage Tab in this video.

M
MettaloCaft
Senior Member
396
03-15-2016, 08:01 AM
#5
Based on your cooling system, you may reach up to 1.35v.
M
MettaloCaft
03-15-2016, 08:01 AM #5

Based on your cooling system, you may reach up to 1.35v.

B
220
03-15-2016, 01:23 PM
#6
Barty1884:
The difference between idle and load temperatures seems quite small to me. How are you measuring your "load" temperatures? With higher voltage, you should notice a rise in temperature. A 0.03V change should definitely affect a few degrees... As for methods to boost your output, it’s limited. More voltage might help, but it’s all about luck with the silicon and there are no guarantees. You could end up needing a whole bit higher, like +0.1V, which wouldn’t make much sense. Regarding your current output, 4.4GHz (+600MHz over Boost) seems reasonable. Considering the cooler you’re using, it probably won’t be much more than that. It appears that people achieving higher outputs with standard cooling typically stayed in the 4.4-4.6GHz range, around 1.25-1.3V. Although it’s frustrating to be at the lower side, that’s why it’s often called a lottery. Your temperatures are nearing the upper limit for comfortable long-term operation, and your voltage (1.3V) is close to the maximum you’d want on air. Thanks! I was searching for a real answer and you’re almost correct—I’m stable at 4.4GHz, 1.26 at idle (previously around 45-50°C), 75/80 under full load, with a max of about 76°C. As you mentioned, there could be a significant jump from voltage to the next stable frequency. I reached 1.3V but wasn’t stable at 4.5, so I reduced it back to 1.26v.
B
BlueSkyHorizon
03-15-2016, 01:23 PM #6

Barty1884:
The difference between idle and load temperatures seems quite small to me. How are you measuring your "load" temperatures? With higher voltage, you should notice a rise in temperature. A 0.03V change should definitely affect a few degrees... As for methods to boost your output, it’s limited. More voltage might help, but it’s all about luck with the silicon and there are no guarantees. You could end up needing a whole bit higher, like +0.1V, which wouldn’t make much sense. Regarding your current output, 4.4GHz (+600MHz over Boost) seems reasonable. Considering the cooler you’re using, it probably won’t be much more than that. It appears that people achieving higher outputs with standard cooling typically stayed in the 4.4-4.6GHz range, around 1.25-1.3V. Although it’s frustrating to be at the lower side, that’s why it’s often called a lottery. Your temperatures are nearing the upper limit for comfortable long-term operation, and your voltage (1.3V) is close to the maximum you’d want on air. Thanks! I was searching for a real answer and you’re almost correct—I’m stable at 4.4GHz, 1.26 at idle (previously around 45-50°C), 75/80 under full load, with a max of about 76°C. As you mentioned, there could be a significant jump from voltage to the next stable frequency. I reached 1.3V but wasn’t stable at 4.5, so I reduced it back to 1.26v.