F5F Stay Refreshed Power Users Overclocking i7-6700k with a 4.6ghz clock speed and 1.345v voltage.

i7-6700k with a 4.6ghz clock speed and 1.345v voltage.

i7-6700k with a 4.6ghz clock speed and 1.345v voltage.

Pages (3): 1 2 3 Next
M
meke58
Junior Member
5
10-13-2016, 03:58 PM
#1
Hello,
This is my second message about Overclocking. I'm new to it and excited to learn more, but I'm concerned about whether my setup is safe and won't cause any issues. About a month ago, I completed my first gaming build and it performed quite well. Now I decided to experiment with overclocking on my rig. I managed to push my CPU to 4.6ghz at 1.345v, but when using AIDA64, the Core Voltage stabilized at 1.376v. I've seen videos suggesting that Intel Skylake safe voltage is around 1.45v and others warn not to exceed 1.35v—what should I really follow?

I'm seeking your advice or thoughts on the specifications I achieved. Any feedback would be very helpful. In the meantime, I saved a screenshot of my stress test results:
http://i.imgur.com/QFhV3kE.png

Here are my PC components:
- EVGA SuperNova 750G2
- ASUS ROG Maximus VIII Ranger
- i7-6700k 4.0ghz
- 16gb RAM G.Skill Ripjaw V (3000mhz)
- Corsair H100i v2
- 500GB Samsung SSD 850 EVO
- 1TB Seagate Barracuda
- ASUS GeForce GTX 1060 6GB Turbo

Finished End:
http://i.imgur.com/DHoINw1.jpg
http://i.imgur.com/MvfbnCw.jpg

Thank you for your help, and please forgive my attempts at cable management—I've tried my best! Lol

Also, could you let me know if the temperature underload is acceptable? I'm not sure whether I should focus on individual CPU core temperatures or just check the overall max temperature. Appreciate your response.
M
meke58
10-13-2016, 03:58 PM #1

Hello,
This is my second message about Overclocking. I'm new to it and excited to learn more, but I'm concerned about whether my setup is safe and won't cause any issues. About a month ago, I completed my first gaming build and it performed quite well. Now I decided to experiment with overclocking on my rig. I managed to push my CPU to 4.6ghz at 1.345v, but when using AIDA64, the Core Voltage stabilized at 1.376v. I've seen videos suggesting that Intel Skylake safe voltage is around 1.45v and others warn not to exceed 1.35v—what should I really follow?

I'm seeking your advice or thoughts on the specifications I achieved. Any feedback would be very helpful. In the meantime, I saved a screenshot of my stress test results:
http://i.imgur.com/QFhV3kE.png

Here are my PC components:
- EVGA SuperNova 750G2
- ASUS ROG Maximus VIII Ranger
- i7-6700k 4.0ghz
- 16gb RAM G.Skill Ripjaw V (3000mhz)
- Corsair H100i v2
- 500GB Samsung SSD 850 EVO
- 1TB Seagate Barracuda
- ASUS GeForce GTX 1060 6GB Turbo

Finished End:
http://i.imgur.com/DHoINw1.jpg
http://i.imgur.com/MvfbnCw.jpg

Thank you for your help, and please forgive my attempts at cable management—I've tried my best! Lol

Also, could you let me know if the temperature underload is acceptable? I'm not sure whether I should focus on individual CPU core temperatures or just check the overall max temperature. Appreciate your response.

W
Wellxam56
Member
61
10-20-2016, 01:05 AM
#2
Very similar configuration to the one I just completed on Saturday. Great work. Regarding your temperatures, please review each core to ensure they remain consistent throughout. After a couple of hours gaming, open each core and monitor the peak temperatures. Those who suggest staying below 1.35v are correct for continuous use; I've adjusted my Vityal memory size to match my RAM capacity. In Local Security Policy, I cleared my page file during shutdown. I also power off my PC almost every day. In my opinion, 1.4 is acceptable. I really need to dig through reported chip degradation cases. With stable temperatures, I don't mind what others say—1.4 vcore won't harm your 2015 Intel SkyLake CPU. Just make sure before upgrading.
W
Wellxam56
10-20-2016, 01:05 AM #2

Very similar configuration to the one I just completed on Saturday. Great work. Regarding your temperatures, please review each core to ensure they remain consistent throughout. After a couple of hours gaming, open each core and monitor the peak temperatures. Those who suggest staying below 1.35v are correct for continuous use; I've adjusted my Vityal memory size to match my RAM capacity. In Local Security Policy, I cleared my page file during shutdown. I also power off my PC almost every day. In my opinion, 1.4 is acceptable. I really need to dig through reported chip degradation cases. With stable temperatures, I don't mind what others say—1.4 vcore won't harm your 2015 Intel SkyLake CPU. Just make sure before upgrading.

C
Catsstate
Member
146
10-20-2016, 09:12 AM
#3
The setup is almost identical to the one I completed last Saturday. Great work overall. Regarding your temperatures, it’s important to inspect each core to ensure they’re consistent. After a couple of hours gaming, open each core and verify the peak temperatures. Those who advise keeping temps below 1.35v are correct for continuous use; I’ve adjusted my Vityal memory size to match my RAM and configured local security policy to clear the page file during shutdown. I also power off my PC almost every day. In my opinion, a 1.4 vcore is acceptable. I need to thoroughly research chip degradation reports before making any changes. With stable temperatures, I don’t mind what others say—1.4 vcore won’t harm your 2015 Intel SkyLake CPU. Just make sure you upgrade eventually.

If upgrading isn’t an option, I plan to keep my CPU running for about five years or more. Would that last me until then?

I do shut down my PC daily, especially since I have an SSD that boots in just 30 seconds. Sadly, I managed to stabilize it around 4.5ghz at 1.345v, but I’m now pushing it to 4.6ghz. I ran some stress tests and set the voltage to about 1.36v, which seems stable on average (around 1.38v max). Current temps are roughly: CPU max 58°C, average 51.9°C, C2 max 71°C, C3 max 58°C, C4 max 68°C, averaging around 54°C. Are these numbers satisfactory?
C
Catsstate
10-20-2016, 09:12 AM #3

The setup is almost identical to the one I completed last Saturday. Great work overall. Regarding your temperatures, it’s important to inspect each core to ensure they’re consistent. After a couple of hours gaming, open each core and verify the peak temperatures. Those who advise keeping temps below 1.35v are correct for continuous use; I’ve adjusted my Vityal memory size to match my RAM and configured local security policy to clear the page file during shutdown. I also power off my PC almost every day. In my opinion, a 1.4 vcore is acceptable. I need to thoroughly research chip degradation reports before making any changes. With stable temperatures, I don’t mind what others say—1.4 vcore won’t harm your 2015 Intel SkyLake CPU. Just make sure you upgrade eventually.

If upgrading isn’t an option, I plan to keep my CPU running for about five years or more. Would that last me until then?

I do shut down my PC daily, especially since I have an SSD that boots in just 30 seconds. Sadly, I managed to stabilize it around 4.5ghz at 1.345v, but I’m now pushing it to 4.6ghz. I ran some stress tests and set the voltage to about 1.36v, which seems stable on average (around 1.38v max). Current temps are roughly: CPU max 58°C, average 51.9°C, C2 max 71°C, C3 max 58°C, C4 max 68°C, averaging around 54°C. Are these numbers satisfactory?

Z
ZakenMannetje
Junior Member
46
10-20-2016, 10:33 AM
#4
Hey there, remember talking about AIDA64 and ASUS realbench? After a few hours of testing, both passed, but I ended up hitting the BSOD. I adjusted to 4.5ghz with the same voltage, hoping it would stabilize. If it does, I’ll try again and boost it to 4.6. Quick question though—what’s considered enough time for a stress test? Some suggest running it for 1 to 2 days, up to 3 or even 24 hours.
Z
ZakenMannetje
10-20-2016, 10:33 AM #4

Hey there, remember talking about AIDA64 and ASUS realbench? After a few hours of testing, both passed, but I ended up hitting the BSOD. I adjusted to 4.5ghz with the same voltage, hoping it would stabilize. If it does, I’ll try again and boost it to 4.6. Quick question though—what’s considered enough time for a stress test? Some suggest running it for 1 to 2 days, up to 3 or even 24 hours.

D
DorianTib
Junior Member
7
10-27-2016, 09:45 AM
#5
I think it comes down to what you're testing with. My last rig used a prime 95 blend for two hours and it worked fine—never hit BSOD. That's my experience. Every time I crashed with prime, it happened within the first 20 minutes. After an hour or two, I realized it was stable enough in real use and never caused problems. But I'm sure the first person reading this will think I did too many drugs when I was younger lol
D
DorianTib
10-27-2016, 09:45 AM #5

I think it comes down to what you're testing with. My last rig used a prime 95 blend for two hours and it worked fine—never hit BSOD. That's my experience. Every time I crashed with prime, it happened within the first 20 minutes. After an hour or two, I realized it was stable enough in real use and never caused problems. But I'm sure the first person reading this will think I did too many drugs when I was younger lol

R
raven_sing
Junior Member
48
10-27-2016, 03:11 PM
#6
Running at 4.6GHz with a voltage of 1.345V demands a capable chip, http://cdn.overclock.net/f/f3/f395401b_Udntitled.png. The typical average is close to 1.35. Some sources suggest it's about 1.35 plus or minus 0.03v, but if you want higher stability, I opted for a lower voltage of 1.34v instead.
R
raven_sing
10-27-2016, 03:11 PM #6

Running at 4.6GHz with a voltage of 1.345V demands a capable chip, http://cdn.overclock.net/f/f3/f395401b_Udntitled.png. The typical average is close to 1.35. Some sources suggest it's about 1.35 plus or minus 0.03v, but if you want higher stability, I opted for a lower voltage of 1.34v instead.

P
Pickmaster12
Senior Member
710
10-28-2016, 11:23 AM
#7
Rain2o would say it varies based on the testing setup. My previous rig used a prime 95 blend for two hours without any problems, and I never experienced BSOD. That was during the first 20 minutes. After extending to an hour or more, it seemed stable enough in real-world use. But I know someone will think I did too much before I was that young.
P
Pickmaster12
10-28-2016, 11:23 AM #7

Rain2o would say it varies based on the testing setup. My previous rig used a prime 95 blend for two hours without any problems, and I never experienced BSOD. That was during the first 20 minutes. After extending to an hour or more, it seemed stable enough in real-world use. But I know someone will think I did too much before I was that young.

H
hahahalo21
Junior Member
46
11-05-2016, 06:57 AM
#8
Thank you for your reply, yes I've seen and read that post. Even when I manually set the core voltage to 1.345, the voltage still jumps up to 1.376—does that seem normal? Should I check the number I entered or the one displayed where the system stopped? Also, what value did you put on your multiplexer? I believe the mines run at 46x. For now, I'm setting it to 4.5ghz and kept the same voltage. I let it run for about three hours last night using AIDA64 and the system appears stable. I’m not sure about in-game time yet.
H
hahahalo21
11-05-2016, 06:57 AM #8

Thank you for your reply, yes I've seen and read that post. Even when I manually set the core voltage to 1.345, the voltage still jumps up to 1.376—does that seem normal? Should I check the number I entered or the one displayed where the system stopped? Also, what value did you put on your multiplexer? I believe the mines run at 46x. For now, I'm setting it to 4.5ghz and kept the same voltage. I let it run for about three hours last night using AIDA64 and the system appears stable. I’m not sure about in-game time yet.

X
xXJay_BugXx
Senior Member
559
11-21-2016, 06:06 PM
#9
Thank you for your reply, I've reviewed the post. Even when manually adjusting the core voltage to 1.345, the boost still reaches 1.376, which is typical. Should I refer to the number I entered or the one displayed where the system stopped? Also, what value did you set on your multiplexer? I believe mine is at 46x. Currently, I'm setting it to 4.5ghz while keeping the same voltage. It ran for about three hours last night using AIDA64 and the system appears stable, though in-game time is uncertain.

It's normal because load line calibration is active. It's better to set a fixed level rather than auto. A higher LLC setting will provide more voltage to counteract voltage drop. It seems your 6700k and mine are from the same batch or similar batches. You just have an average or unfortunate 6700k that struggles to overclock well at lower voltages due to silicon variation.
X
xXJay_BugXx
11-21-2016, 06:06 PM #9

Thank you for your reply, I've reviewed the post. Even when manually adjusting the core voltage to 1.345, the boost still reaches 1.376, which is typical. Should I refer to the number I entered or the one displayed where the system stopped? Also, what value did you set on your multiplexer? I believe mine is at 46x. Currently, I'm setting it to 4.5ghz while keeping the same voltage. It ran for about three hours last night using AIDA64 and the system appears stable, though in-game time is uncertain.

It's normal because load line calibration is active. It's better to set a fixed level rather than auto. A higher LLC setting will provide more voltage to counteract voltage drop. It seems your 6700k and mine are from the same batch or similar batches. You just have an average or unfortunate 6700k that struggles to overclock well at lower voltages due to silicon variation.

M
mumustrak
Senior Member
729
11-23-2016, 03:26 AM
#10
Here’s a revised version of your text:

The response from olrac_86 was appreciated. I’ve already seen and read that post. Even when manually adjusting the core voltage to 1.345, the voltage still rises to 1.376—this is normal? Which value should I use when I manually entered it, or which appears where the system stabilized? Also, what setting did you apply on your multiplier? I believe mine runs at 46x.

Currently, I’m lowering the frequency to 4.5ghz while keeping the same voltage. I let it run for about three hours last night using AIDA64, and the system seems stable. However, I’m not sure about in-game performance.

I successfully achieved a stable 4.5GHz overclock at a 1.34 vcore in BIOS. Usually, you should specify both or clearly state what you mean.

You can find this information on the link provided:
https://docs.google.com/spreadsheets/d/1...edit#gid=0

The table there shows both the BIOS Vcore and the CPU-Z or similar tool’s Vcore readings. The only other option I have in mind is HWmonitor.

It seems there’s a difference between what you set and what you actually measure. If you really want, most motherboards (at least on my system) allow you to accurately measure the Vcore with a multimeter.

When measuring voltage, I recommend using CPU-z first—though I’m not sure if AIDA uses the same driver. For general readings, it’s better to check the actual Vcore while the program is running, which will be more precise. This value should be closer to what you set in BIOS.

Additionally, are you using Manual, Adaptive, or Offset settings?
M
mumustrak
11-23-2016, 03:26 AM #10

Here’s a revised version of your text:

The response from olrac_86 was appreciated. I’ve already seen and read that post. Even when manually adjusting the core voltage to 1.345, the voltage still rises to 1.376—this is normal? Which value should I use when I manually entered it, or which appears where the system stabilized? Also, what setting did you apply on your multiplier? I believe mine runs at 46x.

Currently, I’m lowering the frequency to 4.5ghz while keeping the same voltage. I let it run for about three hours last night using AIDA64, and the system seems stable. However, I’m not sure about in-game performance.

I successfully achieved a stable 4.5GHz overclock at a 1.34 vcore in BIOS. Usually, you should specify both or clearly state what you mean.

You can find this information on the link provided:
https://docs.google.com/spreadsheets/d/1...edit#gid=0

The table there shows both the BIOS Vcore and the CPU-Z or similar tool’s Vcore readings. The only other option I have in mind is HWmonitor.

It seems there’s a difference between what you set and what you actually measure. If you really want, most motherboards (at least on my system) allow you to accurately measure the Vcore with a multimeter.

When measuring voltage, I recommend using CPU-z first—though I’m not sure if AIDA uses the same driver. For general readings, it’s better to check the actual Vcore while the program is running, which will be more precise. This value should be closer to what you set in BIOS.

Additionally, are you using Manual, Adaptive, or Offset settings?

Pages (3): 1 2 3 Next