Welcome Guest ( Log In | Register )

Bump TopicReply to this topicRSS feed Start new topic Start Poll

Outline · [ Standard ] · Linear+

> Skylake/ Kaby Lake OverClocking V1 | Z170/Z270, Show off your Skylake/ Kaby Lake OC here

cattscratch
post Jun 23 2017, 06:50 AM

New Member
*
Group: Junior Member
Posts: 33

Joined: Aug 2012
QUOTE(owikh84 @ Jun 17 2017, 09:30 PM)
Sounds like insufficient voltage issue, specifically the System Agent voltage (VCCSA).
After applying XMP, try increasing it a bit like 1.25-1.35v (depending on strength of your CPU's integrated memory controller AKA IMC).
Weak IMC means you will need to pump more VCCSA.
*
I gonna try this...
1. Press F5 key while in BIOS to load optimized defaults
2. Press F10 key then OK to save and reboot
3. Choose XMP Ai Overclock Tuner setting while in BIOS
4. Move down to "DRAM Voltage" field then type 1.37 <ENTER>
5. Move to next field "CPU VCCIO Voltage" then type 1.20 <ENTER>
6. Move to next field "CPU System Agent Voltage" then type 1.20 <ENTER>
7. Press F10 key then OK to save and reboot

Repeat steps 4, 5 and 6 above using just the (+) key to increase voltage one press at a time if you still have issues.


Btw anyway to know if the RAM overclocking is stable ? because 4/5 times it will boot successfully, but 1/5 will fail.
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Jun 24 2017, 08:17 AM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(cattscratch @ Jun 23 2017, 06:50 AM)
Btw anyway to know if the RAM overclocking is stable ? because 4/5 times it will boot successfully, but 1/5 will fail.
*
I use Hyper Pi 32m or Memtest-HCI: http://hcidesign.com/memtest/

For Hyper Pi, pretty much straightforward. Just run the 32m test with full core/thread

For Memtest, use one instance for each CPU thread.
At least 500% no error
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
acther
post Aug 4 2017, 01:55 AM

Casual
***
Group: Junior Member
Posts: 314

Joined: Jan 2009
Yo anyone here interested in a 4266mhz 8gbx2 trident z rgb kit?
brand new set from rma just got it yesterday.
im changing to x299 so not in use anymore

interested can PM me. Letting go cheap

This post has been edited by acther: Aug 4 2017, 01:55 AM
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
Skylinestar
post Aug 10 2017, 01:11 AM

Mega Duck
*******
Group: Senior Member
Posts: 7,409

Joined: Jan 2003
From: Sarawak
anyone here doing delidding service?
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Aug 11 2017, 08:45 PM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(Skylinestar @ Aug 10 2017, 01:11 AM)
anyone here doing delidding service?
*
delidding service
If you have the courage to do it by yourself, can consider buying this Delid kit

Good luck! icon_rolleyes.gif
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
ah_khoo
post Aug 11 2017, 11:46 PM

- No Action Talk Only -
Group Icon
Overclockers United
Group: Elite
Posts: 7,992

Joined: Jan 2003
From: Tanah Lot @ Dungun, Terengganu.


actually i have a used rock it delid kit (relid kit included) & a brand new liquid ultra to let go, anyone keen? back to amd so it's collecting dust now. tongue.gif
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Aug 12 2017, 09:29 AM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(ah_khoo @ Aug 11 2017, 11:46 PM)
actually i have a used rock it delid kit (relid kit included) & a brand new liquid ultra to let go, anyone keen? back to amd so it's collecting dust now.  tongue.gif
*
Wow didn't know that you have a delid tool bro!
Actually I also have one Rockit as well, haven't used it yet.
I think I will not delid my 7700K, resell and buy 8700K instead bwahhaha biggrin.gif
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
azilazwa
post Sep 13 2017, 09:30 AM

Enthusiast
*****
Group: Senior Member
Posts: 830

Joined: May 2008
From: Setiu, Terengganu
Anyone still active ocing skylake? Needed some help here.

I just setup 6700k with Asrock 170M Pro4. But it seems that i cant oc it to 4.5ghz. I have set the multiplier to 45 in bios, voltage at 1.35v. But in windows, it still said my proc is at 4.0ghz. Or am i doing it wrong? This is my first time ocing skylake.

Set multiplier at 45. Other setting untouched.
Attached Image


Manually set voltage to 1.35v.
Attached Image


Other setting untouched
Attached Image


And, what whats bothering me is the temp. In bios the temp already reach 64c. But, in windows i check with HWMonitor and Speccy, the temp is around 38c during idle.
Attached Image


Booted in windows
Attached Image

This post has been edited by azilazwa: Sep 13 2017, 09:33 AM
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
goldfries
post Sep 13 2017, 11:59 AM

40K Club
Group Icon
Group: Forum Admin
Posts: 43,240

Joined: Jan 2003
From: www.goldfries.com




Do a CPU LOAD test and see whether CPU-Z shows 4.5Ghz.
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
azilazwa
post Sep 13 2017, 03:10 PM

Enthusiast
*****
Group: Senior Member
Posts: 830

Joined: May 2008
From: Setiu, Terengganu
QUOTE(goldfries @ Sep 13 2017, 11:59 AM)
Do a CPU LOAD test and see whether CPU-Z shows 4.5Ghz.
*
Thanks for the reply. I run stress test on cpuz. Within 5 minutes of testing, it crashed. Wonder is it bcoz of the temp?

Attached Image

I'm using custom loop with 360mm rad. GPU temp seems quite cool. Probably i didnt tighten the wb enough? Or the temp is not accurate? hmm.gif

Oh btw, the proc is running at 4.5ghz rite? Coz confuse with one showing the correct speed @_@

This post has been edited by azilazwa: Sep 13 2017, 03:11 PM
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Sep 13 2017, 10:33 PM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(azilazwa @ Sep 13 2017, 03:10 PM)
Thanks for the reply. I run stress test on cpuz. Within 5 minutes of testing, it crashed. Wonder is it bcoz of the temp?

Attached Image

I'm using custom loop with 360mm rad. GPU temp seems quite cool. Probably i didnt tighten the wb enough? Or the temp is not accurate?  hmm.gif

Oh btw, the proc is running at 4.5ghz rite? Coz confuse with one showing the correct speed @_@
*
To be able to see your CPU operating at the desired frequency (4.5GHz in your case), change your power plan to "High Performance":
Attached Image

Use CPU-Z to to check your CPU frequency.
Use Core Temp to monitor your CPU core temperatures.
Alternatively, HWInfo64 can be used to monitor both CPU speed and temperatures etc.
Download links available in the top post of this thread.

System crashing has many possibilities, check the BSOD message.
It could be due to overheat, wrong BIOS settings or insufficient voltages (vCore, vDIMM, VCCIO, VCCSA etc).
Most probably is insufficient vCore that feeds voltage to the CPU.
I saw your're using Load-line Calibration (LLC) level 4, which could cause too much vDroop (load vCore is too low). Perhaps you should try the lower level to reduce vDroop, best would be level 1 (minimum vDroop).
If LLC level 1 still crashing, then you should increase your vCore, notch by notch until your system is fully stable.

This post has been edited by owikh84: Sep 13 2017, 10:33 PM
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
azilazwa
post Sep 14 2017, 12:01 AM

Enthusiast
*****
Group: Senior Member
Posts: 830

Joined: May 2008
From: Setiu, Terengganu
QUOTE(owikh84 @ Sep 13 2017, 10:33 PM)
To be able to see your CPU operating at the desired frequency (4.5GHz in your case), change your power plan to "High Performance":
Attached Image

Use CPU-Z to to check your CPU frequency.
Use Core Temp to monitor your CPU core temperatures.
Alternatively, HWInfo64 can be used to monitor both CPU speed and temperatures etc.
Download links available in the top post of this thread.

System crashing has many possibilities, check the BSOD message.
It could be due to overheat, wrong BIOS settings or insufficient voltages (vCore, vDIMM, VCCIO, VCCSA etc).
Most probably is insufficient vCore that feeds voltage to the CPU.
I saw your're using Load-line Calibration (LLC) level 4, which could cause too much vDroop (load vCore is too low). Perhaps you should try the lower level to reduce vDroop, best would be level 1 (minimum vDroop).
If LLC level 1 still crashing, then you should increase your vCore, notch by notch until your system is fully stable.
*
Ahhh, now i can see the clock speed at 4.5ghz after changing the power plan. Thanks for the tip. I also manage to stress the cpu (using cpuz only) for about 10 minutes with temp around 71-73c. Didnt proceed to with other test as i'm a bit worried bcoz my psu is only bronze rated (evga 500w). Will do a proper test when i managed to upgrade the psu.
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Sep 14 2017, 12:24 AM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(azilazwa @ Sep 14 2017, 12:01 AM)
Ahhh, now i can see the clock speed at 4.5ghz after changing the power plan. Thanks for the tip. I also manage to stress the cpu (using cpuz only) for about 10 minutes with temp around 71-73c. Didnt proceed to with other test as i'm a bit worried bcoz my psu is only bronze rated (evga 500w). Will do a proper test when i managed to upgrade the psu.
*
Easy right? Haha you're welcome bro biggrin.gif
Btw, CPU-Z is not a CPU stressing tool, not even a good software for benchmark.
Run Prime95, LinX, OCCT (these 3 very stressful and too hot), AIDA64 Stability Test (less stressful), XTU, Realbench, etc.
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
azilazwa
post Sep 14 2017, 11:03 AM

Enthusiast
*****
Group: Senior Member
Posts: 830

Joined: May 2008
From: Setiu, Terengganu
QUOTE(owikh84 @ Sep 14 2017, 12:24 AM)
Easy right? Haha you're welcome bro biggrin.gif
Btw, CPU-Z is not a CPU stressing tool, not even a good software for benchmark.
Run Prime95, LinX, OCCT (these 3 very stressful and too hot), AIDA64 Stability Test (less stressful), XTU, Realbench, etc.
*
Quite easy playing with multiplier only. But, havent found sweet spot yet.

A little update. So, i changed my psu to Leadex 650w silver and running Prime95 at stock clock just to test whats the temp at full load @ stock clock. Quite shocking as the temp already reach 80C during 15 minutes of blend test blink.gif Is that normal for stock clock to reach that high temp? i'm using custom loop (360mm magicool rad with tt riing fan).

Attached Image
User is offlineProfile CardPM
Go to the top of the page
+Quote Post
owikh84
post Sep 14 2017, 09:15 PM

i7 Clan
Group Icon
Overclocker's United
Group: Elite
Posts: 8,146

Joined: Nov 2007
From: Butterworth, PG / Machang, Kelantan



QUOTE(azilazwa @ Sep 14 2017, 11:03 AM)
Quite easy playing with multiplier only. But, havent found sweet spot yet.

A little update. So, i changed my psu to Leadex 650w silver and running Prime95 at stock clock just to test whats the temp at full load @ stock clock. Quite shocking as the temp already reach 80C during 15 minutes of blend test  blink.gif  Is that normal for stock clock to reach that high temp? i'm using custom loop (360mm magicool rad with tt riing fan).

Attached Image
*
Unfortunately I have have any good temperature data to compare with yours, but my last 6700K at stock clock (4.2GHz) registered around 55-65C at full load in AIDA64 under a custom loop powered by a pair of 360mm radiators. It's hard to tell unless we have identical systems to compare with each other.

Recent gen Intel CPUs, including the 6700K have thermal issue due the ineffective Intel thermal paste used and excessive gap between inside the chip, which cause inefficient thermal dissipation between the chip die and the integrated heatspreader (IHS).

Anyway Prime95 is an unrealistic punishment of a CPU, generating a lot of heat compared to daily usage. So it's quite normal to see insane temperature during the stress test. Perhaps you should try reducing the vCore to shave off few degrees?
User is offlineProfile CardPM
Go to the top of the page
+Quote Post

14 Pages « < 12 13 14Top
Bump TopicReply to this topicTopic OptionsStart new topic
 

Switch to:
| Lo-Fi Version
0.0744sec    4.30    6 queries    GZIP Disabled
Time is now: 24th September 2017 - 08:10 PM