Close

Register Now!

To Get More Info and Daily Reward.

Please login or register.
Pages: [1]   Go Down

Author Topic: Unstable cpu overclock/crashes no matter what, seeking expert’s advice  (Read 891 times)

0 Members and 1 Guest are viewing this topic.

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Hi everyone, I appreciate you for taking the time to contribute to this topic.

I've had spent dozens of hours trying to diagnose and fix the problems that my pc has with the overclock, until now that I am so frustrated and out of hope that I am reaching out to you all that are far more experienced in such field and hoping that with your help I will manage to resolve my issues.

For weeks I’ve had problems with overclock, no matter what setting I turn on or off/increment or decrement, I am unable to reach a stable overclock that wouldn’t result in system freeze or sometimes a BSOD. (Usually my system just freezes and the only thing I can do is shut down the pc by holding the power button for few secs. Second or so before the system freeze, there’s always this single static electric click noise which is followed by whole system freeze.)
In fact, I am starting to wonder whether there is a hardware issue like PSU failing to supply enough power to CPU or motherboard having some kind of defect that causes this. Unable to investigate this as I do not possess a backup psu, nor motherboard.

In fact, I am starting to wonder whether there is a hardware issue like PSU failing to supply enough power to CPU or motherboard having some kind of defect that causes this. Unable to investigate this as I do not possess a backup psu, nor motherboard.

My Setup;
CPU: I5-8600k
MOBO: MSI Z370-A Pro
RAM: Kingston Hyperx Predator DDR4 3000MHz
PSU: EVGA 600W W1 80+
GPU: Asus Strix GTX 970
Heatsink: Cooler Master Hyper 212 Evo

CPU Overclock


DRAM & Voltage


CPU Features


DigitaALL Power


Advanced DRAM Configuration


The above OC values are the most recent values that I have used.

Is there anything that I am missing or have configured incorrectly? Anyone with same setup mind sharing their OC? I am open to your suggestions and willing to cooperate with you. Could the PSU be the problem as it is old and classified as low-tier? If that so, can I get a stable but lower OC until I get new psu?
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

Hi,

First try these settings:

CPU Ratio: 48
CPU Ration Mode: Dynamic Mode
CPU AVX Offset: -2
Ring Ratio: 45
CPU Core Voltage Mode: Adaptive + Offset.
CPU Core Voltage: Auto.
CPU Core Voltage Offset Mode: +
CPU Core Voltage Offset: Auto
CPU SA Voltage: 1.200V
CPU IO Voltage: 1.150V
Load Line Calibration Control: Mode 4
CPU Lite Mode: Mode 1


Download Link HWiNFO BETA:
https://www.fosshub.com/HWiNFO.html

Download Link Prime95:
ftp://mersenne.org/gimps/p95v298b6.win64.zip

Open HWiNFO Sensors Monitoring and then open Prime95.

Please use only these settings:

Custom.
Min and Max FFT's: 1344
Run FFT's in-place.
Disable AVX2.
Disable AVX.

Run Prime95 for about 10 minutes, if any WHEA errors appear in HWiNFO you need to stop Prime95. After 10 minutes if Prime95 is running without errors please take a full screen screenshot of HWiNFO Sensor Monitoring and post it back here.

Attached are screenshots of HWiNFO while my system is running Prime95 with a 4.8GHz OC. Ambient room temperature is 24°C.
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Hi,

First try these settings:

CPU Ratio: 48
CPU Ration Mode: Dynamic Mode
CPU AVX Offset: -2
Ring Ratio: 45
CPU Core Voltage Mode: Adaptive + Offset.
CPU Core Voltage: Auto.
CPU Core Voltage Offset Mode: +
CPU Core Voltage Offset: Auto
CPU SA Voltage: 1.200V
CPU IO Voltage: 1.150V
Load Line Calibration Control: Mode 4
CPU Lite Mode: Mode 1


Download Link HWiNFO BETA:
https://www.fosshub.com/HWiNFO.html

Download Link Prime95:
ftp://mersenne.org/gimps/p95v298b6.win64.zip

Open HWiNFO Sensors Monitoring and then open Prime95.

Please use only these settings:

Custom.
Min and Max FFT's: 1344
Run FFT's in-place.
Disable AVX2.
Disable AVX.

Run Prime95 for about 10 minutes, if any WHEA errors appear in HWiNFO you need to stop Prime95. After 10 minutes if Prime95 is running without errors please take a full screen screenshot of HWiNFO Sensor Monitoring and post it back here.

Attached are screenshots of HWiNFO while my system is running Prime95 with a 4.8GHz OC. Ambient room temperature is 24°C.
I was unable to run Prime95 for more than 2 minutes. Each time either one or more workers found errors and stopped or the whole system just froze. I will attach screenshots of moments after the workers stopped. Perhaps I need to manually increase the core voltage as 1.224 seems a bit too low for 4.8GHz overclock?
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

Ok try these settings:

CPU Ratio: 48
CPU Ration Mode: Dynamic Mode
CPU AVX Offset: -2
Ring Ratio: 45
CPU Core Voltage Mode: Adaptive + Offset.
CPU Core Voltage: 1.300V
CPU Core Voltage Offset Mode: +
CPU Core Voltage Offset: Auto
CPU SA Voltage: 1.200V
CPU IO Voltage: 1.150V
Load Line Calibration Control: Mode 4
CPU Lite Mode: Mode 1

What is the make and model of you PC case?
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Ok try these settings:

CPU Ratio: 48
CPU Ration Mode: Dynamic Mode
CPU AVX Offset: -2
Ring Ratio: 45
CPU Core Voltage Mode: Adaptive + Offset.
CPU Core Voltage: 1.300V
CPU Core Voltage Offset Mode: +
CPU Core Voltage Offset: Auto
CPU SA Voltage: 1.200V
CPU IO Voltage: 1.150V
Load Line Calibration Control: Mode 4
CPU Lite Mode: Mode 1

What is the make and model of you PC case?
So I did increase the core voltage to 1.300V as you suggested and it has been successfully running for an whole hour. Just before you replied to my reply, I incremented the core voltage to 1.260V and ran the prime95, It ran for 34 minutes without any errors until it finally froze my system. I guess what I do now is decrement the core voltage until I start getting system errors or freezes and increment it again uptill I find stability, right?
The case that I am using is BitFenix Neos ATX Tower.
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

Ok it looks good.

Lower the Vcore with 0.005V increments till either Prime95 starts to give errors or WHEA errors are detected when using the attached batch script.

Extract and copy the script to the Prime95.exe folder. Use it to run Prime95. If there are any WHEA error logged, Prime95 will fail to open and you'll first need to clear the Event Viewer System log.

When you start the custom stress test the script will continue to monitor for WHEA errors in the background. If the script detects WHEA errors it will immediately end Prime95. Increase the Vcore again with 0.005V till no WHEA errors are logged.

Use this command to clear Event Viewer System log:
Code: [Select]
WEVTUtil.exe clear-log System
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Ok it looks good.

Lower the Vcore with 0.005V increments till either Prime95 starts to give errors or WHEA errors are detected when using the attached batch script.

Extract and copy the script to the Prime95.exe folder. Use it to run Prime95. If there are any WHEA error logged, Prime95 will fail to open and you'll first need to clear the Event Viewer System log.

When you start the custom stress test the script will continue to monitor for WHEA errors in the background. If the script detects WHEA errors it will immediately end Prime95. Increase the Vcore again with 0.005V till no WHEA errors are logged.

Use this command to clear Event Viewer System log:
Code: [Select]
WEVTUtil.exe clear-log System
Today, before I went to sleep, I thought I found a stable core voltage of 1.272(reported in hwinfo monitor). I ran prime95 for whole 9 hours over the night with no errors or freezes. I woke up and it was still running, however, now, when I am attempting your suggested script above, the system oddly freezes, sometimes a minute or two into the stress test and sometimes 10-15 minutes into the stress test. I will attach both 10-minute stress test with the script you suggested and the stress test than I ran over the night.

How odd is that? 9hours compared to 10 minutes.. What is this instability? What could cause this? Is there something that I am missing like adjusting other voltage values as Auto sometimes sets them inappropriately high or low?

The first two pictures that I have attached are with the script that you provided(crashed around 13th minute). Last two pictures are the 9hour stress test results that I ran over the night without the script.
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

Higher ambient room temperature can make an overclock unstable even if only a few degrees higher. To compensate you'll have to set the Vcore higher.

I always keep a record of my overclock settings, min max temperatures when running benchmarks and also the ambient room temperature.

Adjust the Vcore to 1,280V.

Also set the case intake fans to spin at maximum speed at about 60°C to help vent out hot air.
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Higher ambient room temperature can make an overclock unstable even if only a few degrees higher. To compensate you'll have to set the Vcore higher.

I always keep a record of my overclock settings, min max temperatures when running benchmarks and also the ambient room temperature.

Adjust the Vcore to 1,280V.

Also set the case intake fans to spin at maximum speed at about 60°C to help vent out hot air.

This doesn't really make sense to me. Nothing is overheating, the temps are fairly safe and the fans turn up as the heat rises so I see no correlation.

Unfortunately, no voltage seems to help. I tried increasing and decreasing Vcore, system still freezes under load.

I still don't get why all of sudden core voltage of 1.272 is unstable after I ran it for 9 hours..

UPDATE: I did some research on my psu and it turns out it's terrible. I will be upgrading my PSU ASAP and will come back to this topic as soon as I get my new power supply.
Logged

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Hi, so the psu has arrived and I have installed it. Unfortunately, the same issues persist, system still freezes on prime95 stress test :( No WHEA errors though.
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

What is the Vcore set to?

Take the case front cover and side panel off, set all the fans to 100% in the BIOS then run Prime95 again.
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Currently, Vcore is set to 1.320 and still crashes. 

Still crashes with side panels off and fans on the max. Could it be problem with motherboard itself?
Logged

Chike

  • LIEUTENANT GENERAL
  • ****
  • Online Online
  • Posts: 4252

Test with memory at default settings.
Logged
Motherboard: ASUS Z97-K R2.0 BIOS ver. 0903 2/26/2016
CPU: i5-4670k
Cooler: CNPS10x Performa
Memory: F3-2400C11D-16GXM
GPU: EVGA GeForce GTX 950 FTW ACX 2.0
SSD: 1. Samsung PM981 512GB, 2.  Samsung 840 EVO 120GB
HDD: WDC WD3200AAKS-00L9A0
PSU: Antec 500W VP500PC
OS: Windows 10 Home 1903 64-bit / Windows 7 Home Premium 64-bit
Case: Compucase 6C60

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Tried it numerous times. It seems that at the moment it doesn't crash under load but as soon as the load eases e.g. stopping prime95, it instantly crashes. Problems with vdroop I guess? Any suggestions?
Logged

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

What errors are logged in Event Viewer?

Set the Load Line Calibration to Mode 3.

If increasing the LLC still results in a crash, try with a lower overclock then:

CPU Ratio: 47
CPU Ration Mode: Dynamic Mode
CPU AVX Offset: -2
Ring Ratio: 44
CPU Core Voltage Mode: Adaptive + Offset.
CPU Core Voltage: 1.300V (If no crash occurs, start lowering the the Vcore).
CPU Core Voltage Offset Mode: +
CPU Core Voltage Offset: Auto
CPU SA Voltage: 1.200V
CPU IO Voltage: 1.150V
Load Line Calibration Control: Mode 4
CPU Lite Mode: Mode 1
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

Krank

  • FIRST LIEUTENANT
  • ***
  • Offline Offline
  • Posts: 402

Please post a F12 BIOS screenshot of the CPU Fan profile.
Logged
MSI Z370 Gaming Pro Carbon (A.90 BIOS)
Intel 8700K 5GHz@1.312V Adaptive (Delidded)
MSI GeForce GTX 1070 Ti Duke Gaming 8GB (OC'd & Undervolted + Shunt Mod)
G.SKILL TRIDENT Z F4-3200C16D-16GTZB Kit (OC'd to 3333MHz CL16-18-18-38-499-2T 1.400V)
Samsung 960 Evo 250GB(OS), Samsung 860 Evo 250GB(Games, Portable programs, Batch scripts)
2x Western Digital Blue 1TB (Temp folders, Data backup)
11x Corsair ML120 Pro fans (Pressure optimized)
Corsair H150i Pro(Modified) (6x Shrouds)
Corsair RM850x
Raidmax Alpha Prime
Samsung 43" LCD TV
Windows 10 Enterprise (1903) 18362.418
https://www.fosshub.com/HWiNFO.html?dwl=hwi_612.zip
Manage Speculative Execution Settings Script v1.9

DensTopic starter

  • PRIVATE FIRST CLASS
  • ***
  • Offline Offline
  • Posts: 9

Thank you Krank for your input, appreciate your contribution!

The problem was one of my ram sticks. Shortly after I restored my bios to default settings I changed one thing at a time until it came to selecting XMP profile. As soon as I selected XMP profile 1, the crashing started again. Then I ran the Windows Memory Diagnostic for like 5th time and for the first time ever, errors were reported. Then I ran the same Windows Memory Diagnostic on each stick separately and discovered that one of my ram sticks kept reporting problems. I sent back the flawed ram stick and received a replacement a few days later. Now I am rocking my 4.8GHz overclock with no problems at all, runs smooth! :biggthumbsup:
Logged
Pages: [1]   Go Up