Solved Diagnosing CPU instability

Status
This thread has been Locked and is not open to further replies. Please start a New Thread if you're having a similar issue. View our Welcome Guide to learn how to use this site.

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
I want to ask if anyone has any ideas regarding what could be going on with my Phenom II X6 T1100 instability.

The instability is difficult to trigger, it only seems to occur in games and with Prime95. Games are very unpredictable, its been happening at most twice an evening or if I'm lucky, once every few days, Resulting in a hard freeze. Prime 95 torture with small FFT's will crash the system within 15 minutes, starting with a lockup then a restart.

These events will leave this in my windows 10 pro event logs:
===============================
WHEA-Logger ID# 18

A fatal hardware error has occurred.

Reported by component: Processor Core
Error Source: Machine Check Exception
Error Type: Cache Hierarchy Error
Processor APIC ID: 2
===============================

There are also some "corrected events" along with the fatal ones.
Attempting to look this up at Microsoft internet using the app provided hyperlink will cause Event viewer to lock up and I need to terminate it from task maanger. >.>

The RAM tests clean on an overnight MemTest64 run, and I'm fairly sure its not related to memory.

I have never Overclocked this system.
I do use the built in turbo mode the Phenom II has.
The system and all mobo components are about 7 yrs old.

I have a Liquid cooling system that appears healthy to physical inspection, pump is working, no leaks, liquid level OK, etc.
Though I do intend to top off the coolant this weekend.

The harshest games max T around 52C and prime 95 can push it to 63 C max.

So, that brings me ot the question:

These symptons are new within the past month, it used to be rock solid, which is what is raising my concern.

Does this mean

Is my Phenom II aging out? What is a typical CPU lifespan?
Suggest anything to stop or mitigate the lock ups?
Will it get worse If I do nothing and deal with an occasional crash?

Details can be given, I'm looking for some generalized advice first.

Thanks in advance.
 

dmccoy

Darrin
Joined
Oct 7, 2017
Messages
4,489
Is it just freezing or are you getting any BSOD’s? If so, please upload the minidumps.

It is best to test memory as follows:

Run memtest86+ version 5.01 for at least 8 passes.
http://www.memtest.org/

This may take hours so plan to run it overnight.

Testing takes approximately 1 - 2 hours /GB RAM
Just 1 error is a fail and you can abort testing
Then test 1 RAM module at a time in the same DIMM each for 8 or more passes
When Memtest86+ has completed 8 or more passes take a picture and post an image and upload to your reply

https://www.tenforums.com/tutorials/14201-memtest86-test-ram.html

MemTest86 isn't the only memory test. I also suggest Prime 95's Blend test - as it will find memory errors that MemTest86+ won't.

Test hard drive if you have not already. Run the long test.
Seagate Seatools for Windows
https://www.seagate.com/support/downloads/seatools/seatools-win-master

After the above, Please perform the following steps to start.

Step 1:
Speccy
http://filehippo.com/download_speccy

1. File > Save Snapshot. This will create a file called [hostname].speccy
2. File > Publish Snapshot. This saves your snapshot to their servers
3 Copy and Paste the link to your next Reply

Step 2:
Click on Start menu
Type in cmd
Right click on cmd in list and select run as administrator
Copy and Paste the commands into the command console
1. sfc /scannow
2. dism /online /cleanup-image /restorehealth

When these have completed.
Right click on the top bar command window
Left click on Edit then Select All
Right click on the top bar again
Left click on edit then copy
paste into your reply

Step 3:
Download Process Explorer
https://docs.microsoft.com/en-us/sysinternals/downloads/process-explorer

1. Save it to your desktop then run it. Right Click and Select Run As Administrator
2. Select View Menu Click Select Columns
3. In the Select Columns Window Check Verified Signer and Click OK
4. Select Options Menu and Check Verify Image Signatures
5. Double Click o the CPU Column to sort by highest CPU usage.
(System Idle Process should be at the top once sorted properly)
6. Wait approximately a minute
7. Select Save or Save As from the File Menu and save System Idle Process.txt to Desktop
8. Upload the file to your Reply

Step 4:
Please download and Run the Event Viewer Tool by Vino Rosso
http://images.malwareremoval.com/vino/VEW.exe

1. Right click on Vew.exe program and select Run as Administrator
2.. Under Select log to query
3. select System
4. Under Select type to list
Select:
* Error
* Warning
Choose the Number of events as follows:
Click on umber of events
Type 20 in the box (1 to 20)
Click the Run button (Notepad will open the log)
upload files to next reply
Rename to SystemLog.txt or it will be overwritten

Run these steps again Except at step 3.
Select Application
Rename log file to ApplicationLog.txt
Upload or Paste to next reply
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
No BSOD, controls and screen freeze, need to hit restart button except in the case of prime95 which reboots the machine. I could not find any prime95 error logs but I am supposing it does not have the oppertunity to write them.

I am working through your advice, might be day or two.
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
Since the RAM is going to take overnight, going to throw in a few things now. More to come tomorrow.

Here are my specs:
http://speccy.piriform.com/results/xvLHZyKBKRFvcGsTj8YCHZl


cmd commands:

Microsoft Windows [Version 10.0.16299.248]
(c) 2017 Microsoft Corporation. All rights reserved.

C:\windows\system32>sfc /scannow

Beginning system scan. This process will take some time.

Beginning verification phase of system scan.
Verification 100% complete.

Windows Resource Protection did not find any integrity violations.

C:\windows\system32>dism /online /cleanup-image /restorehealth

Deployment Image Servicing and Management tool
Version: 10.0.16299.15

Image Version: 10.0.16299.248

[==========================100.0%==========================] The restore operation completed successfully.
The operation completed successfully.

C:\windows\system32>
 
Joined
Feb 7, 2018
Messages
355
any ideas regarding what could be going on with my Phenom II X6 T1100 instability
From said symptom of freezing and no BSOD....I would start looking at an over heating issue.

What are the temps of the CPU at idle and in use? What type of CPU cooling do you have? When was the last time the system was cleaned (blown out with compressed air) from dust and what not? When was the last time the thermal paste was changed?

Is my Phenom II aging out?
It's a possibility.

What is a typical CPU lifespan?
No real concrete evidence about this as they tend to become obsolete first before failing.

Suggest anything to stop or mitigate the lock ups?
Keep CPU fan clean as well as system on a regular basis. Replace thermal paste: I like to replace my Thermal paste every other year. I use Artic Silver. Replace stock CPU cooler with an aftermarket cooler. Best bang for buck is the Cooler Master Hyper 212 evo. Make sure you case has adequate air flow.

Will it get worse If I do nothing and deal with an occasional crash?
It can but that depends on the cause of the freeze....which could be just about anything from hardware to software.
 

dmccoy

Darrin
Joined
Oct 7, 2017
Messages
4,489
Everything looks okay on Speccy and you already mentioned testing your temps so I don’t think thst is an issue. Please continue with the steps I provided and submit when available.
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
12 hrs got through 3.5 passes on memtest86. zero errors. Going to need another overnight run. A pic to be attached.
I'd be surprised if its RAM at this point, I think I would be having larger problems if it was.

The 60C+ from prime95 does concern me enough to entertain it could be overheat related. It shouldn't get above 50C with this cooler. This thread:

https://forums.techguy.org/threads/liquid-cooled-cpu-overheating.1171570/#post-9256318

Was a previous incident about 2 yrs ago when the entire Lq cooling system minus the radiator was replaced. Including thermal paste. CPU may have taken some subtle heat damage at that point.

Currently, idle or light load temps hang 27-31 C.

I plan to add collant tonight and see if that helps prime95 last longer before crashing.

What do you think about disabling turbo mode on the CPU? From the BIOS.. i think.

MemTest86-IMG_5955.JPG
 

Attachments

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
Attached is the pic showing 5 more passes with no errors.
I have done a few more tests with prime95, one of them failed when the Temp was in the 40's, which further goes against the overheat possibility. Two more have continued long enough to reach 60C. However, I put windows into a clean boot mode by disabling non-MS services and disabling startup items. I jsut finished a prime95 test of 15 minutes. All I had time for tonight, but this is looking more like some odd software / OS / conflict issue.
 

Attachments

dmccoy

Darrin
Joined
Oct 7, 2017
Messages
4,489
Yeah those temps should not be of concern. It might be worth testing without turbo mode. I was going to suggest the clean boot as a next option. Did you get a chance to test your games etc for any freezing other then Prime ? If not, please do that next and then we can start trying to narrow it down.
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
The games I'm playing now really do not relyably trigger the problem, maybe once in 10 gaming hours. So I find it easier to use Prime since it will consistantly trigger the crash. I have a feeling that If I get Prime95 stable - the games will also be.

Throughout the last two days, the Clean boot has certainly helped, it did not stop the lock ups. Prime95 made it almost to two hour mark but still managed to lock and restart the machine. So I am purplexed now.

Norton seems to be able to turn its disabled service back on (inside msconfig). I'm considering temp. Uninstalling it. I am running Prime now with the auto protect and silent mode turned on. (ver 22)

If this is indeed software, I am suspecting either Norton or Java, or possibly some obscure driver issue.
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
Still happens with Norton gone.
The trand with the Prime95 is that the system is more vulnerable to the "small FFT" torture test then the Blended test.

I'm running out of ideas.

I've caught Prime 95 crashing anywhere from 40 to 65 C.... mid 60's C is the highest
I've seen before it goes down. < That seems to be the temp that reaches equalibrium with the cooling system.
 

dmccoy

Darrin
Joined
Oct 7, 2017
Messages
4,489
Please run perform the following steps.

Step 1.
Check Hard Drive for Errors using one of the options below:

1. BIOS based system
run Seagate Seatools for DOS (Long Generic test). Boot from disk to run

2. UEFI based system
run Seagate Seatools for DOS (Long Generic test). Boot from disk to run it (will have to disable SecureBoot and enable CSM/Legacy mode).

3. Connect the hard drive to another working computer, and run Seagate Seatools for Windows (Long Generic test).

Seagate Seatools for DOS (free): http://www.seagate.com/support/downloads/seatools/seatools-dos-master/

Seagate Seatools for Windows
https://www.seagate.com/support/downloads/seatools/seatools-win-master/

Step 2.
Stress Test GPU - FurMark
https://www.tenforums.com/tutorials/13081-furmark-gpu-stress-test.html
Make note of the importance of temperature monitoring and aborting the test in the GPU link
Test for approx 1 hour
Upload a picture of the screen when complete using the Microsoft snipping tool or similar method

Step 3.
Open administrative command prompt and type or copy and paste: chkdsk /x /f /r

It is best to run this 2-3times for best results
This may take many hours to run so plan to run it overnight.
Post or upload the chkdsk results
 

Krybo

Thread Starter
Joined
May 19, 2016
Messages
18
Been away some time this week.
FurMark: ran about 5 hours worth of "burn-in" test.
GPU temp rise levels off around 56C. CPU is hardly bothered, ~35C.
- Its stable.
Are there any specific settings for FurMark runs? I'm doing 1280x1024 (the highest res I normally use) with 8x MSAA.

Oddly, I had one lock-up occur this week when nothing was running except Firefox.
Its only happened once, but it is concerning.

Getting to the disk checks next.
 
Status
This thread has been Locked and is not open to further replies. Please start a New Thread if you're having a similar issue. View our Welcome Guide to learn how to use this site.

Users Who Are Viewing This Thread (Users: 0, Guests: 1)

As Seen On
As Seen On...

Welcome to Tech Support Guy!

Are you looking for the solution to your computer problem? Join our site today to ask your question. This site is completely free -- paid for by advertisers and donations.

If you're not already familiar with forums, watch our Welcome Guide to get started.

Join over 807,865 other people just like you!

Latest posts

Members online

Top