Help Troubleshoot Multiple Blue Screens w/ Clean Install

Všechny diskuze > Fóra služby Steam > Hardware and Operating Systems > Detaily tématu

KBoy420 15. dub. 2018 v 16.53

Hi All,

I was getting multiple BSODs while playing any Steam games. The two most common were "Driver Verified Detected Violation" and "Driver Page Fault in Freed Special Pool". After trying to troublshoot for days I decided to do a clean install. I used the Settings -> Recovery -> Reset This PC and choose not to keep any files. Windows 10 reloaded and I proceeded to install Nvidia GeForce Experience, Steam, and Discord. So far, I've only reinstalled Rust & Just Survive. Everything seemed fine. Launched Rust and got in game and everything was fine for about 15 minutes and then a crash - "Clock Watchdog Timeout".

Ive reset my BIOS to optimal (default) settings and have no overclocking or AISuite installed.

Can anyone help me figure out what is going on?

Windows 10 Professional (64-bit)
ASUS P8Z77-V LK
Intel i5-3570K @ 3.4-GHz
Nvidia GTX 750
8 GB RAM

< >

Zobrazeno 1–15 z 20 komentářů

Omega 15. dub. 2018 v 16.54

You should do an actual clean install. Not this Windows reset stuff.

Carlsberg 15. dub. 2018 v 18.02

Omega původně napsal:
You should do an actual clean install. Not this Windows reset stuff.

Agreed, windows reset is not a clean install.

chiefputsilao✖️ping 15. dub. 2018 v 22.30

after installation of windows OS, download and update windows (non-driver) and reboot as needed.
install motherboard chipset drivers and reboot.
install gpu driver and reboot.
install microsoft directx 9.0c, and visual C++ and reboot.
install your antivirus and other favorite programs/games.

(i have older hardware: i3-530+intel dh55tc; asus h61 usb3 + i3-3240; ecs h61h2 m12 + g2030 and had no BSOD so far in windows 10 fcu)

KBoy420 2. kvě. 2018 v 18.24

Ive nowndone this twice and still getting crashes.

SenMithrarin85 2. kvě. 2018 v 18.33

KBoy420 původně napsal:
Ive nowndone this twice and still getting crashes.

driver verified violation is usually caused by a problematic driver. try the program "whocrashed" it might point you to the driver in question.

Omega 2. kvě. 2018 v 18.33

KBoy420 původně napsal:
Ive nowndone this twice and still getting crashes.

Then run memtest86 to check if your RAM may be bad.

if you RAM is fine something else is likely deffective.

Naposledy upravil Omega; 2. kvě. 2018 v 18.34

Carlsberg 2. kvě. 2018 v 18.42

look if there is a bios update for that board. After windows is installed get the updates and update all drivers. Your device manager list should be clean and showing no errors, do not rely on windows to update drivers, do it yourself especially chipset and onboard devices, when the OS is running stable go on and install your other stuff.

The clock watchdog timeout is a cpu error and is more than likely caused by a bios or driver issue.

KBoy420 3. kvě. 2018 v 4.56

I can confirm that my BIOS is up-to-date and all system drivers are as well. I've also usd all the Windows 10 tools to check drivers, memory test, my SDD, everything is checking out. There are also no notifications or errors in Device Manager under any devices.

I also got the WhoCrashed tool and it is reporting on the last 2 crashes that occurred last night. Both are Machine Check Exceptions, probably caused by the following module: hal.dll. The crash took place in the Windows kernal. Possibly this problem is caused by another driver that cannot be identified at this time.

Conclusion: 2 crash dumps have been found and analyzed. No offending third party drivers have been found.

As I said I ran the Windows memory diagnostic tool but will DL MemTest and run that.

What next?

Naposledy upravil KBoy420; 3. kvě. 2018 v 5.00

igloosfolly 3. kvě. 2018 v 7.07

Check the version of your windows 10 1803 17134.1 is nasty mess I personally have four pages of BSOD errors from testing that version. Microsoft fixed that mess pretty much by ver 17134.4 and 17134.5 is completely stable. If you have the first version roll back and wait for the newest .4 or .5 to be released to the general public.

KBoy420 3. kvě. 2018 v 8.52

MemTest just completed and no problems were found. Running Version 1709 (Build 16299.402). Any other suggestions?

#10

igloosfolly 3. kvě. 2018 v 8.54

Try windows, settings,check for updates lets get those patches in there.

#11

KBoy420 3. kvě. 2018 v 8.56

All Windows updates are installed, except for the Version 1803 update which just downloaded last day or two.

#12

Buck 3. kvě. 2018 v 9.05

""Driver Verified Detected Violation" and "Driver Page Fault in Freed Special Pool".

These are both driver problems. Make sure you have the latest WHQL drivers for all components installed. Remove/reinstall if you do.

"CLOCK_WATCHDOG_TIMEOUT" is probably hardware related. overclocking or a CPU/Memory defect are strong common suspects. A CPU with insufficient or cooling or a cooler mounting problem can be a possible cause as well.

If you are overclocking, reduce to defaults.
You can try underclocking your memory for testing purposes
you can also try massaging the CPU and memory voltage (if your board allows this, which most do) to see if that stabilizes the system.

KBoy420 původně napsal:
MemTest just completed and no problems were found. Running Version 1709 (Build 16299.402). Any other suggestions?

How long did you run it for?

The thing about memtest is that it doesn't emulate real world memory usage. It runs patterns which can be good for shaking out double faults (Uncorrectable Errors), but single faults (single bit errors) may just quietly pass through because the hardware is simply incapable of detecting them. Memtest SHOULD, but single bit errors can happen so infrequently you may need to run the test for days before you see one crop up, even when they might be happening more frequently and/or have more of an impact on a real world operating environment.

Naposledy upravil Buck; 3. kvě. 2018 v 9.21

#13

KBoy420 3. kvě. 2018 v 9.11

These are both driver problems. Make sure you have the latest WHQL drivers for all components installed. Remove/reinstall if you do.

Alrteady done and couble checked.

"CLCOK_WATCHDOG_TIMEOUT" is probably hardware related. overclocking or a CPU/Memory defect are strong common suspects. A CPU with insufficient or cooling or a cooler mounting problem can be a possible cause as well.

Temp readings all seem to show no over heating issues and all fans are spinning and unobstructed.

If you are overclocking, reduce to defaults.
You can try underclocking your memory for testing purposes
you can also try massaging the CPU and memory voltage (if your board allows this, which most do) to see if that stabilizes the system.

No overclocking being used now or in the past.