Computer Intermittently Freezes for 1-5 Seconds

Brian_Adams · May 6, 2022, 4:47pm

My Equipment:

Intel Core i5 9600k
ASRock Z390 Pro4
32 GB DDR4 (XMP 2.0)
- 1. DDR4-2132 G.Skill
- 1. DDR4-2132 G.Skill
- 1. DDR4-2401 Team Group Inc.
- 1. DDR4-2401 Team Group Inc.
NVidia GeForce RTX 2070 Super

I notice this most often when I’m watching videos as it usually results in a prolonged or stuttering noise that accompanies the freeze. When it freezes, my mouse and keyboard also freezes. It can occur as often as once per hour but usually more rarely. My graphics card was recently upgraded from 1070ti and it occurred then also so I think that the GPU isn’t the issue. Also, I recently updated my GPU driver. After the freeze, everything seems to continue as normal however in some cases, I get kicked out of CP2077 or Path of Exile.

I tend to keep many tabs open in my browser of things that I intend to get to eventually and Chrome often is the highest consumer of cycles and memory, but I don’t seem to have issues with concurrently running other things. So, I guess the paging and/or swapping is ok.

I am running Private Internet Access which definitely adds noticeable latency but doesn’t seem to affect overall throughput.

I’m not skilled with the event viewer or sysmon, so I’m unsure how to investigate after a freeze. I use HWInfo64 to monitor temps and those seem to be reasonably steady and mid-lo.

Can anyone detect issues in this report and/or offer tracking/investigation techniques?

Thank you,

Brian

risk · May 7, 2022, 12:43pm

Sounds like a bad driver, cursor<->gpu code path is typically highly optimized to not induce motion sickness… if you’re seeing freezes, it’s likely something very low level that’s having system issues.

Do you see anything in logs?

Brian_Adams · May 8, 2022, 9:05pm

I don’t know “cursor<->gpu code path”. I have installed the latest gpu driver.

when I look in the device manager, I don’t have any yellow triangles or other symbols indicating a problem.

Where do I find the logs that you referred to?

risk · May 8, 2022, 9:19pm

anon7678104 · May 8, 2022, 10:40pm

could be your stressing the cpu’s mmu by running 4 sticks of ram at different speeds.
more stress=more heat=more chance of throttling due to stress.

stick em all at 2133 and see if the problem persists.
if it does pull 2 sticks and run at 1 speed.

while you can mix and match ram, when you do, you run it all at the same speed and try to run it at the same latencies.
you are trying to run 2 sticks at one speed and 2 at another. this will hammer the mmu and likely cause it to overheat, causing the mmu to throttle the cpu.

a couple of negatives running mixed speeds. the ram will run at the lowest speed of all the sticks.
you have 2 sticks of 2133 and 2 sticks of 2400.
the ram will try to run the 2400 at 2133 but will likely try to mix and match latencies.
and this is where the stress comes in for the mmu.

other reasons you dont want to mix and match timings. the amount of bad data written back to your o.s will increase. to the point of in a few months you will notice stuff just stops working (typically settings dialogs and the likes)
then you will be running sfc/scannow and dism fix’s

(you might wanna run it now anyways. likely result is repairs needed )

i looked up 2133 and 2400 they can both run at cas 16
the rest you will have to figure out from cpu-z or similar.
but compare the jdec’s of both sets of ram and see if you can match the other timings.
the more you can match without losing stability the less stress on the mmu.

but really. for stability, dump the 2133 if you dont really need it, and run the 2400 at the lowest latency you can.

Brian_Adams · May 8, 2022, 11:56pm

OK, so, I look for recent activity in the Event Viewer immediately after a freeze that eventually recovers. I’ll see what shows up.

Brian_Adams · May 9, 2022, 2:24am

Risk:

I had a freeze at 7:03. There were two error events at the same time: one in system and one in security.

--------------------------------------
System 5/8/2022 7:14:03 PM
Display driver nvlddmkm stopped responding and has successfully recovered.
--------------------------------------
Security 5/8/2022 7:14:03 PM
Special privileges assigned to new logon.

Subject:
	Security ID:		SYSTEM
	Account Name:		SYSTEM
	Account Domain:		NT AUTHORITY
	Logon ID:		0x3E7

Privileges:	SeAssignPrimaryTokenPrivilege
			SeTcbPrivilege
			SeSecurityPrivilege
			SeTakeOwnershipPrivilege
			SeLoadDriverPrivilege
			SeBackupPrivilege
			SeRestorePrivilege
			SeDebugPrivilege
			SeAuditPrivilege
			SeSystemEnvironmentPrivilege
			SeImpersonatePrivilege
			SeDelegateSessionUserImpersonatePrivilege
--------------------------------------

I don’t know what either of these indicate. I double checked with my NVIDIA GE Force Experience which told me that I already had the latest driver for my device. Not sure what to think about the simultaneous security error.

Any thoughts?

Mach3.2 · May 9, 2022, 2:33am

Display driver crashed. Are you overclocking or undervolting your GPU?

Try running factory defaults for your GPU if you’re overclocking/undervolting your GPU.

Brian_Adams · May 9, 2022, 3:19am

Mach
I don’t know enough to overclock or undervolt though I know Afterburner can adjust some stuff.

Mach3.2 · May 9, 2022, 3:49am

I’d try uninstalling nvidia drivers in safe mode using DDU, then install it again.

Also what are the temps like on your GPU while you’re gaming?

alkafrazin · May 9, 2022, 4:51am

Power supply perhaps? Do you have an uninterruptible power supply you can test with? Small fluctuations in wall power can lead to some odd stability issues. Plugging into a UPS or power filter can help to at least eliminate that possibility.

Double check your memory speeds. Set all sticks to the lowest common clocks/highest timings(2133 looks like)
Small memory instability can end up doing things like this.

It could also just be that nVidia’s drivers have been unstable garbage lately, and you’re one of those unlucky enough to see it.

anon7678104 · May 9, 2022, 10:00am

lol you really dont wanna blame the ram config do yah!.
ok have fun trying to figure out all the rest of the crap your gonna be suggested thats wrong.

your ram config IS THE PROBLEM!.

Brian_Adams · May 11, 2022, 4:36pm

Hexit

It took a while to get the memory and put it in place. Here is my new configuration:

Intel Core i5 9600k
ASRock Z390 Pro4
32 GB DDR4 (XMP 2.0)
- empty
- DDR4-3596 Nanya Tech
- empty
- DDR4-3596 Nanya Tech
NVidia GeForce RTX 2070 Super

Brian_Adams · May 12, 2022, 10:26pm

Mach

Used DDU then specifically re-installed from NVidia. Freezes continue to occur, but perhaps less frequently. Not gaming currently, but with quickhash excavator running for a few minutes:

When a freeze occurs, I don’t find any errors in the branches of the event viewer log.

thetazman · May 12, 2022, 11:45pm

Have you tries a sfc after replacing the memory? If it was bad memory it could have cause a file corruption!

windows sfc

anon7678104 · May 13, 2022, 12:30am

yeah open an elevated cmd (run as admin) and run

sfc /scannow
then
DISM /Online /Cleanup-Image /CheckHealth
DISM /Online /Cleanup-Image /ScanHealth
DISM /Online /Cleanup-Image /RestoreHealth

also you never mentioned what storage you have. your sure thats not a bottle neck?.

lastly run hw info and grab a screenshot of your ram timings temps volts and the likes.

Brian_Adams · May 13, 2022, 6:55am

Here are the drives I use:
C Samsung SSD 860 EVO m.2 2TB (504.02GB Free)
E HGST HDN726040ALE614 (785 GB Free)

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 found corrupt files and successfully repaired them.
For online repairs, details are included in the CBS log file located at
windir\Logs\CBS\CBS.log. For example C:\Windows\Logs\CBS\CBS.log. For offline
repairs, details are included in the log file provided by the /OFFLOGFILE flag.

C:\WINDOWS\system32>dism /online /cleanup-image /checkhealth

Deployment Image Servicing and Management tool
Version: 10.0.19041.844

Image Version: 10.0.19044.1706

No component store corruption detected.
The operation completed successfully.

C:\WINDOWS\system32>dism /online /cleanup-image /scanhealth

Deployment Image Servicing and Management tool
Version: 10.0.19041.844

Image Version: 10.0.19044.1706

[==========================100.0%==========================] No component store corruption detected.
The operation completed successfully.

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

Deployment Image Servicing and Management tool
Version: 10.0.19041.844

Image Version: 10.0.19044.1706

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

Here is all the info that HWiNFO64 could report:

E-Wasted · May 13, 2022, 7:33am

Your core VIDs are locked to ~1.1V when the Vcore is idling down, (Is that just the voltage the cores are requesting?)

Are you using the balanced power plan in Windows? Using any “performance” power plans disables low power mode and prevents the CPU from idling down without any real benefit.

Also there’s that one Aux mobo sensor that reads 100C, usually they’re false or inaccurate, but if that is actually your VRM/Mosfets they could be thermal throttling. Seems unlikely for a Z390 and an i5 though, usually they peak out at around 115C.

Brian_Adams · May 13, 2022, 7:47pm

Windows was on a “High Performance” but I have switched to “Balanced”.

I don’t know enough to respond to VIDs, Vcor, Aux mobo sensor, or VRM/Mosfets.

I need more specific (simple) action items.

Brian_Adams · May 13, 2022, 7:51pm

I note that on my most recent freeze, the only thing that the Event Viewer reported at that time occurred in security in a manner nearly identical to when I first reported on this.

Is this of consequence? Should I continue to monitor the event viewer events after freezes?