As was reported in the main PrecisionX 5.3.11 Release thread back in April, PrecisionX is causing the WMI Provider Host to be in constant use and is resulting in an unacceptable amount of CPU usage. With PrecisionX minimized to the system tray, its own processes (0-2% CPU usage each with 3 processes) along with the WMI Provider Host (1-3% CPU usage) have a far greater impact than running HWiNFO64 with sensors and summary windows open (1-2% CPU usage) along with the WMI Provider Host (0% CPU usage).
 
This is the task manager with an idle system up until I close PrecisionX:

 
This is the task manager with an idle system but with HWiNFO64's summary and sensors windows open and updating:

 
There's clearly something going on with PrecisionX that isn't right and this is an issue that was reported almost 4 months ago. Even a quick look with Process Monitor shows a bunch of interactions with tzres.dll:

 
Looking at the process stack for wmiprvse, I can see why it would be accessing that particular file, but I have no way of knowing what PrecisionX is doing that would require it:

13 KERNELBASE.dll ConvertTimeZoneMuiString + 0xe6 0x7ffa84b51f42 C:\WINDOWS\System32\KERNELBASE.dll
14 KERNELBASE.dll ConvertTimeZoneMuiStrings + 0x168 0x7ffa84b51abc C:\WINDOWS\System32\KERNELBASE.dll
15 KERNELBASE.dll GetTimeZoneInformation + 0x83 0x7ffa84b51923 C:\WINDOWS\System32\KERNELBASE.dll
16 framedynos.dll CInstance::SetDateTime + 0xbe 0x7ffa747c84be C:\WINDOWS\SYSTEM32\framedynos.dll

 
Querying the SystemSetupInProgress registry key also seems to happen quite often in conjunction with the WMI Provider Host activity.
 
That's pretty much all the information I've been able to gather so far. The fact that nothing has happened regarding this issue doesn't exactly instill me with a lot of confidence... :(
 

I still don't know why EVGA refuses to comment on this bug or fix it. Ugh. 

I think its been mentioned before. It was causing my games to have micro stutter. Pretty sure its related to this bug. Using Afterburner solves the problem and no more micro stutter. 

To repeat what I said in the 6.0.6 release thread so that all this relevant information can be found in one place...
 
 
Well I finally figured out how to prove that PrecisionX is the cause of this problem. From within the thread stack of PrecisionX_x64.exe:

13 KernelBase.dll ConvertTimeZoneMuiString + 0xe6 0x7ffe48f61f42 C:\Windows\System32\KernelBase.dll
14 KernelBase.dll ConvertTimeZoneMuiStrings + 0x15a 0x7ffe48f61aae C:\Windows\System32\KernelBase.dll
15 KernelBase.dll GetTimeZoneInformation + 0x83 0x7ffe48f61923 C:\Windows\System32\KernelBase.dll
16 msvcr120.dll _tzset_nolock + 0x16c, f:\dd\vctools\crt\crtw32\time\tzset.c(180) 0x7ffe3db79b7c C:\Windows\System32\msvcr120.dll
17 msvcr120.dll __tzset + 0x26, f:\dd\vctools\crt\crtw32\time\tzset.c(94) 0x7ffe3db796f6 C:\Windows\System32\msvcr120.dll
18 msvcr120.dll _make__time64_t + 0x1c9, f:\dd\vctools\crt\crtw32\time\mktime64.c(228) 0x7ffe3db7b169 C:\Windows\System32\msvcr120.dll

This is from a Process Monitor capture that I started before running PrecisionX OC 6.0.6 and ended after it was done loading. This is also the first time that tzres.dll is interacted with at all (and tzres.dll.mui shortly thereafter).
 
It obviously hands this task off to the WMI system at some point, but I don't feel like sifting through all 149,763 events in the capture when this is something EVGA should have done months ago when it was first brought to their attention.

Again is this happpening on all versions of windows or just win 7? 

I have the same problem. Windows 10, 1080 FTW, version 6.0.6

I have this problem as well across two builds. Win10 Pro x64.  Happens on both my build with a 2600k + 780sc (with the original) and my build with a 6700k and a 1080 FTW (With XOC).
 
I've also had this 1070 in that other build and it did it there.
 
 

Brief update:
 
Got a reply from the Director of Global Customer Relations confirming that they are looking into the issue.

Chiming in to say I have this problem as well. Recently decided to do a clean installation of my graphics driver, including removing/reinstalling PrecisionX 16. WMI has been in constant use ever since I installed 5.3.11. Running 64-bit Windows 10 Professional with a 980Ti. I didn't have this problem before with 5.3.10. Does rolling back fix it? I may just do that.
 
As an aside, I've had varying degrees of success by restarting Windows Management Instrumentation from the Services panel. Sometimes it makes WMI stop using my CPU, other times it eventually starts up again.

Just a quick update to note for people who don't mind using an older version: 5.3.10 doesn't appear to have the WMI problem. It does have the old bug where Precision Server runs all the time regardless of OSD, but you can just change the server.exe filenames to get around that.
 
Hopefully EVGA gets the new versions working properly soon.

OMG a reply. Its like a Sasquatch sighting just happened. 

Wiser87

EVGA_JacobF
Just an official post in here that we are looking into this, we have not yet identified root cause but it is being worked on.

Thank you for the reply Jacob.

+1 

Does the server have anything to do with this?
 
I had asked previously if there's anyway to install just PrecisionX without the server but there's not. You can still install Afterburner without the rivatuner server and I do see constant activity from the server in Process Monitor even when there's no OSD.
 
I wish there was an option to simply install PrecisionXOC without the server for those of us not using the server features.
 
But anyway, I'm still seeing the behavior mentioned in this thread in 6.0.7. Afterburner uses about .5% CPU for me while PrecisionXOC is always at 2% with spikes to 5 or even 6%.