PLEASE HELP - New PC Problem

the0utcast

Reputable
Jan 12, 2015
15
0
4,510
Hey all,

So I built a new PC. Here are the parts and specs:
https://ca.pcpartpicker.com/user/the0utcast/saved/7LQhP6

NO overclocking. All updated drivers. Windows 7 Home Premium x64.

The issue:
My pc will randomly go to a blank light blue or dark grey screen and then become unresponsive. If there's a video or music playing, the sound can still be heard. No hardware changes are noticed... the system continues to run without change. The only way to resolve the issue is to hard reboot. I have also noted that when the 'freeze' occurs, my HDD light doesn't come on at all.

Windows event logs show nothing except the kernel-power error from when hard reboot.

I stress tested all the components using Prime95, OCCT, memtest, ect, ect... Everything was fine, the only thing that resulted in strange activity was the PSU test in OCCT. I have the graphs that it created, I'll post them below. What happened during the psu test was at first everything seemed fine, fps was stable at 200, temps were all normal, everything was fine. Then all of a sudden, the fps dropped to 14 or so after about 10 minutes in. No temp changes. No voltage changes that I can tell. I happened to have CPUID HWMonitor running as well, and noticed the GPU utilization jumping from 0% to 100% within seconds. That's the only thing I saw that seemed strange.

I'm not sure if that means I have a GPU issue (because fps dropped) or a PSU issue (because it was a PSU test). When I did the GPU test, it passed with flying colors and no issues, so I don't understand.

Here are the graphs from that test in a rar folder:
http://www.megafileupload.com/en/file/597776/OCCTgraphs-rar.html

There were no issues when building the pc... Please help. I have no idea what's happening.

EDIT: I noticed one of the graphs is of the voltage... and there is a severe voltage drop... hmm...
 
Is the PC plugged in directly to a wall or is there some kind of surge protector or ups in between?
It could also be some kind of power saving mode that kicks in and fails to recover afterwards, have you tried another powersaving profile in the bios? Or disabling this feature altogether? You can check this also in windows power profiles.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


Okay, this has turned into a ghost hunt. I found a setting in the bios to turn off any power saving. The windows power saving has always been set to full performance. I then connected the pc directly to the wall... When I booted back up, it booted into a totally different user account, or so it seemed... All my settings were changed. Areo was re-enabled, all my desktop icons were gone, I couldn't even find google chrome installed. Like WTF. I'm reformatting now. This is turning into the weirdest thing ever.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


Okay, so I'm back up and running... Everything seems faster this time around. I haven't gotten to get in game or anything, but it's been running for a while now with no issues. It'll be a few days before I can confirm if it's fixed or not though as it happens pretty randomly...
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510
Happened again... I was playing and streaming Alien Isolation and I got a BSOD saying BAD_POOL_HEADER. I rebooted, started streaming and playing again and it did the other kind of crash where the screen just goes grey and becomes unresponsive, except now the sound freezes up too.
I can't figure this out and it's really ruining this new PC I spent all my money on... SOMEONE PLEASE HELP :'(
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


I'm not using malwarebytes. Not installed, it's a fresh format. I have AVG, steam, VLC player, and CCleaner installed. That's about it.

The last crash produced this error in the event viewer:

Source: WMI EventID: 10

Details: Event filter with query "SELECT * FROM __InstanceModificationEvent WITHIN 60 WHERE TargetInstance ISA "Win32_Processor" AND TargetInstance.LoadPercentage > 99" could not be reactivated in namespace "//./root/CIMV2" because of error 0x80041003. Events cannot be delivered through this filter until the problem is corrected.

There are a good number of these errors.

Here are some other errors in event viewer:

Source: PrintService EventID: 512
InitializePrintProvider failed for provider inetpp.dll. This can occur because of system instability or a lack of system resources.

Source: SideBySide EventID: 80
Activation context generation failed for "c:\program files\CCleaner\CCleaner.exe".Error in manifest or policy file "" on line . A component version required by the application conflicts with another component version already active. Conflicting components are:. Component 1: C:\Windows\WinSxS\manifests\x86_microsoft.windows.common-controls_6595b64144ccf1df_6.0.7601.17514_none_41e6975e2bd6f2b2.manifest. Component 2: C:\Windows\WinSxS\manifests\amd64_microsoft.windows.common-controls_6595b64144ccf1df_6.0.7601.17514_none_fa396087175ac9ac.manifest.

Source: BugCheck EventID: 1005
Unable to produce a minidump file from the full dump file.

Source: e1dexpress EventID: 27
Intel(R) Ethernet Connection I217-V
Network link is disconnected. ******THERE ARE A BUNCH OF THESE
 
Try booting indo safe mode and see if the problem still occurs. If it doesn't it probably is a software issue. Try to uninstall ccleaner or even AVG. Download another antivirus first, I can recommend avast.
If the problem still occurs in safe mode it probably is a hardware failure.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


Okay, I'll try this... I've noticed something else. No matter which program I use, I can't find a monitor that will tell me what my +12V is at. HWmonitor only shows my -12V. Any ideas?
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


Okay, the +12V fluctuates between 12.168 and 12.096 when I'm stress testing with OCCT... Not sure if that means anything. I'm going to set up a camera and try to get it to crash again.
 
That fluctuation is normal, if it were varying over 0.75v either way then it could be considered like an anomally.

Having said that, I just took a look at the graphs you uploaded, clearly the PSU is failing, every time your CPU tries to ramp up its clock speed to 3.90Ghz (due to Turbo boost) the vcore drops down way deep instead of keeping itself at the same value or increasing as needed to sustain such clock speed, this results in those FPS drops and could easily lead to system instability you're having right now.

Temps are fine so either the PSU is the problem or the CPU is faulty (though in that case crashes and instability should be more often).

EDIT: this won't help to isolate the problem, but I guess it's also possible that the VRM/Phase modulators in your mobo could be the reason of this vcore dropping. Best thing to resolve this would be to first swap the PSU for another known working unit and see if problem persists.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510
Okay, so... I looked into the case just to make sure everything was set up alright... I decided to take off the psu intake filter to clean it, and that's when I noticed something... The bottom of the case doesn't give a lot of breathing room for the PSU. I cleaned the filter and decided when I put the computer back, I'm going to elevate the case so the PSU can breathe... So I did that and have not had a single crash or hiccup yet. I've thrown games at it, stress tests, and everything. It seems solid now. There's barely a drop in voltage during OCCT's psu test now. I think it's fixed. I'll let you know within a few days to see if it happens again.

So, assuming this fixed it... the problem was the PSU not getting enough air. If not, it might be something else.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510
I played and streamed Tomb Raider for 6 hours straight today... No hiccups or crashes or anything. If I make it through tomorrow without a crash, I'm going to close this thread.
 

the0utcast

Reputable
Jan 12, 2015
15
0
4,510


I spoke too soon. I got hit with this just now:

http://oi57.tinypic.com/2yy1myx.jpg

This is really taking the wind out of my sails about having a new PC. I'm starting to think it's hardware... But what the hell could it be, it's a brand new rig with all new parts in it. And everything does fine on stress tests. Like...
 
Looking at your graphs I think this can be:
1. A failing psu, not able to provide enough power to the system under load. If possible get another psu to test the system under load
2. A cpu cooling problem, to prevent the cpu from permanent damage the cpu is being throttled down. Double check your cpu cooler, if the thermal paste inbetween cpu and cooler is applied well and if the cooler doesn't get too hot (faulty fan).

I think you should look in the direction of your psu though, since it keeps dropping vcore and also the fps goes down after a few minutes of stress testing. If you have another smaller psu you can test the cpu alone by removing the gpu and running on the integrated graphics. See if OCCT finds vcore drops again this way.
 
Noone will be able to help, you will need to find a way to test the different components of your system separately as suggested before. This is the only way to isolate the problem. If you don't have any spare parts you can borrow, consider taking the computer to a computer repair shop and let them test it with a different psu or run some stress tests, each time changing out components.