PC Freezing and odd lines on display

[ukman408]

I'm not sure if this is a graphics issue or what but here's the deal. Just built my first upgraded PC. (original was built by ibuypower.com) Here's the Specs:

CPU: Intel Core i5-2500K 3.3GHz Quad-Core
CPU Cooler: Cooler Master Hyper 212 EVO 82.9 CFM Sleeve Bearing
Motherboard: ASRock Z75 Pro3 ATX LGA1155
Memory: G.Skill Sniper Low Voltage Series 8GB (2 x 4GB) DDR3-1600
Storage: OCZ Agility 4 128GB 2.5" SSD
Video Card: EVGA GeForce GTX 650 Ti 2GB
DVD: HL-DT-ST DVD RAM GH24NS90

I've also loaded after completely installing/patching the OS and drivers, two additional standard harddrives. Both out of my old PC. Ones being used as a data and game storage, the other is storing my system backups.

Five times since building the system on Friday 10/26 the system has randomly frozen on me. Straight up lockup, no BSOD, no errors on the lockup.... just goes unresponsive.

There's only one error I'm seeing in the event logs AFTER the freeze. There's nothing really showing what happened at the time of the freeze or just before. Just shows "The Previous system shutdown at (time here) on (date here) was unexpected"

Error #1 from event log (System)
----------------------------------------------------------------------------------------------------------
The ScRegSetValueExW call failed for FailureActions with the following error:
Access is denied
Source: Service Control Manager
Event ID: 7006
Level: Error
Keywords: Classic
OpCode: Info
----------------------------------------------------------------------------------------------------------

The first time it happened, I only had 1 monitor plugged in to the 650. I was fiddling around with a flash game on facebook while installing another game at the same time.

The second time happened again during a flash game. After the 2nd freeze, I disabled chromes built in flash player and installed the standard IE flashplayer on the system.

The third time it happened while in the middle of installing a game while doing nothing else.

The fourth time was while launching a game. It forced the monitor to change resolution, which is normal but a few seconds in it froze. Rebooted and the game has launched fine several times since then with no problems.

Now the 5th time was a little different. While waiting on an adapter for my 2nd monitor to come in so that I can run dual display off the 650, I decided to plug the 2nd monitor into the on-board video. I ran this setup fine for several hours last night with no problem, left the PC on all day/night to run game updates and came home from work tonight and the system was still running fine. I rebooted the system to make sure i was starting fresh and started doing some light webbrowsing. Just a few sites like huffingtonpost, espn, etc.....Again froze up. The only real difference is the left (primary display from the 650) did not change in its appearance. The 2nd one that I had plugged into the on-board, which again I had not been using previously, went off color. Sorta hard to explain but sorta grungy looking. On top of that, there were 4-5 squigly lines running across the top quarter of the monitor. They were moving slightly but not a whole log. I'm sorta lost.

Here's a few things I've tried.

updated all drivers and bios
still running all systems on factory speeds
disabled chromes internal flash player and installed the standalone IE flash player
reseated the video card and memory

CPU temp is reading anywhere from 28C and 35C
Motherboard temp is reading about the same

 

[ukman408]

Hrm...I wonder if I found the issue (i hope) I downloaded some stress test tools. One in particular kept causing crashes about 10 seconds in. I was using Prime95 and selecting the "blend test" which tests a little of everything, put lots of heat and uses lots of Ram. I took everything on the PC apart again and noticed I'd used a little too much compound on the heat sink. A very very little had sort of bled over the side of the silver cap of the processor onto the black part of the processor around the edge. I'm talking miniscule amounts. I cleaned all that up according to the processor instruction book, reseated every piece of hardware and cable in the box and started the test again. It's now been running that same test for 15 minutes with no errors. Gonna let it run at least a few more hours to see if that cleared it up. I know it was likely something that wasn't seated right or whatever but could that little bit of compound cause problems or do damage? It never spilled over the side onto the socket or anything, but again just spilled every so slightly in one spot over the silver cap onto the black board around it.

 

[ukman408]

Scratch that last post. About 10 seconds after I submitted it, the PC crashed again. Issue still on-going.

 

[ukman408]

Further update. I ran memory tests all night and they came back with no errors. I was praying it was memory as I'd seen several other issues like this be related to memory. The other thing I was thinking to check when I get home tonight is there's a software that came with the PC that handles integration of a discrete card and on-board gpu. I didnt really think about it at first since I wasnt using the on-board video the first 4 freezes but if it was still running, I guess it could be causing problems? Just thinking out loud hoping to spur some conversation and help

 

[nukemaster]

1. I did not know that had 1.25 volt memory or I would have some of that

I recommend getting hwmon and checking your temperatures. since you mention worry over the heat sink installation

http://www.cpuid.com/softwares/hwmonitor.html

You are right officially SB is only good for 1333 memory speeds. Most chips will take MUCH more(better with 2 sticks then 4).

It never hurts to get CPU-Z to see what speed your memory is running at(just for a second opinion.)
http://www.cpuid.com/softwares/cpu-z.html

If you want to eliminate the video card, it may be worth removing the dedicated card and running JUST the onboard for a bit to see if it helps.

From the bios, you can disable the onboard while testing the dedicated, but I have never seen it cause issues(anything is possible).

You may wish to test the memory @ 1.35 volts to see if it is any better. Those are just very good sticks picked with the ability to run @ 1.25. 1.35 will not hurt them.

Many things can freeze a system.

 

[ukman408]

Well I got home and actually checked the bios. It's up to date. On the memory, it was set to auto, and was set to the 1.25. I uninstalled that discrete/onboard gpu software. Rebooted and started another prime95 test. It froze almost instantly with the same thing. Monitor 1 (discrete) freezes, monitor 2 (onboard) does this funky thing. Link below.

http://www.youtube.com/watch?v=AzQNgxIYR9s

I really don't think its temps. I ran the HeavyLoad program with the asrock utility to watch the temps. Even at 100% cpu usage and the memory getting slammed, the pc was responding slowly, but was still responding. The temps never went above 45C. I'm just not sure what that first test on prime95 is doing. Whatever it is, is deffinitly showing the source of the problem. Guess I need to keep researching.

 

[nukemaster]

Prime is a load stability test. It finds prime numbers . One thing is for sure, you should not crash on it.

I would try to run some extended ram tests with memtest86+ just to be sure

http://www.memtest.org/#downiso

 

[ukman408]

I actually had already ran that memtest program for over 8 hours the other night with no failures. Last night I tried the things mentioned above and one other. Of all things I think maybe AVG was the problem? I uninstalled AVG after reading that someone else was getting false failures on it from Prime95. The test ran and completed without a single failure. Go figure....

 

[ukman408]

Best answer selected by ukman408.

 

[nukemaster]

The AVG thing is VERY strange. I mean I run AVG and just did a 13H prime95 run with high ram use to test my new memory.

http://imageshack.us/a/img822/1109/prime9413h.png

Then I failed IBT. needed 0.005 more volts to the cpu.