Solved BSOD at random times, at any time. Unsure of error.

I apologize for the poor title, i will try to give a better explanation of the issue.

Initially the issue started 3-4 months ago, at which time i was using Windows 7. Around the same time one of my 2 CPU fans stopped working but because my temperature wasn't going above 50 degree celsius i was hesitant to replace any part so i kind of just ignored the issue because every test i ran at the time came back as everything is fine and the only log i could find was Kernel-Power "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."; which doesn't tell me much because i was forced into manually restarting my pc. Around January of 2015, i upgraded the OS to Windows 8.1 but the issue persisted. A week ago my CPU fan died and i replaced it with the EVO Hyper 212 as well as installed a new SSD drive (samsung pro 850).

The issue is hard to explain because it isn't a BSOD with a message to contribute. The timing seems random and not linked to when the pc is under stress (gaming). Sometimes it can happen right after booting, other times it won't happen for hours and sometimes it can happen several times in a very short time period.

Now for the issue itself. I have two monitors, each monitor will pick a random color from the rgb scale and take on that color for the entire screen. Monitor one might be blue and monitor two might be pink, moving the mouse or using the keyboard does nothing and forces me into manually rebooting the system.

I have screenshots if that will help and have attached the zip file. I am also currently speaking to EVGA to run tests on the graphic card and i had the pc at a shop all week, they had been stress testing the card and said that there was no issues with it.

I received my pc back from the shop yesterday, spent all of yesterday and today doing installs/updates. Everything was fine, it was on for over 8 hours without any issues and the problem did not occur again until i installed the optional/recommended updates from windows update. I went back and removed the most recent ones, not sure if they are the cause or not.

Removed: attachment

You guys are my last hope, if you can make some sense out of any of this i would greatly appreciate it! Thanks in advance =D

Edit 1 8:36 PM EST April 2nd: Following EVGA's suggestions i ran Heaven Benchmark 4.0 and 3D Mark tests. I tried to upload the html file xps/pdf files but it gave me an error saying invalid file type. Any recommendations?

Edit 2 1:11 AM EST April 3rd: So the issue happened two more times today. There is no dump file that was created but the event view does have the following.
No mini dumps were created

Fixed!

So for anyone having a similar issue when playing games, your screen will either go black or any other random color and usually forces you to manually reboot.

Manually rebooting is what causes the Critical Kernel-Power 41 (63) error, this does NOT mean that your cpu is going bust. First off lower the settings of all games and switch to full screen mode. The error will persist and get more frequent in fact. The actual culprit which wasn't being logged beforehand is nvlddmkm. Now there are two ways that windows will address this, it will try to resolve the issue and hang in a loop until you manually reboot causing kernel-power failure or it will recover creating a warning only and repeating this until it either hangs in the loop or you quit the game.

The normal cause of this issue is defective ram, psu or gfx. As it turns out for me my graphic card was not stable unable it's native clock speed. Using EVGA Precision, i under-clocked the Core Clock and and memory clock by 20 and kept under-clocking until the issue stopped.

Both my GPU clock offset and Mem Clock offset are now set to -105 MHz. This is NOT a fix, this is more of a band aid solution. The graphics card is unstable and needs to be replaced.

If anyone else has anything to contribute or feels that i made a mistake in my explanation somewhere, please correct me.

Edit; If under-clocking by 100 MHz doesn't solve your issue then i would stop looking at the gfx card and look at ram, take out all the stick and try them one by one. If the issue persists with each individual stick, replace the psu, make sure all cables are properly attached as well!