Sudden black screen, Unrecoverable, Fan spins up, Requires reboot

Toggle sidebar Toggle sidebar
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540
Been having this issue for a little while now. Looked for solutions all over and my hunch (and my fear) is that my mobo is dying.

My graphics card will sometimes randomly black screen and the fan gets really loud. The windows error noise plays after a few seconds and then the monitor declares that no input is detected. The only way to recover is a hard reboot. I've only had it happen on a couple games, all of which can be graphically intensive: Skyrim, WoW, Arma 3. It's happened on rare occasion during some program errors, but those have been very rare.

I've tried confirming that all the power cords were securely in place, even swapped some around (my GFX card has two power ports, an 8 pin and a 6 pin, and my PSU cables are 6+2 pins) and for a couple days that seemed to work, but Im having issues again today. I even set up a custom fan curve through MSI Afterburner to make sure the card wasnt overheating.

I was able to record my graphics card stats via GPU-Z the last time it happened, and to the best of my knowledge its not the GPU's fault. I uploaded the log file to mediafire, since pastebin screws up the formatting. Windows Event Viewer lists an "unexpected shutdown" at "2016-06-23 00:09:21", so I looked in the logs within the minute and a half before that and didnt see anything denoting overheating or drawing too much power. Log file: https://www.mediafire.com/?08fc8abkbx20i88

My PSU is an Antec 1200W Platinum, so I have a hard time believing Im running out of power, especially since a rough estimation using a power requirement calculator said I really only need around 600W.

The only thing I can think of other than my mobo dying or my ram being faulty is my CPU's overclock. For some reason my CPU came overclocked to 4.3 GHz. I've tried telling the CPU to go back to factory defaults, but that only works until I reboot the computer, then it goes back to overclocking.

My hardware stats are below:
OS: Windows 7 Professional 64 bit, SP1
MB: ASUS P9X79 LE LGA 2011, BIOS version: 46.04
PU: Intel i7-4930K @ 3.40GHz (12 CPUs)
RAM: G.Skill 2 x 8GB DDR3 1600
GFX: NVIDIA GeForce GTX 770
PSU: Antec TPQ-1200 1200W

Thanks for all your help.
 
Solution
hang-the-9
Fast way to see where the issue is, test the video card in another system that can run it. Without removing variables as to what may be the issue you are just left with replacing parts one at a time till you fix it. Need to test the parts first unless you just want to go into buying new things. Does not sound like a motherboard issue. Bad video card or power supply are the top things to check on, maybe RAM. Test with one stick of RAM at a time.
Sort by date Sort by votes
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040
sounds like an electrical issue.

Now before you think i'm saying it's the power supply, a LOT of things can cause an electrical issue. Shorts, loose wires, metal touching parts it shouldn't touch (grounding things out) a dying psu, a dying part (any number of things can die, or start to die and cause random power issues), failing power/reset switches on the case, failing fans, not enough power to the CPU or RAM; a failing GPU

as I said. a LOT of things can cause this issue.

 
Upvote 0 Downvote
hang-the-9

hang-the-9

Titan
Mar 25, 2010
60,387
1,081
152,940
Fast way to see where the issue is, test the video card in another system that can run it. Without removing variables as to what may be the issue you are just left with replacing parts one at a time till you fix it. Need to test the parts first unless you just want to go into buying new things. Does not sound like a motherboard issue. Bad video card or power supply are the top things to check on, maybe RAM. Test with one stick of RAM at a time.
 
Upvote 0 Downvote
Solution
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540
I should add that this has never occurred while doing graphics card stress tests.
To test the RAM all I need to do is remove a stick and run the trouble programs I mentioned before, right?

Thanks for all your answers so far.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


What's the proper way to go about testing that?
I know that's a setting I take care of in the BIOS, but I've never touched the voltages for anything.

Thanks.

 
Upvote 0 Downvote
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040


well, since your problem looks electrical the FREE and EASY tests almost all include using the bios to boost voltages, you're gonna have to jump over that barrier.

Unfortunately i'd been half a decade since I overclocked an intel extreme edition chip or played in the bios for one. So i can't be too specific about what you need to do

1) bump the vcore a little bit.
2) bump the dram voltage a little bit.

not much else to touch. I would start with the vcore cause if it is an undervolting problem that's gonna net the most results.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


So this is what Im working with, if this is any help.

3d74294c76.JPG

According to the specs for my RAM, it already has the proper voltage and timing. However, as best I can tell, the VCore is 0.2GHz below what it should be. Apparently "Offset Mode" plays into that.

I checked up on what "Offset Mode" is, supposedly increases the available voltage to the CPU as the load increases, preventing the CPU from running higher and lower workloads when unnecessary.

Should I take the CPU out of Offset Mode and switch to Manual? Or should I just adjust the Offset Value? Any ideas?

Thanks again.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540
Another thing to add is that, like I said, for the most part these occurrences are random, but they always take place while the game is done loading, but still rendering. The few times I've been able to replicate it has been with the occasional outdoor Skyrim save with some moderately intensive scenery. The black screen doesnt happen if I revert to an earlier save that is indoors or less intensive.

I didnt start getting these with Skyrim until I changed the values for memory allocation to higher (but still recommended) values. I think this supports the idea that the RAM is the culprit in one way or another.

[Memory]
DefaultHeapInitialAllocMB=768
ScrapHeapSizeMB=256
 
Upvote 0 Downvote
M

maxwellmelon

Splendid
Oct 2, 2013
2,511
0
24,010
" The only thing I can think of other than my mobo dying or my ram being faulty is my CPU's overclock. For some reason my CPU came overclocked to 4.3 GHz. I've tried telling the CPU to go back to factory defaults, but that only works until I reboot the computer, then it goes back to overclocking."

this is exactly what my pc does while overclocking the cpu and going to far. the fans go to full and the screen goes black. so I just don't get why your cpu keeps defaulting to overclock settings as the default for that cpu is only 3.4 ghz so to run 4.3ghz stable you have to add a lot of voltage.

so I think we need to get your cpu to run default settings first

so brings me to following question have you reset the cmos by moving the jumper on your mainboard for 5 sec then putting it back? (plugged into wall but system off) does it auto overclock after this?

have you tried disabling "turbo mode" a from of auto overclocking?




 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


I haven't tried messing with any specific BIOS settings other than setting it to factory defaults. I've never tried or messed with Overclocking, so I was afraid to touch anything other than that.

On a side note, I also just tried moving my GPU to a different 12V rail than my CPU and radiator. I have yet to test it, and am not sure if that's even something that will help, but since power seems to be the theme of possible issues, I figured it couldn't hurt.
I also uninstalled my aftermarket sound card, since I was no longer using it. I also reseated the GPU, the RAM, and all my power cables.

I'll figure out how to reset the CMOS after I test the changes I've already made, just so we're not combining too many variables.

Thanks.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


So as it turns out the automatic overclocking is part of the "Performance" preset in the BIOS. The reason it kept overclocking itself is because when I hit "factory defaults", I would then turn around and set the preset back to Performance, which is why it only stopped overclocking until I restarted my PC.

I've enabled the "Standard" preset, but it instead turned on Turbo mode. I know how to turn that one off, but it looks like Turbo is only maxing at 3.9 GHz instead of 4.3 GHz, so I figured I would test it before I did anything.

I also stepped up my DRAM to 1600 MHz, since it was only running at 1333 MHz, and was marketed as being tested and stable at 1600 MHz.

I'll test again with these settings and report back.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


I've turned everything to defaults now. CPU is running at 3.4 GHz and DRAM at 1333 MHz.
Still getting fan spin up and black screens.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


Alright, so I guess I'll start on trying to bump the VCore and DRAM voltage next.
If that doesnt work, then thankfully my PSU, GPU, and Mobo are all still under warranty.
 
Upvote 0 Downvote
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040


I know some overclockers who swear by offset mode overclocking. I'm not one of them. Change the mode from offset to normal, that should give you the option to adjust the vcore normally.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


I went for two or three days without an issue, but then randomly got two occurrences within an hour, hour and a half while playing WoW, so Im about to try and tweak the VCore a little bit and see what happens.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


Okay, I've set VCore to Manual, and bumped it to 1.1V.
On BIOS screen it was sitting at 1.014V, so I dont know if a ~0.1V increase is too high or not. I figured worst case I could tune it down until I encountered issues again. BIOS said 1.7V was maximum.

I also noticed a setting called VCCSA Voltage. It claims to affect the PCIE controller. Should I adjust that as well, or stick to just VCore? Pic below.

12935dec23.JPG
 
Upvote 0 Downvote
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040
just work with the vcore. and you're good with the vcore up to 1.25V without any real danger. frankly, you'll probably need as much as 1.30-1.35V on the vcore for 4.3ghz (I looked it up); I wouldn't go much higher then that without some knowledge of what you're doing. Keep an eye on temps as you add to the vcore.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


I've turned the overclock off for now, until I can figure out what's going on.
VCore at 1.1V and still experienced a crash. I've turned it up to 1.2V, and the BIOS is informing me I'm running the setting above specs, but like you said, Im not in the danger zone yet. We'll see what happens.

This issue just occurs at the most random times, and I dont understand it at all. I just had it five minutes ago while loading into the main menu of Overwatch, a game which has NEVER given me the issue before. And just before that I was running Skyrim, which is modded like crazy and really puts my system to work. Had no issue playing that today.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


1.2V has not helped. The last crash caused me to lose all my option settings in WoW. Everything has been set to defaults.
I'm starting to pull my hair out over this.
 
Upvote 0 Downvote
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040
Get the vcore voltage up to 1.35V, and try bumping the dram voltage a little (i think it was at 1.5V, try for 1.55V). I saw the setting in one of the screenies you posted, so it's on the same page as the vcore.

If you still have an issue then my next look would be to the gpu's voltages, hop into MSI afterburner and push the power limit to max.

See how that works for you. If all of this doesn't work then you probably have something failing. and the next step will be to breadboard the whole system and see if you can't isolate the part or wires that are failing and causing the power issue. If breadboarding doesn't turn up anything then your issue is likely power supply related; replace the power supply, see if the problem goes away.

Problem you're going to have is if you reach this step, without discovering a culprit, is it still could be a faulty motherboard or gpu or even cpu behind all of this. My money on likelihood would probably be PSU>Motherboard>....>GPU>...>...>...>Ram>CPU. those ... represent a steep decline in likelihood.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


Should I continue to bump the VCore despite my CPU not being overclocked anymore?
Just double checking. Thanks.
 
Upvote 0 Downvote
ingtar33

ingtar33

Glorious
Dec 17, 2012
9,024
213
46,040


ah... it should be fine at the voltage I suggest.
 
Upvote 0 Downvote
TheSelverFang

TheSelverFang

Honorable
Nov 30, 2013
49
0
10,540


Alright so, my PC did alright for a while at 1.25V

Then today I experienced about 5-6 occurrences back to back within 30 minutes.
Decided I wasn't getting anywhere with voltage and swapped out the GTX 770 for an older GTX 660 Ti I still had lying around.

The occurrences have stopped for now. Im going to do some more testing before I call anything conclusive, but it looks like it was my GPU. Which is good, because that's easy to replace.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom