Random crashes to a randomly colored screen, computer unresponsive

Status
Not open for further replies.

vreazan

Honorable
Jul 16, 2013
10
0
10,510
This is my first post and so I hope it is in the right section and if not hopefully a moderator can move it. I have been having some issues with my desktop since I initially built it last november, while most have been insignificant I have had one glaring issue. At times when gaming the computer will crash to a randomly colored screen( not bsod). The computer will be fully unresponsive and requires a hard reset to turn it back on. This only occurs when playing games; however, this has a caveat. It crashes during league of legends primarily, occasionally during fifa 13, but never does during higher stress games like battlefield 3, running it on max settings. The crash would generally be accompanied by a whirring sound, not from the tower itself but through the audio, which I suspect was the last fragment of audio just on a very frequent short loop repeat since it was the lat sound input. The action center would show that there was a display error when reviewing the crash. Sometimes the crash would be after a few minutes of play, sometimes after hours. It could crash, I could power it off and let it sit for a short while, power back on, boot up league to play, and within a couple minutes it crash. So time does not seem to be a factor in when it crashes(though it did seem to happen more often early or that it would be a decent amount of time before it would happen. So either <5min or>1 hour)

Initially I was confounded and could not determine the problem, using programs to monitor the components my cpu temperature was never exceeding 42 degrees, ruling out over heating. The voltage stayed steady and seemed to not be experiencing any issues. I ran stress tests on the graphics card itself running it at 90% max capacity for two hours without even a hiccup. At this point the logical response would of course be some form of driver issue and it turned out it was. I had thought the drivers were up to date, using the disc that came with it but silly me, I of course needed to go to the site. Once I updated the drivers using the site,this problem completely disappeared. I continued to vigilantly check to make sure I had the newest drivers to avoid a repeat of this issue. A month of no crash no problem experience, but that did not last.

A month after I properly updated the drivers the problem has reappeared. My drivers for the video card are the most up to date available. The problem now solely appears when playing league of legends where before it would also sometimes occur when playing fifa 13. Other notable differences are that now windows does not report a display error, only reporting an unexpected shutdown, from the hard reset. Also where the crash would generally be accompanied by some form of noise with it(loud buzzing audio), the computer now solely crashes to a random color and is unresponsive. I have done clean installs of the latest drivers in case they did not install properly. I have done a clean install of league of legends to see if the issue was some weird client to gpu glitch. I then did a system restore to the day before the crashes began along with another clean install of League and gpu drivers. After the restore I managed to play 2 full games without a crash, after that it began crashing after a few minutes of play. I have probably played 15 hours of other games without crash in the last several days, but in league it is a guaranteed crash. Also have ran malware bytes to scan for anything malicious and used cc cleaner to clean up the computer some.

This leaves me pretty much out of ideas as to what to do or what could be the cause. And any help would be so very appreciated.

List of Hardware

Thermaltake TR2 430W Silent Power Supply

Samsung 24X DVDRW SATA OEM

WD Caviar Blue 500GB SATA HD 7200/16MB/SATA-6G

EVGA GeForce GTX 650 Ti 2GB GDDR5 PCIe

Two- Kingston 4GB 1333MHz DDR3 CL9 HyperX Blu

AMD PHENOM II X6 1045T THUBAN 2.7GHZ AM3 CPU

Asus Amd 970 Atx 4 Dimm Ddr3 Crossfire

Microsoft Windows 7 Home Premium 64bit



 
Solution
Case ventilation first: either online or at local computer shop buy a couple more 120mm fans. Put one as intake on the bottom and one on the front panel as intake.
Second: The way we figured it was that when we played a game or something with graphics (3DMark11 looped for a couple hours) the fan on the GPU would spin up and keep it cool enough. We are pretty sure it wasn't the GPU chipset itself that was getting to warm, but some other peripheral part on the video card. When the card was just idling along while he surfed the web or emailed someone there wasn't enough air flow to keep that part cool, when the card was pushed the case fans AND the graphics card fans pushed more air.
That's the best we could come up with. About 6 months...
What case is this housed in? How many intake and exhaust fans?
Make certain you have enough air flow into and out of the case. Don't count your PSU fan, it really is just for the PSU.

Had this problem on friend's Alienware with an OEM GTX460. His would randomly crash (not a gamer) to a pink screen.
The solution took some investigating and we never POSITIVELY identified the problem, but believe the Alienware did NOT spool up the case fans soon enough when the case ambient temperature rose; and the stock fan settings on the 460 did not spin up fast enough nor spin up soon enough.
We used MSI's Afterburner software and set the GPU fan profile to run at a minimum 42% and spin up way earlier then the stock profile (which had the fan at about 30% most of the time). It solved the problem.
Worth a try.
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510


Anything is worth a shot at this point, I know one issue with my rig set up is that it does not have the best ventilation/fans. My only thought about this has been though that why would this only be the case with a very graphically unintensive game, when I have no issues running a very intensive game like battle field 3 on max settings without any issue. Or in other words why would there be heating issues/fans not cooling enough for League but not for other more demanding games?

Also how exactly does one change the minimum fan speeds? Is it a simpel case of download the program you mentioned and play with settings?

And the case is a Thermaltake V3 Black Edition Mid Tower Case
http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=5520044
 
Case ventilation first: either online or at local computer shop buy a couple more 120mm fans. Put one as intake on the bottom and one on the front panel as intake.
Second: The way we figured it was that when we played a game or something with graphics (3DMark11 looped for a couple hours) the fan on the GPU would spin up and keep it cool enough. We are pretty sure it wasn't the GPU chipset itself that was getting to warm, but some other peripheral part on the video card. When the card was just idling along while he surfed the web or emailed someone there wasn't enough air flow to keep that part cool, when the card was pushed the case fans AND the graphics card fans pushed more air.
That's the best we could come up with. About 6 months later he replaced it with an ASUS 660. (And THAT cost me money, cause the performance was so much better I needed a couple)
Third: Yes it is simple to set up. EVGA has basically the same software named PrecisionX, but I found it harder to work with and to read on the monitor. Down load either version, EVGA or MSI. (If MSI don't install Kombustor). Install the software, then click on settings, click on the Fan tab. You will be presented with a graph and you can set the minimum (no less then 40%) fan speed and at what temperature it will start increasing. This explanation is harder to understand then the actual software.
You'll want to play around with it as there are things I turned off (hints and tool-tips) and, well, you get the idea.
I would set a minimum fan speed of about 43% @ 40C, 50%@50C, 60%@60C, and I think that 650 will max out at 73% shortly thereafter (about 65C).

Repost please, let us know how it goes. Working on my friends PC was a Mutha! (I took his 460 home and ran it for two days straight on my PC without a hiccup, including 14hrs. of 3DMark11 looped on highest settings; he had one of my 460's and never had a problem; returned his card, and by the time I got home 10 minutes later he had a pink screen!) I think we just lucked out. I reallyhope this works for you.
 
Solution

vreazan

Honorable
Jul 16, 2013
10
0
10,510


Thank you so much for trying to help. I just tried adjusting the fan settings before booting up league. I started up a custom game and let it sit there for about 30 seconds then switched over to the precision tool to look at what it was showing. It wasn't even breaking 30c which is about what it sits at when first booting up. I don't know if it means anything but the gpu clock was running at 900ish mghz and memory clock was sitting at about 2700mghz with game running. The gpu clock setting shows it set to 324mghz and voltage to 875mn Like I said I don't know if this means anything just what I observed. The comp crashed again about 1.5 minute into the custom game when minions spawned, which would be at a sudden increase I guess in stress. I just went through the full system event logs from the last hour, there is not a single event log from the moments coming up to the crash, just logs from once it boots back up.


Just ran fifa and the card is running at the same temperatures and mghz as with lol, yet it doesn't have crashes
 
It's very late for me and have a long busy day ahead, so this will be my last post for the night; I won't get much time tomorrow and won't get home before 10p.m. but WILL get back on here as soon as I can.
I DON'T think this is a powersupply issue,but may end up being such. Your Thermaltake is not a reliable psu; it is in Tier 5 in this list:
http://www.eggxpert.com/forums/thread/323050.aspx
I guess if it was acting up and/or not producing enough power for the video card AND system it could do something strange like this, but have my doubts.
Just the same, do you have access to a good known working 400-500 watt PSU (preferably Corsair, XFX, Seasonic, PC Power&Cooling)? If so, try swapping out your Thermaltake with that unit and see if anything different happens.

It could be the video card. Can you borrow someone nvidia card and try it in your machine, and have them try yours in theirs?
(Ideally in their machine playing LoL).
Man, I have to go to bed. Can't think straight. We'll work on this tomorrow.
Anyone else out there have any ideas plllleeeeeaaaassssseeee jump in and help.
'night
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
Ok so tried a handful of things today. I noticed quite a few error reports involving adobe air, tried up dating that to no effect. Stress tested gpu, ran for 2 hours no problems. Did a full malware bytes scan of the whole computer and nothing showed up. Tried uninstalling my graphics drivers and installing an older version, didn't help any. Reinstalled current drivers and something did happen though. Right after the newest drivers installed, about when they went to take effect, it crashed like it has in game but with no other programs running then geforce experience. Hard reset and rebooted and it crashed again in the same manner within a couple minutes this time with only action center programs running and chrome. Reset again and no crashes since, outside of games. I was starting to really think this had to be something with league and software related; however, the two crashes with no stressful programs running seems to rule this out.

At this point I am thinking it might well be the psu. If it is the psu dropping its power level causing the crashes that would maybe explain the crashes post boot with no other things running. But honestly I just don't know. I already tried a couple days ago recovering to a point before the crashes, that didn't help. Since that did nothing this is probably indicative of either hardware or windows itself, something that isn't changing or the recovery I'd imagine should have fixed this.

I've looked up so many similar cases online but they are rarely resolved, or at least not posted the solution if they are resolved. One common theme seems to be that the card itself ( many different cards not just nvidia cards ) have had similar issues and in the end the card seems to not be the problem, at least tests, heating, etc don't seem to indicate the card itself being the problem. Often times people's advice point to either the psu or the motherboard being the ultimate issue, which is what I'm leaning towards now.

The way I see it is that I have tried and tested about everything I can think of short of the following:
Swap psu
motherboards issues
updating bios

Talking to a friend he recalled a friend with a similar problem that persisted even after getting a newer card, but after thinking about it he stopped seeing the problem after getting a new psu. So I guess that will be the next thing I try. My only thought is that I still don't understand why it would be that the least stressful game I play encounters psu voltage dropping but not more intensive activities. My other thought is that if this same issue was resolved before via a graphics driver update, it leaves me unsure on if it is really a hardware issue. Short of the driver being more efficient and reducing voltage supply or something then that problem seemed to be purely software. This is of course counterbalanced by the fact that the problem has resurfaced when the issue this time is clearly not the gpu drivers.

Or in other words ultimately I have no real clue what the exact issue is and everything seems to conflict and point to different problems. I'm lost. I guess I'm going to try a psu swap and see what happens but ultimately the fact this problem has recurred makes me totally unsure if this is hardware or software. The first instance was software( at least seemingly) but this time the same issue seems to be hardware.

If anyone else has any clues I would love the help! This is so confusing

Edit- one last thought I just had is that after each crash I've looked at the full windows logs and all system events and with every crash there are no logs before leading up to the crash point spare standard system messages. I'm not sure but this is making me think even more that it might be the psu. If there is something malfunctioning I would think that there would be some form of error log. There aren't any though meaning that the moment whatever happens happens the system is fully dropping, so hard the system doesn't even have time to log anything. If its voltage dropping, this could explain that. Voltage drops-system fully goes down immediately- end of story, log wise, till reboot.
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
Ok I just updated the motherboard chipset drivers and made it through a 50 minute game of lol without crash, whereas before it was crashing within the first 3 minutes. I'm going to cross my fingers and play some more games to see if this solved it. The other day before it was full crashing in first 2 minutes I made it though 2 games before it crashed so this may or may not have done the trick. Will post tomorrow if things are still running smooth or later tonight if it crashes again
 
Time to try a different power supply. IDEALLY it should be a psu that you KNOW is working properly and in good condition.

Also time to try a different video card. If you have to, buy a new nvidia card, your model or a 660, with the thought that you can return it if it doesn't solve the problem.


It's probably a good idea to install this program
http://www.hwinfo.com/
Run it and look at your voltages. Post a screen shot if you can.
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
http://s10.postimg.org/was075iy1/Capture.png

This is a screen shot of everything about 10 seconds before the computer crashed. I had a game of league running. As soon as I saved it it crashed.

One other oddity that has popped up is that my windows explorer, aka the desktop, has crashed and been bogged down a couple times immediately post boot. This started occurring a few months back when I had these crashes before and stopped when the crashes did. It's back now, not sure what this is indicative of.
 


Can't see any temps nor voltages in your picture. Want to try again?
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
http://s9.postimg.org/4amsxz3mn/Capture2.png

Something unsure if I noted earlier is that since these crashes started when games first load up on league I have a whopping 1 fps. Till it stabilizes out after about 10 seconds back to my normal 60 fps I have it set for.

Screen cap was taken during the 1 fps, promptly crashed about 20 seconds in, so 10 seconds after I saved the image
 
CPUID HW monitor indicates the 3.3 voltage is way down, just at 5% low. Normally (!) equipment can tolerate up to 10% and though this sounds like it is o.k. at 5% we have no way of telling the quality of the voltage on that rail. PrecisionX shows everything within specification, so which one to believe? Well, the PCI-E x16 slot does feed 3.3 and, if memory serves me correctly, 5 volt to the video card (Older top notch PSU carried a lot of amp. on those two rails and not as much on the 12v).
So the only theory I can make is that there is some problem on the 3.3volt rail.
Brings us back to trying another PSU.
And after that a different video card.
You are running out of options and you have TRULY worked your butt off trying to isolate this problem.
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
Ok i'll have to hunt around to see if I can find a psu which will work, most people I know around here have laptops, and we do have a few old desktops here but my uncle says that they are so old they may not even work with my rig. I'll see what I can do and let you know how it goes, thank you so much for the help! I have no clue about voltages or half of what you posted meant so I never would have picked up on that haha.
 


REMEMBER, it needs to be a working PSU; preferably something out of an actual working PC. And it needs to have at least total power of 400W, and 20amps. on the 12V rail/circuit, preferably a single 12v rail.
 

vreazan

Honorable
Jul 16, 2013
10
0
10,510
Going to try a psu swap when I can. I checked the voltages on the bios just to confirm though, and the bios showed the 3.3v running at 3.298, which seems to conflict with the cpuid's reading, currently at 3.132. Oh well, still going to try a psu swap when I can find one to use.
 

shnizzaga

Honorable
Oct 18, 2013
1
0
10,510


Vreazan, did swapping the PSU ever stop the crashes? Ive been having similar problems with my computer and am desperately looking for a solution
 

ThereAre

Honorable
Nov 28, 2013
1
0
10,510
Hi vreazan,

I am experiencing the same problems that you described. i have tried a clean install, runned memory tests etc, but so far no clser at finding the solution. Could you please let us know if you have managed to solve the problem?

Regards,
Tim
 

dToast

Honorable
May 4, 2013
45
0
10,530
Hi,
I'm bumping this thread because I have EXACTLY the same problem as stated by OP. Funny thing is we are running on the same PSU... I'm wondering if this confirm the issue is PSU related?

Any help is appreciated.
 

DSMonaco

Reputable
Aug 23, 2014
1
0
4,510
I'm having a similar issue, if not the same. It seems to only be happening when i play ArmA 3.
This has been getting on my nerves and i've been searching the internet with no luck for days now. I've installed all the up-to-date drivers for everything. I've gone down to the basics when i first turned my PC on and ran the set-up and nothing has helped. In fact, arma 3 seems to crash more and more, especially in Multi-player.

My Specs/Build:
EVGA Geforce GTX 650Ti Boost (Overclocked)
MSI Z87-G43 Motherboard
Intel i5-4670k @3.40GHz
Corsair GS700 Gaming series
8GB Rip jaws gaming series ram
Roswell challenger medium case
Windows 7 Ultimate SP1

When playing ArmA 3, nothing out of the ordinary happens before a crash. i could be patrolling kavana in a hunter mrap one minute and the the next second i would have a white screen and a locked up PC requiring a forced restart.

i doubt its something related to the CPU because i can do a 8+flight on FSX and get 120+FPS to even 400+FPS at 38,000Feet.
and FSX is a CPU intensive program not to mention x32 running on x64.

My issue in depth:
Visual: Random Solid color screen (White, Cyan, Pink, Green, Black) White and Black happen the most pink and green are very rare.
Audio: Silent/No sound just a crash. OR when talking in TS3 i would hear them for a second and it would just cut out into silence.
(example: So wh----at a----r- y-- -oin- *silence*)

Every time this happens, i spam caps lock as i watch the light light up on my keyboard. then after roughly 1/2 a second my system is at total lock-up. the caps lock light doesn't go on/off no input nothing. always resulting in me holding down the power button untill my pc is off.

is this practically what you're experiencing too? has anyone figured out a solution?!
 

Cheronik

Reputable
Sep 8, 2014
1
0
4,510
Hey there, i had these problems over a year now and never found a solution that worked, until last week.
I downclocked the graphiccard by 150Mhz and suddendly i had no issues anymore, it even increased the FPS performance.
So, if you are out of solutions and lost all hope, try this.

Fiddling with the Graphiccard may lose you the warranty and you could mess up the whole card. if you want to do it its on your own risk but i also have to say, i never heared of anyone who experienced hardware failure after a mild downclock of the graphiccard.

Get some information how to downclock your card and what problems may occur.

i hope i could at least give you another way to solve the problem. tell me if it worked when you tried it.
 
Vereazan, did you try another PSU or replace the one you have? sorry I haven't been around, but as you can see others have hi-jacked this thread (to the others, that is a no-no. If Vreazan's answer after this doesn't solve Your problems please start a separate thread).
I just realized that you are using a tier 5 power supply. What's Tier 5 you ask? Please look at the last tier in the attached two links which talk about dependable and not-dependable power supplies.
https://community.newegg.com/eggxpert/computer_hardware/f/135081/t/45344.aspx?Redirected=true
http://www.tomshardware.com/answers/id-1804779/power-supply-unit-tier-list.html
Might be necessary to upgrade to a tier 1,2, or 2b supply.
 

rayleighjean

Reputable
Dec 24, 2014
1
0
4,510


man ive been having the same issue intermittedly for a while now. sometimes ill be good for a week or two and then other times ill crash three or four times at various stages of boot and applications in the same night. its crazy because sometimes ill be able to game or run apps for 14 hours no problem and other times ill crash on windows boot, login, or just nothing. same thing everytime. warm black screen randomly. then ill spam caps or numlock until theres no response and hard reboot.

my specs:
asus r9 290x
asus m5a97
amd 8150 fx 3.6ghz 8-core
corsair 850 psu (upgraded from corsair 600 because i thought that might be the issue, didnt help)
patriot xms viper 3 2x8gb @1600mhz
cooler master xb haf workbence case (upgraded from some other one, didnt help)
corsair h80 liquid cpu cooler
win7 64 bit ultimate (upgraded to a legit copy because i thought it might help, didnt)

just ran a memtest64, took 6 hours, came back clean.
gpu drivers should be good.
tried two corsair PSUs (As above), with nothing

i dont know. any solutions would help. might try the underclock the gpu tonight.
 
Status
Not open for further replies.