Another issue. EVERY SINGLE DAY

Hardware Other Hardware
1

Hey guys. As some of you might have seen, I had an issue with my computer. It just kept restarting and restarting, but I solved it. Which is right here: https://teksyndicate.com/comment/1727068#comment-1727068

BUT, another problem popped up and I f'in hate it. This morning, I've played BF4 for about 2 hours without any problems at all. The computer was working like a charm, then after a couple of rounds, I left the game. An hour later, I decided to play another round, so I joined back to another server. The problem started after 5-10 mins. The screen started flickering with white net looking stuff, and both my monitors shut themselves off. The thing though, computer kept working but screens first said "no signal", then went to sleep mode. So I restarted the computer, and it was like nothing's ever happened. I used Chrome for a while, then played a couple of videos on VLC player; nothing happened. I joined back to the game, and it happened again. The same exact thing in again 5 to 10 mins. I tried other games to see if it's a problem with my graphics cards, and yes, it's happened again.

The other things I tried:

- Prime95 for 30 mins for CPU, RAM, and HDDs. No problems there.

- Clean reinstall of AMD drivers.

- AIDA64 on CPU and GPU separately, and bam. On GPU, monitors shut themselves down (without flickering), so I had to restart the computer once again.

- Called ASUS Support, they told me it's a PSU issue and hung up on my face.

- Called Corsair, they sort of relieved me when I told them the steps I took. The representative first told me that if it was a PSU related issue, the computer would've failed CPU and RAM stress tests. Then he told me to check out the VRAM temps while I do the testing, so I ran AIDA64 once more while watching the temps with both HWMonitor and CPUfan (fan speed on automatic). The temps went up to 90 from 50, gradually. And it again, shut the screens off. I noticed something, though. CPUfan kept GPU fan speed at 20% to 30% even though it was burning hot. Maybe that was the issue?

- Called ASUS Support, again. The gentleman told me that it's definitely a graphics card issue, after I explained all of the steps I took, and that they would replace it. So I filed an RMA.

Now, I have two more days before they send me an email asking for debit card information. Is there anything I could do to test it? It just started happening out of the blue.

Things I can do:

- CLEAN drivers installation. (I only uninstalled the drivers from Program Files, I could use extra software to make sure everything is gone this time)

- Asus GPU Tweak Fan Profile check? Since it stayed at around 20-30%, I thought maybe it's just shutting itself down because of too much heat?

- Different PCI slot test, to see if the issue is the motherboard. (By the way, motherboard's posting A0 on startup, which means a-ok, I think)

- Onboard graphics card test. The problem with this is, my card works fine on desktop unless I do heavy working. How am I supposed to test to see if it's the PSU without another graphics card?

 

Thanks a bunch, guys. This is the second major issue I'm having in just one week. Thanks!

2

3

how hot is your GPU getting? def sounds like a defective GPU

4

The last number I saw was 79 degrees celsius before it shut the monitors down (while stress testing it with AIDA64). I filed an RMA, but we'll see. I still wanna run the other tests before calling it a defective GPU. It's been two months and I only had two issues in one week.

By the way, I am almost absolutely sure that it's a drivers problem. I googled it, and found people talking about the same exact issue on different forums. I realized that I might have reinstalled the drivers after the first gaming session (the one I didn't have problems during gaming). I probably failed to delete all of the files before installing the new drivers. I'll try to figure it out, meanwhile, I'm open to suggestions as I still have two more days before RMA form comes in.

5

There's a few program's out there that remove left over driver files, I use CCcleaner, try it. Worth a shot.

6

i use display driver uninstaller from 3d guru

7

Well, you guys seem to have it down to GPU overheating, which sounds about right. Could it possibly be due to the PSU not being able to handle it under full load though?

8

Thanks for the answers! I'll reinstall the drivers, fresh and clean.

I thought about PSU not being able to handle it. But it's worked fine for 2 months, and now it started happening? That doesn't seem right. I have a RM850, and it should be enough to run two of those babies, maybe just barely, but yeah. IF it's a PSU problem, I'll make a deduction to see if it is. By the way, I'm using a UPS. Could it be possible that it's clogging the power?? I'll also run it without the UPS, and see what happens.

9

850W PSU should handle just most realistic setups. Not sure sure what your setup is, but unless you're sporting multiple very powerful GPUs...

UPS doesn't sound like a realistic problem, but I've seen stranger things happen, shouldn't be hard to check, just in case.

70-80 degrees Celsius doesn't necessarily sound too bad. My GPU was running at 70 with the stock cooler, and it's a fairly well ventilated case with dedicated intake and outtake fans. Nevertheless, I would try moving the GPU further away from the CPU if that's an option to see if it helps at all.

10

yeah, gpu heat doesnt seem so bad especially for a R9 290x. could be a bad vrm thermal contact? have you tried adjusting the fan to 100% and see if that keeps your system stable.

11

My setup is right here: http://pcpartpicker.com/b/Psbv6h

I run a single R9290X DCUIIOC and a 4770K with H100i cooling it. It sounds reeeeaally weird right now. I even replaced the power cables that are feeding the graphics card.

It runs at around 50-55 on desktop, and around 70ish while in game. I'll do all of the steps above, then if nothing comes out of it, I'll replace the card. And if that doesn't solve it, then I'll be sure to replace the PSU.

But thanks a bunch! It really helped!

12

Shouldn't be hitting half the PSU capacity. Temps look perfectly normal for a stock cooler. Hope it's just a piece of flawed hardware and gets sorted smoothly.

13

Alright guys. Tried a lot of things today. Nothing seems to work out. Here's a list of things I tried in order:

Try a different PCI slot > Clean Drivers Install > DX Update > Clean Format PC > Install Everything from 0.

The thing is, between 'Clean Drivers Install' and 'Clean Format PC', I decided to run 3DMark 11. It ran all the tests up to that gliding lady. Both screens went black right after she landed (tried 3-4 times, the same white net flickering happened). I was monitoring the temps and fan speed, and they were at 80 degrees and 35-40 % respectively. So I decided to run the test again without the smaller monitor installed. It again went dark right after she landed, but this time before restarting the PC, I decided to plug the other monitor back in. And bam. Small screen gave me a screen, and I saw that I was getting an error (like before, computer kept running even though the other screen was black; it went back to normal after I plugged the second monitor in, though): "DXGI_ERROR_DEVICE_REMOVED." So this time, I decided to clean format the PC, and reinstall everything from the beginning.

So I did. I installed everything back again, including DX from Microsoft's website, and Battlefield 4. I ran 3DMark 11 again, and it crashed a little bit after that scene. Before, I ran 'Run All', and waited for the other two tests that come before the-gliding-lady-thing to be completed. This time, I ran the gliding lady only. It crashed at the same spot, but this time, both screens didn't go black, only the big one did. So I'm thinking, other tests ran, then the lady and it crashed. I directly ran the lady test and it crashed again. So I don't think it's heat related(?) This time around, I got another error: DXGI call IDXGISwapChain::SetFullscreenState failed. I did my research, and saw that it is DX related, some people suggested to run DX setup from Battlefield 4's folder, so I did. And ran the test again. It crashed at the same spot, but this time, I got both screens black, and my speaker going crazy and making a screaming-like sound. I shut the computer down and came to work. :P

More info: The smaller monitor is Dell E2011H running at 60Hz, and the large one is Asus VG248QE running at 144Hz.

I have one more idea as to what's going on. I feel like graphics card can't handle two different sized monitors running at different frequencies? If so, how come it worked all morning (yesterday) and started crashing afterwards? Also, the second error I got is apparently related to Directx 11, but I didn't mess around with it before it started crashing? And isn't it strange that it only fails at one test at a very specific spot? Might be a coincidence but still, I really don't wanna believe that it's the hardware.

Thanks guys!

14

Also, Dell monitor is connected through HDMI to DVI cable, and Asus is DVI.

15

I have the same thing with my dying r9 290 :(

Have yet to figure out what it is  , so tell me if you see a solution !

16

Take out the GPU and use integrated graphics.  Play around with it and see if it crashes.  It sounds like your GPU is starting to die.

17

Will do! Thanks!

Could it be related to failed BIOS update? I'll flash it tomorrow if it is. Basically what happened was when I was dealing with the previous problem, the random shuts off because of the disconnected 8 Pin CPU cable, EZ utility detected a BIOS update, and around 80% in it, the computer restarted. It runs on desktop and all, but on games and 3DMark 11, it fails. Do you think that might be the issue?

Also, Asus will send me an email confirming my RMA request about tomorrow. I'll give it one last go, and if it doesn't work, I'll send my card in!

Thanks for the answers!

18

Um, a failed BIOS update could cause no end of weird problems and I imagine you must be lucky that it runs at all.

Crashing on specific spots instead of generally after some time sounds suspicious, but there's still the chance of a faulty GPU.

I think you could try some things about checking if the temperature is an issue - open the case up, point a fan towards it... Could even try a hair blower with the coil turned off. Increased airflow might help. Could cool down the side of the GPU that doesn't have the fans.

19

Update: Flashed the BIOS. No luck. I, again, tested all of the components individually (after a clean format), and when I tested the graphics card, it failed, and all of the monitor shutting down happened.

Contacted Asus on last tuesday, still waiting for them to send me the confirmation e-mail. They are a PAIN. Urgh. Long story short, it's already been 5 days and they, according to them, are still checking their stocks for availability. Jesus.

20

Are you able to manually dial the fan speed up on the GPU?