1 GPU dead, brand new one already not POST'ing!

aanders50

Honorable
Dec 20, 2013
25
0
10,530
*EDIT AT BOTTOM WITH NEW INFORMATION - 1/23/18*

Hello! I am having a seriously annoying, time consuming, and expensive problem with my GPU(s). I am a gamer, and have been playing with my current rig for about 4 years now (will post specs below). I run a 2560x1440 monitor and in November I got 2 side monitors (1920x1080). So I currently have a triple monitor setup. Anyways, mid-december I am playing a game and suddenly all my displays goes dead. After spending about 3 hours of panicking, re-seating stuff, trying different PCI-e slots nothing is happening to my monitors at all. I sent it in for an RMA and they also concluded that it was dead (no POST). This was for a 3-4 year old R9 290x-edfd.

Anyways, I've been on the integrated card till then and I FINALLY got my new RMA replacement card today (an R9 390x-8256). I was super happy and excited! So, I hooked up all my monitors and did some updates and ran out for dinner (leaving my displays up). I come back 2 hours later and everything is just as it should be. So I boot up a game and after about 5 minutes, BLACK. Everything went dark like it did in December and I thought I was literally going to throw up / cry on my office floor. Luckily, I was able to force power it down and turn it back and it worked! I thought okay...weird, lets just go back to play super low graphic games. 5 minutes into game, BLACK. All monitors go dead again. (Keep in mind, PC is still on, there is just zero signal response on my monitors, like no POST).

This time however, the reboot did nothing to the displays. They stayed completely dark without any POST while the PC ran. I tried a reboot again. Nothing. I tried re-seating the card. Nothing. I tried the second PCI-E slot, nothing. I walked away, started tearing up a bit and now I am back about 15 minutes and IT BOOTED UP FINE!

I am writing this message as fast as I humanly can to see if I can get help on what the heck is going on. I am not sure what killed my first card and what is about to kill this second card and drive me into spiral of sadness/depression with the another RMA and useless computer again.

What should I be watching out or checking for? The only new change was the two side monitors, but I feel like a 290x and even a 390x should be fine with that. I don't do a massive resolution on games using them, they are just there for side windows and such while I play on the main screen.


My current setup is.
3 monitors
2560x1440p (HDMI to HDMI)
1920x1080 (DVI to HDMI)
1920x1080 (Display Port to HDMI)
CPU: INTEL|CORE I7 4790K 4.0G 8M R
RAM: G.SKILL Ripjaws X Series 16GB (2 x 8GB) 240-Pin DD
Motherboard: GIGABYTE|GA-Z97X-SLI Z97 ATX R
PSU: EVGA|850W 110-B2-0850-V1 R
HDD: 1 HDD
SDD: 2 SSD's (OS sits on one of them)
GPU: R9 390x-8256 (Driver: Adrenalin 18.1.1)

OLD GPU THAT DIED: R9 290x-edfd (also latest drivers)

MSINFO32: https://nofile.io/f/dJ0JbCAD5JB/ComputerInfo.txt

I am really scared to try or test anything out of fear of killing this new GPU. :'( I will be watching this thread like a hawk and am willing to chat on Discord or anything as my life is basically at a standstill until I can fix my computer.

Thank you so so so much in advance!
-Austin

SIDE NOTE: I should mention, that when my 290x was in for RMA, I was able to use the PC with the integrated graphics card on the motherboard. I could play some super low intensity games, and while they would feeze or crash now and then, they never killed the display to where everything went back. Just a typical integrated graphics card crash. I could hop right back and usually continue playing for another or two before a small crash. But again, no black screen or anything on the single monitor I was using.


EDIT (1/23/18)
Some new information, I have been running MSI Afterburner ever since I received my new card on the 18th. I have noticed most of my low end games run the GPU temp at high 60 or low 70, which seems okay, and the CPU is about low 70's.

Now, I tried Civ 6 out which I think is a little more intense, the GPU was about 77-79 C (which I "think" is okay for R9 390x) but the CPU was in mid 80's, with AI turns (cpu crunching) going to mid 90's till it was my turn again. After an hour of playing Civ all my monitors went dark randomly, I also noticed I lost connection to my voice chat with friends, so I assume the GPU didn't die but maybe other components also turned off/crashed?. Although the computer itself appeared to still be running (lights on, fans going), just nothing was happening visually or audibly, could this be the CPU overheating?? If so, could a CPU overheat kill a graphics card like my previous one and if it was the CPU overheating, should my computer fully shut down or no?
 
Solution
From the pictures, that PSU puts out 70A on the 12V line, which should be plenty for the GPU.

I would try to test out that system with another PSU from someone else which is well rated and at least 650W for that GPU. Sounds like the PSU may be dying.

homeybaloney

Prominent
Jan 18, 2018
7
0
510
https://www.youtube.com/watch?v=cXvd8lyw8Lc

maybe try taking out your ram, taking out your bios battery, removing all power from the pc, plug pc back in but keep the battery out. power down pc, put bios battery back in and 1 stick of ram in slot 1 or 2. play games and see if this fixed it if so then somethings wrong with one of your ram slots or one of your ram sticks.

Also use cam web monitor to monitor your gpu temperature if its too high it will crash;
https://camwebapp.com/
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


Any particular reason why RAM would cause the GPU to die? The PC runs, but no signal inputs are detected...if RAM was messed up wouldn't the PC being having additional problems or not work with the integrated graphics (which do work?)
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


I shall give that a shot. Any guess as to why this may work? A 390x should be far more than capable of supporting triple displays unless somehow the monitors can cause issues to the GPU itself?
 

homeybaloney

Prominent
Jan 18, 2018
7
0
510
The gpu temperature isn't showing in your picture. try open hardware monitor and play a game with your 3 monitors and look at the gpu usage; http://openhardwaremonitor.org/

leave the sw open when you play a game to let it record the gpu core load usage. 100 percent means your maxing out you gpu.
 

D JAY Cva

Reputable
Feb 12, 2015
152
0
4,760


since you mentioned the issue started after the installation of additional monitors , there is a possibility that they could cause this .. if i were you , i would remove those monitors and test with the primary monitor only to be sure .. if it worked fine the second step will be to hook up one of the additional monitor (without converter if possible) and test that too for few hours ,,the troubleshoot might take longer time .. if you keep having the same issue with single monitor connected .. the issue is due to some other reason ...

 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


Yup! Although I have always wondered if I hooked it up correctly? But it worked for the 290x for 4 years so it seems okay? The PSU has a giant array of power lines that go into stuff, but it also has named ports on it as well. Like Perifs, SATA, VGA, etc. So I just plugged stuff in that sounded correct with the motherboard and used the VGA ones for the GPU itself and it has all worked fine for 4 years since then.

My exact PSU is this guy, if that helps understand my explanation: https://images.evga.com/products/gallery/png/110-B2-0850-V1_XL_4.png

When I got RMA'ed this new R9 390x, they did send me some cables, but they just look like adapters, here is an image of them: https://imgur.com/a/yUpZX
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


In that image I posted previously, the GPU temp is the top left graph, isn't it?
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


I do not have a spare one on me, but could probably arrange to test with a friends over the next or so.
In the meantime, is there any safe way to test that it is faulty?

So far, the computer has been fine since last night, but I haven't touched any games. I did screen capture all of MSI afterburners tests while trying the 3DMark Demo on Steam though to see if it crashed it, which it did not. I think the only crazy things in the graph were that the Pagefile usage was super high, my CPU's peaked in the high 90's, and I think my GPU got up to the 80's, I would have to go back and look at the videos though. (All in Celsius).

I would love try a game and watch the graphs, but I am too scared to risk killing the card. :/
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530
UPDATE (1/23/18)
Some new information, I have been running MSI Afterburner ever since I received my new card on the 18th. I have noticed most of my low end games run the GPU temp at high 60 or low 70, which seems okay, and the CPU is about low 70's.

Now, I tried Civ 6 out which I think is a little more intense, the GPU was about 77-79 C (which I "think" is okay for R9 390x) but the CPU was in mid 80's, with AI turns (cpu crunching) going to mid 90's till it was my turn again. After an hour of playing Civ all my monitors went dark randomly, I also noticed I lost connection to my voice chat with friends, so I assume the GPU didn't die but maybe other components also turned off/crashed?. Although the computer itself appeared to still be running (lights on, fans going), just nothing was happening visually or audibly, could this be the CPU overheating?? If so, could a CPU overheat kill a graphics card like my previous one and if it was the CPU overheating, should my computer fully shut down or no?
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530
UPDATE 2 (1/26/18).
Just put in two more case fans + a corsair 75 for the CPU. Issue just occurred again while in a game. Temperature logs show the temp being about 15 degrees than usually for this game, so I do not think it is the CPU overheating.

Could this be a like a PSU issues?
 


What PSU do you have? You did not answer that.

Till now, it is sounding like a PSU issue. How much does it output through the 12V?
 

aanders50

Honorable
Dec 20, 2013
25
0
10,530


My PSU as posted initially is a "850W 110-B2-0850-V1 R" which I believe is this: https://www.evga.com/products/product.aspx?pn=110-b2-0850-v1

How do I look at, see, or test the "output through the 12V?"

Thank you!