980ti Artifacting and Black Screening

Teranus

Reputable
Jun 12, 2014
93
0
4,640
About 3 weeks ago I bought a second hand EVGA 980ti SC on Ebay, the previous seller had it tested by Scan.co.uk and it came back with zero faults.

This week I installed GTA V and experienced serious instability, with the application often crashing or the entire system black screening (sound could still be heard). Now my system seems to be unstable whenever it is under load. Tonight, after closing Hearts of Iron IV, my system started artifacting. Temperature readings in HW monitor show a max of 69C. GPU chip voltages maxed at around 1.10V. I tried running the card at 100% fans through Afterburner but the artifacting still occurred.

I have noticed that most of the time, my cards fans are at 0RPM, never exceeding 800RPM.
I'm running driver version 388.0.


PC Specs:
AMD Ryzen 1600 @ 3.20Ghz (Stock)
MSI B350 Tomahawk Arctic
16GB (2*8Gb) Corsair LPX 3000Mhz (Running 2933Mhz)
Kingston UV400 120GB SSD
WD 1TB HDD
EVGA 980Ti SC
Aerocool (Cannot Confirm) 80Plus Bronze Non-Modular 650Watt PSU

Any help would be greatly appreciated. :/
 
D

Deleted member 217926

Guest
Power supply is trash. I don't think there are any decent Aerocool models. That's unlikely to be causing artifacting but everything else points to the power supply so it's possible it could be causing that as well.
 

Sam Hain

Honorable
Apr 21, 2013
366
0
10,960


1. Ensure previous GPU drivers/software are deleted/wiped ESPECIALLY if they were not NV ones
2. Recheck GPU installation (seating into slot) and recheck that the PSU cables to GPU are correctly installed
3. Perform a CLEAN install of current NV Drivers minus GeForce Experience (GFE); Install GFE separately if used
4. Ensure your motherboard BIOS is up-to-date. I would assume the 980 Ti would pose no issue here but worth a look-see
5. Check that RAM sticks were not disturbed during install of GPU (or anything else for that matter)
6. Try running your own bench of the card, i.e. 3DMark programs and see what results/feedback you get

Do you or someone you know have a spare PSU that can be temp-installed if the above steps fail to eliminate yours as a suspect?

If the BLACK SCREEN is still present, I'd be looking at a new more "vibrant" PSU: i.e. Superflower, EVGA, Seasonic 650w

http://www.tomshardware.com/reviews/best-psus,4229.html

Never skimp on a PSU, as it's the life-blood of your rig.
 

Teranus

Reputable
Jun 12, 2014
93
0
4,640


Will run some benches now and get back to you, if it does end up being a PSU issue, are there any models you could recommend for me?
 
Two things I'd try:

1) Download and run HWiNFO64 and look at the +12v, +5v, and +3.3v readings between Minimum/Maximum/Average while under load. If they are all within +/- 5% variance between each other, chances are it's not your power supply even though it's poor quality (for example if the +12v reading doesn't dip below 11.4v or exceed 12.6v under load).

(Screenshot example here, look towards the bottom readings to find them: http://static.filehorse.com/screenshots/benchmarking/hwinfo-screenshot-03.png). Your chipset may not have voltage sensors to support those readings, so it may not work.

2) Underclock and undervolt your 980Ti and turn down some game settings and run it under load and see if it removes the artifacting. It's possible whoever owned it abused it in overclocking. Nobody knows how Scan.co.uk tested the GPU and determined it "passed."
 

Teranus

Reputable
Jun 12, 2014
93
0
4,640
OK, just ran Unigene Superposition and got 22-45fps, score of around 3300 and a max temp of 79C but as soon as I opened snipping tool to screenshot the results the system crashed to a black screen.
 

Teranus

Reputable
Jun 12, 2014
93
0
4,640


1) All of voltage rails were inside a +/-5% margin from the average and the ideal voltage (5v, 12v, 3.3v).
2) Artifacting hasn't happened again since the first time. However, system is freezing and/or black screening when I try and interact with the pc AFTER a game or becnhmark has exited.
 


Just out of curiosity, have you tried moving the GPU to the second PCIe x16 slot? That's a brand new chipset you have, and it's highly unlikely anything is wrong with it, but at this point I'd be trying anything. Can you borrow someone's Nvidia GPU to test and see if it still behaves the same?

This is just very unusual behavior. If it was a PSU, GPU, or even memory problem, it mostly likely would be crashing in game, not post-game. It almost sounds like you have a software conflict somewhere. Like someone else said, I'm assuming if you had a previous GPU in there, especially an AMD one, you completely wiped out all drivers using an aftermarket utility. Removing them through Windows de-install is not enough.

 

Teranus

Reputable
Jun 12, 2014
93
0
4,640


Using AMD's Clean Uninstall Utility to see if it can catch any residual AMD voodoo nasty still in the system. Will do some more extensive testing tomorrow.
 

Teranus

Reputable
Jun 12, 2014
93
0
4,640
[UPDATE]

After running the AMD Clean Uninstall Utility it seems that the crashing has stopped, when running Unigene Superposition and 3DMark Timespy. I'll mark the thread as solved for now but if the problem returns I'll open it again.