Asus GTX 980 Strix - Artifacting, Blackout & Crashes in Games and 3DMark

[tzelic]

Hi all,

I was very impressed with the reviews and benchmark results for Asus 980 GTX Strix - as such I have purchased one myself just 4 days ago together with a brand new system (specs at the bottom of this post). Unfortunately I have since spent the last 2 days trying to work out the root cause & resolution (but to no avail) to various issues I am experiencing with this card which are mainly related to artifacting/blackouts/crashes. The issues started happening as soon as I used it straight out from the box, and at the default factory-OC'd settings which I have maintained (no change)

This issue is present in all the games that I have tried both slightly older games such as Crysis 3, and the latest ones such as Dragon Age: Inquisition. It also occurs in 'less intensive' games such as World of Warcraft, as well as benchmark program such as 3DMark. Movie playback on standard video player such as VLC player also has artifacts. (easier to see during end-credits, because the screen is mostly dark and the artifacts are grainy red dots)

Please see below for a few videos recorded of when running 3DMark and Dragon Age Inquisition (all running on the latest Nvidia driver 344.75):

3DMark - Test1 (Artifact, Crash)


3DMark - Test 2 (Artifact-red, Blackout)
https://www.youtube.com/watch?v=Kf9wNjB7vpE

3DMark - Test 3 (Artifact, Crash after 6 secs)
https://www.youtube.com/watch?v=kfZCz8bAHSs

Dragon Age: Inquisition - Test 1 (Artifact-bright coloured spots, Crash after 8 seconds)
https://www.youtube.com/watch?v=FVPuuORsIMM

Dragon Age: Inquisition - Test 2 (Artifact-bright co
https://www.youtube.com/watch?v=D3H9YUE2coE

Other things which I have also tried but to no avail:

1) Driver roll-back/tested all 3 versions of Nvidia Driver but to no avail: 344.75, 344.65, 344.60
-no evidence of driver issue

2) Formatted / clean install OS (Windows 8.1 64bit)
-no evidence of system or OS issue

3) Formatted / clean install OS (Windows 7 64 bit)
-no evidence of system or OS issue

4) Swapped between DisplayPort and Dual-DVI
-didn't think this is related, but tried it anyway though with no evidence of any correlation.

MISC/Error messages: Whenever one of the games or applications ended up crashing to desktop, it often resulted in either 'dxgi_error_device_removed' or dxgi_error_device_hung'

Please see below for one of the error messages captured when 3DMark crashed:

Unexpected error running tests.
Workload work failed with error message: eva::d3d11::rendering::deferred_scene_renderer::re nder(): draw_depth_task for thread 0: File: device_context.cpp
Line: 818
Function: struct ID3D11CommandList *__cdecl eva::d3d11::deferred_device_context::do_finish_com mand_list(bool)

Expression: native()->FinishCommandList( restore_deferred_context_state, &result): DX11 call failed [-2005270522].

Device hung due to badly formed commands.
DXGI_ERROR_DEVICE_HUNG: ID3D11DeviceContext::FinishCommandList:



In the end, I ended up removing the gcard and plugged back in my previous XFX Radeon HD 7970 Ghz - and everything worked fine with no issues. (Games, 3Dmark, movies)

This has so far given me more and more evidence that the issue might actually be with the card itself (hardware) - however I just thought I should post my findings here and seek further counsel from the experts..

Thank you so much in advance for any assistance you can provide..

Machine spec:
CPU: Intel i7-5820K 3.3 Ghz
CPU Cooler: Corsair H80i High Performance Liquid CPU cooling
Motherboard: Gigabyte X99-UD3
Memory: 16 GB G.Skill Ripjaws DDR4 2400
OS: 64bit Windows 8.1 (Activated)
Monitor: Asus 27" PB278Q
SSD/HDDs/Optical drives: Kingston 240GB SSDNow Sata3 SSD,
2TB Seagate Barracuda 7200 RPM

Case: Thermaltake Black Core V71
PSU: Thermaltake Toughpower 750W 80+ Gold
USB Devices: Logitech G15 keyboard
Mad Catz Rat9 wireless mouse
Astro A40 Mixamp & Headset
Netgear A6200 WIFI Adapter

[Chino]

Have you tried reseating your GPU or using it in a different PCIe lane?

[tzelic]

Yup I have tried reseating the GPU as well as on a different PCI-E lane. Moreover, my other gcard which is Radeon 7970 ghz ed did not have any issues with either one of the tested PCI-E lanes.

[sde444]

Its just faulty card, not stable factory OC. RMA itd. I RMA my Asus Strix 980 because was artifacting in Far Cry 4 like yours on 3dmark.

[Chino]

Ok.

Open GPU-Z or the integrated monitor if you're using GPU Tweak, run the 3DMark test again. It doesn't matter if it crashes or artifacts. I want you to look at the temperature and the speed for the fans for me.

[tzelic]

Thanks for the suggestion. As per instruction on the Worldwide RMA guide, I ended up contacting my local retailer as to be mindful of the return timeframe. It turns out they have their own team of techs who tested out this gcard and confirmed it's faulty due to the artifacts.

They have since replaced it with another brand new Asus 980 Gtx Strix, which I have tested now and confirmed to be running perfectly with no artifacts, crashes or blackout.

This is such a fantastic product - however I'd like to thank all of you for all the thoughts and suggestion!

Merry Christmas and Happy New Year