version 1.0.3 error; IDXGISwapChain::SetFullscreenState failed

  • Problem
  • Updated 2 years ago
  • Acknowledged
I get this error with version 1.0.3:
Workload Single init returned error message: DXGI call IDXGISwapChain::SetFullscreenState failed:

The requested functionality is not supported by the device or the driver.

DXGI_ERROR_NOT_CURRENTLY_AVAILABLE
I am running the exact same system as I was before I upgraded to the new 3dmark version. I was running 1.00.02 and everything was fine. my nvidia driver is 285.62.

Easy-E

  • 4 Posts
  • 0 Likes
  • confused

Posted 2 years ago

  • 1

FM_Jarnis, Official Rep

  • 6250 Posts
  • 400 Likes
This error can happen if you click somewhere on screen after starting the benchmark run but before the first test starts rendering.

It can also happen if some background application is interfering with fullscreen 3D rendering

Third thing to check would be to see if switching between stretched and centered makes any difference on your system.

Easy-E

  • 4 Posts
  • 0 Likes
my system:

Asus x79 Sabertooth
i7-3960x
16GB Corsair Vengeance 1600MHz
2x 128GB Crucial M4
2x EVGA GTX580 SLI

Easy-E

  • 4 Posts
  • 0 Likes
This error happens immediately following the first underwater test. I am not running any other applications and I'm not moving the mouse or doing anything else during the test. I tried both centered and stretched, yet it does exactly the same thing either way. I'm puzzled because the previous version worked perfectly but nothing physically has changed. The ONLY thing that has changed is my monitor, I went from a 1080 to a 1200. Its weird because the entire first test completes, and then it gives the fullscreen error. Any insight would be great, otherwise, no big loss. I got the 3dmark11 for free with my evga 580's so it was just a bonus. Thanks for your time.

FM_Jarnis, Official Rep

  • 6250 Posts
  • 400 Likes
Just to rule out some things, since you appear to have Advanced Edition, could you try running the benchmark in Windowed Mode and see what that does (yes, that makes it a custom run so it isn't a solution, just interested in narrowing down potential causes)

Also what exact driver version you are using?

Easy-E

  • 4 Posts
  • 0 Likes
I tried it this morning, and in "windowed-mode" it worked just fine. My driver for the gpu's is the latest whql nVidia driver 285.62

As I was writing this, the whole window mode thing made me think. I have a high-end HP monitor, and part of the driver package is the ability to rotate and partition the screen. So, I exited the application and 3dMark11 works fine. I don't know why it allowed the first test and it interfered with the second, but it doesn't matter. Problem solved. Thanks for the help and sorry for the trouble.

FM_Jarnis, Official Rep

  • 6250 Posts
  • 400 Likes
Glad you got it solved. I believe the issue here is that the application that does that auto-rotate is not fully working with fullscreen DX11 applications and when, between tests, the screen switches back to desktop for a second (before loading screen comes up), the software does something that causes 3DMark 11 to lose focus and that pops up the error when test 2 starts.

(Reason for the error is that 3DMark 11 must not lose focus during testing or the benchmark result would not be valid - it is part "anti cheat", part ensuring that the score is not influenced by outside factors. Windowed mode is a custom benchmark mode so there it isn't a factor and the test doesn't get tripped up by the strict validation)

...and that's why any background applications that might interfere with screen rendering (like the on you had running) should be closed for the duration of the benchmark.