Questions and Answers :
Windows :
New Nvidia 3080 graphic card showing 0% GPU use
Message board moderation
| Author | Message |
|---|---|
|
Send message Joined: 17 Jan 21 Posts: 2 Credit: 7,971,617 RAC: 2 |
I just upgraded my GPU from a 1060 to a new 3080 and I expected the card to be much faster at doing the work but instead the card is running at 0% and the work units are not progressing as a % of work complete. I deleted all the work units I had an redownloaded thinking it could be an issue switching cards with the work unit started but still nothing is running. Has this project been updated to use the new GPUs from Nvidia? If not any idea when this will be ready? |
|
Send message Joined: 17 Jan 21 Posts: 2 Credit: 7,971,617 RAC: 2 |
Okay it seems to be working or at least the % complete is going up after a reboot. But it still shows the GPU working at 0-2% and I'm not even sure what is being used on the GPU. |
|
Send message Joined: 17 Jul 20 Posts: 12 Credit: 7,473,347 RAC: 65 |
try to disable other cpu intensive tasks. |
|
Send message Joined: 9 Jul 20 Posts: 142 Credit: 11,536,204 RAC: 3 |
another idea is to open a third party software such as GPU-Z and look for GPU load (in %), GPU clock (MHz), bus interface load (%), memory load as well as power draw (in W) in the sensors tab. If these numbers are close to the max numbers reported in the tech spec sheets for your GPU card model on sites such as "techpowerup", you should be good. Alternatively look for the GPU specs in Win task manager and select one of the measurements to be "CUDA". A high number indicates a large compute load on the card. Finally, the app is not optimised for the RTX 3000 series as far as I know. To boost the overall load, you could bump up the number of tasks that run concurrently on you card via an app_config.xml file that you put in your project directory/subfolder. You should keep in mind to reserve one CPU core for each GPU task that you run so that you don't overcommit your CPU, causing your GPU to always wait on the CPU. Hope that helps. If not, just report back on the numbers that these tools show. |
|
Send message Joined: 5 Dec 21 Posts: 2 Credit: 82,420 RAC: 0 |
I just upgraded my GPU from a 1060 to a new 3080 and I expected the card to be much faster at doing the work but instead the card is running at 0% and the work units are not progressing as a % of work complete. Have you tried removing the 1060, fully deleting and re-installing the NVIDIA drivers? I have a seemingly similar problem with a new RTX 3070 Ti. Using a 12900K PC with Windows 10 Pro. Latest NVIDIA Game Ready drivers installed (497.09). I've set set the MLC@HOME preference to not use the CPU and to only use the GPU but it continues to use only the CPU! The 3070 Ti is working fine with folding@home (which I stopped when trying MLC). In my case the onboard Intel graphics is still available but not currently in use and I'm not willing to disable it at the moment (if this is even possible it sounds a bit risky). |
|
Send message Joined: 5 Dec 21 Posts: 2 Credit: 82,420 RAC: 0 |
Oops, I did not see all the other replies when I found the original post! Sorry |
|
Send message Joined: 30 Jan 21 Posts: 4 Credit: 1,526,031 RAC: 0 |
It may be also a problem of the support of faster cards or Ampere in general. My GTX 1060 and 1660 both need about 64 minutes to complete an MLC wu which had me wonder already why there wasn't a performance increase but my RTX 3060 (mobile) needs between 68 - 95 minutes, usually starting over from scratch once it reaches a certain percentage. |
|
Send message Joined: 5 Oct 20 Posts: 3 Credit: 1,213,744 RAC: 0 |
Yeah, seems this GPU app is highly inefficient. Shame too as so much more work could be done if the app was optimized to use more GPUs and their power. |
©2022 MLC@Home Team
A project of the Cognition, Robotics, and Learning (CORAL) Lab at the University of Maryland, Baltimore County (UMBC)