I've been trying to get work for my 780 for several hours now, it keeps telling me that no work is available yet my 10 series cards have no problem getting work....

This is a linux machine..... the other machines are windows machines....

I've been trying to get work for my 780 for several hours now, it keeps telling me that no work is available yet my 10 series cards have no problem getting work....

This is a linux machine..... the other machines are windows machines....

The app requires driver version 418.39 or higher. It looks like you are good there, according to what's in the database.

Do you know which version of Cuda you have (I cant tell from the database)? The plan class requires cuda 10.0 or higher.

I too use linux and have had no problem downloading work for Nvidia GPUs.

I've been trying to get work for my 780 for several hours now, it keeps telling me that no work is available yet my 10 series cards have no problem getting work....

This is a linux machine..... the other machines are windows machines....

The app requires driver version 418.39 or higher. It looks like you are good there, according to what's in the database.

Do you know which version of Cuda you have (I cant tell from the database)? The plan class requires cuda 10.0 or higher.

I too use linux and have had no problem downloading work for Nvidia GPUs.

Sorry about responding to an old thread, but I'm seeing the same issue with a GTX 780 under windows 7. I'm running driver 445.87, Opencl 1.2 and cuda 11 and getting no work. I think I read in another post that the card may be "too old" to run this project, though it replaced an older card that did run NF. Seems to me that if the the project specs are met it should run the project. Any ideas on how to get this card to run NF would be appreciated.

Sorry about responding to an old thread, but I'm seeing the same issue with a GTX 780 under windows 7. I'm running driver 445.87, Opencl 1.2 and cuda 11 and getting no work. I think I read in another post that the card may be "too old" to run this project, though it replaced an older card that did run NF. Seems to me that if the the project specs are met it should run the project. Any ideas on how to get this card to run NF would be appreciated.

So the minimum required "compute capability" is 3.5. According to this:
https://en.wikipedia.org/wiki/CUDA#Version_features_and_specifications
The 780M is not supported, but the other versions are. According to the server, you have the regular 780 which should work (assuming I'm looking at the correct host).

Is this the same host that used to have an amd card? If so, I noticed many WU abortions. It's possible you exceeded the daily "error" quota (yes, abortions are flagged as errors); however it looks like that was on June 30th, and the quota should have reset.

You could try enabling the scheduler debug options in the event log and see if that gives any hints. It's possible the server is confused since this machine used to have an amd card (again, assuming I'm looking at the right host), in which case you could try resetting the project.

Sorry about responding to an old thread, but I'm seeing the same issue with a GTX 780 under windows 7. I'm running driver 445.87, Opencl 1.2 and cuda 11 and getting no work. I think I read in another post that the card may be "too old" to run this project, though it replaced an older card that did run NF. Seems to me that if the the project specs are met it should run the project. Any ideas on how to get this card to run NF would be appreciated.

So the minimum required "compute capability" is 3.5. According to this:
https://en.wikipedia.org/wiki/CUDA#Version_features_and_specifications
The 780M is not supported, but the other versions are. According to the server, you have the regular 780 which should work (assuming I'm looking at the correct host).

Is this the same host that used to have an amd card? If so, I noticed many WU abortions. It's possible you exceeded the daily "error" quota (yes, abortions are flagged as errors); however it looks like that was on June 30th, and the quota should have reset.

You could try enabling the scheduler debug options in the event log and see if that gives any hints. It's possible the server is confused since this machine used to have an amd card (again, assuming I'm looking at the right host), in which case you could try resetting the project.

Yep, same host. I aborted prior to switching out the AMD card for the Nvidia. Project has been reset and I even detached and re-attached to the project with no joy. Thanks for checking. I'll just run something else on it and try back from time to time.