Despite me leaving my machine on 24/7 I very rarely seem to get work units. Is the demand for units really that high or am I missing some trick? Is my machine low down the pecking order perhaps?

I've also not been able to download more than one unit for a very long time now. I assume the two are related.

Hi, I'm seeing the same. Have you figured out why is this happening?

Hi, I'm seeing the same. Have you figured out why is this happening?

this project only has Nvidia GPU work available, and you either do not have an Nvidia GPU installed or you do not have the drivers properly installed for BOINC detection of it.

you need a relatively modern nvidia GPU with driver 418.39 or newer for Windows.

---

Do not install BOINC manager in "protected mode" aka system service (if you do, it won't see the GPU in your system).
Set a low work cache (less than 5 days total, preferably less than 2 days total).

Despite me leaving my machine on 24/7 I very rarely seem to get work units. Is the demand for units really that high or am I missing some trick? Is my machine low down the pecking order perhaps?

I've also not been able to download more than one unit for a very long time now. I assume the two are related.

Could you post any abnormal messages from the server that appear in the event log when you update the project manually? That would help much in diagnosing the problem.
Thanks.

One thing to check-
If you didn't quit and rejoin this project when it migrated to a secure URL, you will not get any tasks (just in case you missed that somehow). It's something to try again, even if you did it at that time.

Another thought-
I found that I had to give GPUGRID much more priority than the other GPU projects I juggled or they would push themselves to the front and flood the que.

looks like the OP resolved his issues some time ago. when he first posted, his host wasn't reporting any GPU installed, which is likely why he wasn't getting any tasks.

he fixed his driver or hardware config, and now has his GPU recognized and is getting tasks.

---

Looking at Ben's hosts I see that his windows host is operating normally so I figure I gave useless advice on client configuration. His Linux host has only one task, run on Dec 22nd, so that appears to be the troublesome host. Looks like identical hardware to the W10 host.
Linux host issues are beyond my knowledge for now.

By the end of the week it will be difficult for all of us (that don't run scripts) to catch any tasks as we are currently crunching model run 48 of 50. Now is a less than ideal time to be troubleshooting task deliveries.

I see I forgot to check the original posting date. I'm appropriately embarrassed 😳

Hi, I'm not receiving any tasks either. Are there no tasks to send out?

Are there no tasks to send out?

No.
http://www.gpugrid.net/server_status.php

In the work column, I'm seeing only 25 tasks in progress and everything else is 0.

10/01/2021 11:32:31 | | CUDA: NVIDIA GPU 0: GeForce GTX 1050 Ti (driver version 461.09, CUDA version 11.2, compute capability 6.1, 4096MB, 3374MB available, 2235 GFLOPS peak)
10/01/2021 11:32:31 | | OpenCL: NVIDIA GPU 0: GeForce GTX 1050 Ti (driver version 461.09, device version OpenCL 1.2 CUDA, 4096MB, 3374MB available, 2235 GFLOPS peak)
10/01/2021 11:32:47 | GPUGRID | Sending scheduler request: To fetch work.
10/01/2021 11:32:47 | GPUGRID | Requesting new tasks for NVIDIA GPU
10/01/2021 11:33:16 | GPUGRID | Scheduler request completed: got 0 new tasks
10/01/2021 11:33:16 | GPUGRID | No tasks sent
10/01/2021 11:33:16 | GPUGRID | Project has no tasks available

That’s right. 25 (now 23) tasks that’s are out in the field with other computers waiting to be returned, and none to send to anyone else asking for more work.

You need to wait for the project admins to add more work, or get lucky and get a resend of one of those 23 tasks if one of them has a computation error or times out with the host that has it now.

---

I see, thanks for clearing it up for me!