Soooo is there a time frame before I can throw my Radeon cards at this?
It's a bit of a shame to see my hours and hours of CPU crunching now being smashed by NVIDIA GPU work :(

Oh...and you're going to need to rework your badge system with all these credits. 100Mil is nothing now.

The OpenCL stuff is turning out to be more work than I thought, but maybe a few more days.

And yes, the badge system has needed an overhaul for a while now. If only I had more time...

Soooo, should I just give up trying on my 760 GTX? I read there are some issues on the 6xx series of cards, but does that extend to 7xx as well?

on Fedora 29 i got lucky ...
once in a time the BOINC Client and the drivers worked (cuda 10.1 ? no. ).
Restart, all functions are gone.

Nice...not.

You could try upgrading to the 10.1 version of cuda. Other than that, I wouldn't waste your time, this exe may just not work on "older cards". Maybe the OpenCL version will fare better.

I did it,

dnf install cuda

long before, it installed everything BUT it don't generate the needed "lib64" folder and the pathes to the libraries.

lol, I have cuda 10.1 and newer drivers and still no beuno. I guess I will wait for the OpenCL version.

Eric, fixed credit per WU works for me as long as it gives "reasonable" points when compare to CPU in regards to amount of "GFLOPs" done though there could be algorithm efficiency differences in GPU vs CPU which I'm not sure. Faster GPU card will be able to churn out more WUs per hour and vice versa, so they collect more points.

I looked at some asteroid WUs, they sometimes do vary a lot but most of the time less, from what I've seen. Nevertheless, I think on average, the run time will settle to some mean value. At least you are not alone.
.

There is only one GPU in the Top Computers list older than the Maxwell series successfully crunching the GPU WUs. It is a GTX 780. My GTX 1050 Ti has an average run time 20% faster than the GTX 780 and it uses less than half the electricity. Time to consider upgrading those Kepler and Fermi GPUs.

There is only one GPU in the Top Computers list older than the Maxwell series successfully crunching the GPU WUs. It is a GTX 780. My GTX 1050 Ti has an average run time 20% faster than the GTX 780 and it uses less than half the electricity. Time to consider upgrading those Kepler and Fermi GPUs.

Regardless of the cards I am running, or the power you think I use, the card is compute 3.0 capable AND is running 418.56 drivers. It should work.

There is only one GPU in the Top Computers list older than the Maxwell series successfully crunching the GPU WUs. It is a GTX 780. My GTX 1050 Ti has an average run time 20% faster than the GTX 780 and it uses less than half the electricity. Time to consider upgrading those Kepler and Fermi GPUs.

So it works for compute capability 3.5 or higher. I will change the plan_class to reflect this.

Regardless of the cards I am running, or the power you think I use, the card is compute 3.0 capable AND is running 418.56 drivers. It should work.

Yes, it successfully built for compute capability 3.0 and higher, so it "should" work. But unfortunately it does not.

so it "should" work. But unfortunately it does not.

Exactly, as i will have to know it about , maybe 20 years of linux..not working errorless.
By reason, programs are not always they "look like".

Credit is acceptable for now. More important is OpenCL and Windows implementation. Credit can always be adjusted [upwards :)] later... IMO this is where Eric's resources should be applied to enable more users and move the project's scientific goals forward.

Eric, thanks for listening and changing the credit scheme. Credit screwnew was not the way to go.

little good news

as discriped here, i set the folder variables BEFORE installing again:

http://icl.cs.utk.edu/magma/forum/viewtopic.php?f=2&p=6510

Look at the lower end of the page.

sudo dnf remove cuda
sudo dnf install cuda

ok, now the "lib64" folder is filled up with libraries and source-code plus tools
and all everything u need. Cheers !

Out of tasks again?

EDIT: Okay, one of my hosts is getting tasks again, but I can't seem to get anything for this one. Exact same OS, two Pascal cards instead of one Maxwell and one Turing. The driver is a bit newer. What am I missing here? The event log gives the usual "got 0 tasks" without any elaboration.

EDIT2: Nevermind, posted too soon. After half an hour or so I finally received some tasks.

Out of tasks again?

No, there's enough.

I too have noticed the scheduler being stingy with GPU jobs. My client had no GPU jobs this morning and after clicking update, it only gave me 5 new tasks.

This might be related to the "daily result quota" on the server. The day before I had hundreds of jobs in my queue that all failed after an auto update that changed my Nvidia driver. Ever since that, the server does not want to give me tasks. If you aborted a bunch of jobs, you might be in the same boat as me.

I see. Yeah, I have 380 aborted tasks from a few days ago, when I was still playing around with things. Thanks for the response.

Excellent stuff for the GPU app,

Is there a daily quota?
I ask as I can not get any tasks.

Thanks

Any estimate on when we'll have an app that will run under windows?