Reached daily quota of 100 results ?

[none1]

Just checked some of my machines, seems I get this message "Reached daily quota of 100 results" and then can't get any more cuda units, although the CPU seems to have work units for days.

Berkley servers are shut down and I can't see their forums, anyone have a solution for this?

Thanks!!

 

[Sephis]

I'm in the same boat. I don't seem to have a problem getting CPU units (when the servers are up). I've just been using the rescheduler to turn those cpu units to gpu units. Then BOINC just gets more CPU units.

 

[none1]

... using the rescheduler to turn those cpu units to gpu units. Then BOINC just gets more CPU units.

How do you do this?

Thanks!

 

[Sephis]

Its a program that somebody over at lunatics made. I'm trying to find the download link right now. It originally just turned the very low angle cuda units into cpu units, but now it has the ability to transfer any units back and forth. I'm gonna go dig some more to see if I can find the download. If not, I'll zip it up myself and send it your way.

EDIT:
Found the download link. http://lunatics.kwsn.net/12-gpu-crunching/cpu-gpu-rebranding-perl-script.msg19454.html#msg19454 go there, its attached to that forum post.

Close BOINC, unizp the .rar file, run the ReSchedule.exe in the unzipped folder, hit the settings tab and uncheck "Only VLar+VHar to CPU", go back to the main tab, move the slider to whatever percent you want to move to the GPU, and then hit run (you may have to hit test first, I don't remember). That should move whatever percent of the work you selected to cuda work.

 

[col_sanders]

I never knew that existed. Very cool tweak

 

[Sephis]

I never knew that existed. Very cool tweak

Yeah, it comes in handy for the people who don't run the vlar kill that comes with the optimized apps. The vlars can take a very long time on the gpu, but the cpu crunches them just fine. You can set the the rescheduler to run at given intervals as well, to avoid getting any new vlar work sent to the gpu. I just run the vlar kill apps, but to each his own I guess.

 

[QuietIce]

There's a limit of 100 units on CUDA??? I didn't know that. This isn't a new rig is it?

I know the number of units allowed per day starts out small then works it's way up as a rig "proves itself" - at least it used to be that way, maybe not anymore with all the high-powered machines out there now ...

 

[Sephis]

This is the same rig I've been running for over a month now. It just started this recently.

 

[Codeman05]

I ran into this as well, but I haven't been able to get ANY work units (cuda or cpu) all day.
So 1/2 of my active farm is completely idle...doesn't look like I was getting a full cache built up over the last several days

If I had to guess, I'd say the new CUDA WU limit was imposed due to an insane amount of bad data being returned by Fermi cards.
There was several mentions of Fermi WU issues by the staff that have triggered some of the recent problems. I guess this was their fix for the time being.

Hopefully that's temporary, I'll chew through the 100 unit limit in a couple of hours =/

Oh well, I'm sure it'll get straitened out at some point