I used this batch script

cd /d c:\Program Files\BOINC
:loop
boinccmd.exe --project https://www.gpugrid.net update

TIMEOUT /T 600 
goto loop

I have had to add

watch -n 300 ./boinccmd --project https://boinc.bakerlab.org/rosetta/ update

to my Linux boxhttps://boinc.bakerlab.org/rosetta/show_host_detail.php?hostid=6306257 to reliably get tasks.

That just makes it much more likely you'll pick up resends when they become available.

But several top performing Windows system running 256 cores are clearly not getting tasks reliably. The reason I am claiming this is my smaller 128 core system is currently parked at #1. That doesn't seem reasonable given the specifications of the other systems.

Even for systems that run 24/7/365 Resource share & the number of other projects they do will have a massive effect on just how high a RAC a particular system will have for any given project they are doing, along with the size of their cache & work availability- with sporadic work available in small batches and as large as possible cache, that will keep a system crunching for as long as possible when it does have work. However, the large cache will also mean that if the system is out of Rosetta work, then it will fill up on other projects, and won't be able to get any Rosetta work during the short time it is available.
Hence huge core count systems with relatively low RACs for their processing capacity.

---

Grant
Darwin NT

On the main page in the Server Status section, we've got that error again

Notice: Undefined variable: stats in /projects/boinc/rosetta/html/user/index.php on line 81

---

Grant
Darwin NT