Thank you computezrmle. I will look into it...

Curiously - as if further evidence of strangeness were needed, one of the ~200% Theory tasks has just completed. Here's the listing from my LHC@home Tasks web page

Task Work unit Computer Sent Time reported Status Run time(sec) CPU time(sec) Credit Application430147198 237250672 10860321 30 Oct 2025, 0:01:57 UTC 30 Oct 2025, 15:34:55 UTC Completed and validated 55,978.00 77,546.53 845.81 Theory Simulation v301.00 (vbox64_theory) x86_64-pc-linux-gnuIt may be useful to observe that other Theory tasks had Run time and CPU time about the same.

The new Theory task that replaced this behaved properly and ran on just one core - initially. Now it's on 2 cores.
And now the ATLAS is on 2 cores only - as are both the Theories... That's the 75% of my CPU I'm aiming for in the settings. Fun. Well I've got a lead now. That VirtualBox is a hoot.
Thanks again!

Yes, this task got stuck and should be cancelled.
Unlike other Theory tasks Sherpas are known to have a higher risk for this.

.... Instead, leave it at 100 % and limit the # CPUs via <ncpus>N</ncpus> in cc_config.xml..... or in BoincTasks -> Menu Extra -> BOINC Preference -> On multiprocessor systems, use at most ...... % of the processors

Hello,

I downloaded the script theory-vbox.sh from link https://github.com/seanr22a/LHC-Theory/blob/main/theory-vbox.sh
Then I tried to run it with the following Terminal command:

sh theory-vbox.sh
And I got the following result:

-e

--- LHC Theory - USER ---- 2025-09-24 16:27:48 ------------------------------------------------------

| Events |
Slot Job id Total Processed Remaining Elapsed time Completed % Err
------------------------------------------------------------------------------------------------------
theory-vbox.sh: 101: Syntax error: word unexpected (expecting ")")
Some kind advice would be very appreciated

... What is it about this project's crunching that produces such wildly varying runtimes?...
It's a result of the scientific input which you get (by random) from the backend queues.
More details can be found here:
http://mcplots-dev.cern.ch/
currently in progress:
http://mcplots-dev.cern.ch/production.php?view=revision&rev=2922

Crystal Pellet, many thanks for the valuable information :-)

The issue resolved itself with new tasks.

So the issue must have been bad tasks.Your issue and of other volunteers is solved by implementing Theory v301.00 application.

This version fixes the upload issue.

Is this being addressed?
Yes, but not yet solved.