I see strange runtimes and credits for "..._13start_test_..." WUs - for example on my I7-980X

https://boinc.bakerlab.org/rosetta/result.php?resultid=455268995

https://boinc.bakerlab.org/rosetta/result.php?resultid=455229675

Default cpu-time is ~ 10000 sec (as you can see here) - the "..._13start_test_..." WUs run 25000 sec and pays extrem low credits.

---

Some WUs run longer than they should, because sometimes a single decoy simply takes longer than your preferred time (you must return at least one for a WU to be scientifically helpful). If I remember right, Rosetta gives your credits per decoy, not runtime. So sometimes, you'll get a higher granted credit than claimed and vice versa. It "should" even out eventually.

If I'm wrong, please correct me.

Nice processor btw.

---

As Chilean says, it's luck of the draw with tasks to some extent, because some tasks produce very few decoys whereas others will produce many, and credit is assigned per decoy. The totals do average out very well after running a few tasks though. I have two Q6600s and they consistently have very similar RACs.

HTH
Danny

---

Thanks for the answers.

I had the suspicion there is something wrong with this "_test_" WUs,
but your explanations are obvious.


@Chilean - "Nice processor btw.".

Right, but the cost-performance ratio of my i7-2600k is much nicer...

---