I agree with what you explain. It is fully correct about my WU, credit, ....
But still no any answer about all other questions about ecm_ru.
Questions about computing size, access to stderr (when WU is running, when reported, no any timestamp...), credits (compare ecm_ru with other)
Why by Boinc setting to not accept more than 12 hours cache (to prevent bottleneck), we receive tenth of WU who require one day (Boinc)
How longer running time how less credit...
Once again you are right. Like you write (all on the same host)
It used 9.4 cpu hours, your host claimed 330 credits but got 564 (avg credit 60/hour)
OK so far.
Then look this WU (fresh returned)
http://www.rechenkraft.net/yoyo/workuni ... d=44755624 .
1.2hours for 92 credit. (average 76.6 credit /hour)
It is what I repeat from beginning, why to let run very long who block host restart/logoff to have less credit.
I will not speak about one or two. But here about ecm_ru... it is about 20%
Today, I received a lot of ecm_hc_1551949827_12p5_241....
Of course, i will not cancel it. No any problem with it.
Happy to be able to crunch without problem. And if my host need to restart, I let finish WU in reasonable time.
So please, once again, reconsider the ecm_ru WU.