Hello Movieman, and thank you for telling us who you are: it is always simpler to know whom we are talking to.

It was predictable that those monster WUs would have strong effects on the stats for members who would get them, and I warned one teammate two days ago that he should not worry too much if daily results are rather hectic during this period.
Regarding daily results it is even worse for smaller machines than for top end ones: when a single core machine spends more than two days on a single WU there are obviously one or two days where the daily results are null, and not only 25% below. If in addition this machine hits the maximum CPU problem mentioned in several other threads the problem is even worse, at least until the techs are able to compensate the loss eventually. They are also lengthening the time WUs can be waiting for their partner to get validated, and that does not only affect machines hit by one of those monster WUs.

Regarding specific effects on WUs after validation it is business as usual, with a magnifying effect because of the size. If you are paired with a low claimer it is bad luck, if you are paired with a high claimer it is the jackpot, and we both know that there are low and high claimers at all levels of the performance scale.

What is important here is that those WUs are a particular event, not a regular new feature, and things will come back to normal when they will have been processed..

And finally the most negative consequence is those WUs which hit the Max CPU time limit, because that work was lost and will have to be resubmitted.

At least there are two clues that show that this was neither foreseen nor intentional: all this is happening just when the techs decided to start drying FAAH for switching this project to single-redundancy, and at the same time the only team challenge ever announced by WCG was starting, and it is a challenge on results!

Cheers. Jean.

Hello Jean.
Thanks for the response.
Yes, I didn't see it as intentional just as a concern.
I do both the aids and cancer projects so it was easy for me to move over until this was finished.
Movieman

== Gain after the pain - how high can you fly? ==
MM/globalit and all: Our results don't get points credited until they have been validated. Lately, the other copy of many of the FA@H WUs sitting in the "Pending Validation" list in your WCG "My Grid" >> "Results Status", and needed for your validation, has been sitting in someone else's BOINC "Ready to start" list for a much longer time than usual while they nibble slowly through their monster WUs. When and while these WUs are finally being returned, your daily points will shoot up and restore your long-term average.
2 days ago, I had 45 WUs in PV, normally it's about 15, and now it's 27. The chickens are coming home. Yesterday's points were 37121 (global ranking 268th ), current LT average about 22000.

So the techs or programmers or someone made a boo-boo and these extra-long WUs got out ... I hope the results are still usable, but even if not, lessons will be learnt. I've written, run and debugged enough number-crunching software to know that it's not easy. Mind-boggling amounts of logic, and you can get even the simple bits wrong. A myriad of parameters to juggle. The dreaded "floating divide by zero" and its friends always threaten. Floating-point numbers are very ornery critters. I got it wrong far more often than these guys.

Us mere crunchers are partners in this research effort. We may not get so far, but when you've done a certain amount of research, an educational institution may declare you a Doctor of Philosophy (PhD) even though you've never studied "philosophy". Why? Well, when something goes wrong in your research, which happens 99% of the time, you learn to be "philosophical" ...

KEEP CRUNCHING