http://climateapps2.oucs.ox.ac.uk/cpdnboinc/workunit.php?wuid=6069892

The results page states: \"Too many total results\".
Does that mean, that the crunched data is not taken into account?

Thanks for your answers!

PS: This is my result.

The crunched data will always be used.

The \"Too many total results\" message is related to how the workunit has been set up. It looks like the total (and error?) results has been set too low.

---

"The ultimate test of a moral society is the kind of world that it leaves to its children." - Dietrich Bonhoeffer

I understand that we can disregard these Boinc messages, as they aren\'t suited to CPDN (the deadline is another example). We just ignore the Boinc restrictions.

However, one thing concerns me: The new WUs are sent out 10 times (e.g. this one), even if only 1 finished result is needed (probably because the project knows from experience that 10 must be sent out so that 1 is crunched until the end). But as soon as 2 results are in error (such as here, CPDN stops sending out new WUs.

I\'m afraid I don\'t understand the logic. If 10 crunchers are needed, OK. But if 2 fail before all the WUs are sent out, then only 5 crunchers are needed.

That many is NOT automatically sent out; it\'s just a possibility.
Mostly this multiple sending is done by the researchers, if they see somethng interesting happening, or if that dataset was created especially to test something, and they want to make sure it does have a chance to run to completion on at least one machine.
On it\'s own, the server will only re-issure a dataset if the original issue failed.
And once a couple of models issued from the same dataset have failed, it may be obvious that the set of parameters in question will not create a model that runs for a long time. You\'d need to examine the results for each to know this.

To add a bit more, this thread is about the results of some early research that is about to be published.
There is talk there about the use of identical models.