How comes that everyone does the same research? Shouldnt the tasks be distributed? Right now the research is done over and over again, isn\'t it?

Just asking

---

The tasks are distributed. In the BOINC/CPDN jargon a work unit is the fundamental and unique unit of work - if each \'result\' from a work unit were run on an identical computer then the work unit would produce a number of duplicate results. In practice, results from the same work unit run on different computers (e.g. different operating system) produce slightly different results, which are then combined to produce a composite result for the work unit. Of course, many/most results are never completed.

If the CPDN work units were shorter then as each result fails or expires another result would be issued until at least one result is completed for each work unit. However, CPDN work units are too long for that kind of subtlety - so the consequence for us is some duplication among popular platforms (i.e. Windows/Intel) within work units.

Work units themselves should not be duplicated (but sometimes are).