I looked at completed work units using "Your results" on BOINC and I when I click on a work unit ID, I other computers (other than mine) have worked on the same work unit. Sometimes they have identical credit claimed and other times they do not. Does this mean that we are duplicating work --- that several SETI@Home participants are repeating the same effort? If so, is this because there are more participants than required to process the available work?

Thanks,
Steve

The reason this is done is for continuity. The numbers are so vulnerable that they need at least 2 matching (or close to matching) to figure the crunched data is correct. With SETI being open source, there are people that can rewrite the cruncher and send back fake information, if you only have one person crunch. This gives credibility to the data crunched.

Currently, yes if they gave us one each, we'd be out of work. This is being remedied with the new antenna(s) that are starting to pump data for SETI, and being Alpha/Beta tested. When these are in the mainstream, I believe we will have a lot of data for a long time to crunch, because it'll be a lot more sensitive data. Astropulse and a multibeam are the two we know of now.

See Science Status Page and Future Plans about some of the new things coming online.

---

My movie https://vimeo.com/manage/videos/502242

I looked at completed work units using "Your results" on BOINC and I when I click on a work unit ID, I other computers (other than mine) have worked on the same work unit. Sometimes they have identical credit claimed and other times they do not. Does this mean that we are duplicating work --- that several SETI@Home participants are repeating the same effort? If so, is this because there are more participants than required to process the available work?

Thanks,
Steve

Seti initially sends out 4 copies of a WU (Work Unit). There are several reasons for this. The main one is so that the results can be compared and verified to be accurate.

Different CPUs and OS produce slight variations in the results. From a scientific point of view only two very close results are required to validate the work. To speed up the process and rnsure accuracy additional WU are issued.

Also, differing systems do not have the same the benchmark on which credit is based, so a third copy is sent out so that the credit issued is the middle number of credit claimed is granted. This reduces the disparity between systems and secondly discourages or at least reduces the effect of the credit cheating that was rampant in Seti-Classic. The variation in credit claimed has recently been reduced considerably with a newer Boinc manager and the introduction of the enhanced Seti client which beside doing a better, deeper analysis of the WU, now counts actual work done (flops) instead using the benchmark (which has proven itself to be too variable and easily manipulated).

The initial fourth WU is sent out to speed up the credit granting system. Since result are sometimes not returned by due date or not valid for a variety of reasons, it was decided to issue an extra initial copy to reduce the complaints about delays in granting credits when additinal copies need to be sent out to validate the work.

See Credit in the Boinc Wiki for a greater understanding of the Boinc credit system. More links at the bottom of that page.

---

Boinc V7.2.42
Win7 i5 3.33G 4GB, GTX470

Thanks for the rapid explanations -- it all makes a lot of sense now. Before it just seemed like there was waste going on that was not necessary. Thanks again!

Steve