Hoping someone can give a little help with the algorithm that BOINc is using for the RAC badges (which I dislike anyway!)

This is what i posted over at the Convector pages, but have not had a response.

Can you give more explanation of the various levels of badges? I see 1%, 2%, 7%,25% and a broken image for 40% ?

also, I'm trying to calculate the badges myself with my database over at Free-DC, but I don;t seem to be getting quite the same results.

for the top 1% my calculation shows the top 6 should be getting the 1% badge, this is the way I'm calculating it, might just be difference between real time here and using the xml data in my database.

Top 1% is ((#users with RAC >1) * 1/100) + 0.5

This comes out to 601 * 1/100 + 0.5 = 6.51 so any user with an RAC rank higher than this should have the 1% badge which is the top 6.

userid 1807 is 7th in RAC rank but appears to have the 1% badge.

I kind of guessed at the calculation so it may well be incorrect, but it seems to work with the data over at NFS@Home

I looked at the source code a little for BOINC to help and it does seem to only use users with a RAC > 1.

Thanks

Bok