Quote Originally Posted by Werinbert View Post
It looks like there is a rounding error between the WCG run time data (their site) and the Free-DC run time data (looking at the project user stats). It looks like WCG uses 1 year = 356 days where as Free-DC uses 1 year = 365.25 days. Thus the Free-DC data falls behind by about 6 hours run time for each year of run time. My sample set is small, so I could be wrong, but it seems to be so.
I'd noticed that too when I implemented it. I'm using standard php functions to derive it from the numbers they give, so I would say WCG is actually wrong. I guess I'll just try basic arithmetic instead.