#215: Driscoll&Healy integration for Multipole ------------------------------------+--------------------------------------- Reporter: eschnett | Owner: hinder Type: enhancement | Status: review Priority: major | Milestone: Component: EinsteinToolkit thorn | Version: Resolution: | Keywords: Multipole ------------------------------------+---------------------------------------
Comment (by rhaas):
this seems to be buggy. The results differ by what looks like Pi from the results of test_simpson.par which is otherwise an identical parfile. Testing this with a constant value fr[i]=1.0*sin(th) in utils.cc also yields an answer 3.14 times too large when driscoll&Healy is used as compared to Simpson (which gives the correct 4*Pi result).
I don't know how to fix this since I have no idea where the extra Pi creeps in.
Do not apply in this form. :-)