#2398: Multipole: Default (midpoint) integration method incorrectly implemented, yielding wrong results... especially with m=odd modes
| Reporter: | Zach Etienne |
| Status: | new |
| Milestone: | |
| Version: | |
| Type: | bug |
| Priority: | major |
| Component: | EinsteinToolkit thorn |
Multipole works by first interpolating (e.g., psi4) data onto a uniform grid in theta and phi, such that
theta = [0,pi], where theta = 0,dth,2 dth, …, N dth=pi
and
phi = [0, 2 pi], where phi = 0, dph, 2 dph, …, N dph=2 pi
Clearly these are not midpoints. Further the integration for midpoint method proceeds as (https://bitbucket.org/einsteintoolkit/einsteinanalysis/src/26d17d9edfecafd0c516e38fe8343bdc1d11eb5d/Multipole/src/integrate.cc#lines-51):
for (iy = 0; iy <= ny; iy++) for (ix = 0; ix <= nx; ix++) integrand_sum += f[idx(ix,iy)]; return--
Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2398/multipole-default-midpoint-integration