"Alle Meije " <firstname.lastname@example.org> wrote in message <email@example.com>... > > Brilliant. So, what would you (in general, or more specifically for this problem), recommend in terms of speed: do as you've done, combine as many dimensions as possible (Hxy and Hz), or to separate out every dimension (have separate Hx, Hy, and Hz)? ==================
I haven't combined any dimensions. What I've implemented does separate convolutions along x,y, and z. That is the recommendable thing to do.