That was it, Ian. Setting those symmetry boundaries fixed the NaNs.
Thanks again for spending time on this.
Bernard
P.S. I think a binary WF-extraction testsuite would be a good idea. If you like, I’ll apply what you’ve shown me to the qc0_mclachlan.par file, and send in the results ...
(do I have commit privileges for EinsteinAnalysis/Multipole? I suspect I’m doing everything anonymously right now)
On 10/22/10 1:15 PM, "Ian Hinder" <ian.hinder@aei.mpg.de> wrote:
On 22 Oct 2010, at 17:09, Bernard Kelly wrote:
>
> I've tried your "offset" parameter, but that doesn't seem to help,
> I'm afraid (actually, it's on by default anyway). I've
> just enabled Multipole's "out_1d_every" option, and in fact the data
> has nan at the -equator- (that is, at points
> theta = 1.530519 & 1.611073 -- those closest to the x-y plane).
Sounds like the symmetry boundaries are not being applied. Aha! You
need to set the parameters
WeylScal4::Psi4r_group_bound = "flat"
WeylScal4::Psi4i_group_bound = "flat"
Have a look at the test suite parameter files for WeylScal4. This is
not necessary in the current development branch, as this is done
automatically by Kranc now.
It occurs to me that we could do with an example for the use of
WeylScal4 and Multipole, e.g. by using WeylScal4 in one of the
McLachlan QC0 example parameter files.
--
Ian Hinder
ian.hinder@aei.mpg.de