Closed UmrathSt closed 6 years ago
Thanks for the bug report. This is the same issue as #159. It affects calculations with periodic boundary conditions with multithreading enabled. Non-PBC geometries are not affected, and the issue should be eliminated by disabling multithreading, although then things may be slow. I am working on a fix, but unfortunately I am in Japan with almost no free time at the moment so it is going slowly. I hope to address it this weekend.
Ultimately this is really a symptom of the longstanding failure to address #22, which would have caught the problem sooner. We really need simple unit tests with TravisCI automation. Any volunteers to help?
Fixed in 284b202. Please confirm and feel free to post further questions, or close the issue if resolved.
Thank you for the fix. I can confirm that its working.
Dear Homer, I am unable to reproduce the ThinFilm example and also had no luck in reproducing the DiffractionPatterns and PerforatedThinFilm examples. The transmission results don't look how they should. I am using scuff from your last commid: "overhauled spherical-moment calculation and scuff-tmatrix application…".
Best regards Stefan