search

marcuspetschlies / cvc

3 stars 7 forks source link

for twisted boundary conditions, random vector needs to be the innermost dimension #2

Closed kostrzewa closed 5 years ago

kostrzewa commented 6 years ago

In order to save on redoing the MG setup for each change of theta, the random source must be the innermost dimension if multiple sources per gauge are to be used.

marcuspetschlies commented 6 years ago

I'll take it into consideration, when putting together the tbc-related prog.

Am 05.11.18 um 10:29 schrieb Bartosz Kostrzewa:

In order to save on redoing the MG setup for each change of theta, the random source must be the innermost dimension if multiple sources per gauge are to be used.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/marcuspetschlies/cvc/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AMcioZhLmfrkKpvgwZsQUe4V1o2xMZKWks5usAUCgaJpZM4YOD_7.

marcuspetschlies commented 5 years ago

current implementation in cpff_tbc_invert_contract uses minimal number of mg resets; but this depends on the number of fermion fields to be stored;