search

SunnySuite / Sunny.jl

Spin dynamics and generalization to SU(N) coherent states
Other
86 stars 19 forks source link

Sf ff revisions #33

Closed ddahlbom closed 2 years ago

ddahlbom commented 2 years ago

Main changes:

  1. Structure factor interface changed slightly. For dynamic_structure_factor, user no longer sets meas_rate. Instead, they may set a omega_max, which will determine the down sampling. Number of frequencies resolved is specified with num_omegas (not dyn_meas). dynΔt is now just dt.
  2. Form factor calculations have been revised. FF information is specified with SiteInfo. The corrections are precomputed prior to structure factor calculations and applied after calculating magnetization.
  3. Some simple tests have been added for the structure factor material, but they are very preliminary: they don't test for correctness of results, just for errors. I have them commented out for now, but I run them before pushing. They would have caught the errors I introduced after doing the indexing refactor.
  4. sf_slice, a function for cutting paths through a dynamical structure factor, now works on structure factors for which reuduce_basis and dipole_factor are false.

In general, this portion of the code is a little bit difficult to modify and expand because the structure factor data has a different number of indices depending on the stage of the calculation and the value of various optional parameters. At some point it would good to find some set of abstractions to make this a bit easier. Have various thoughts. To be continued.

kbarros commented 2 years ago

These changes look great to me. I cannot review in detail but I suspect any changes needed can be made as follow up PRs.

ddahlbom commented 2 years ago

Ok, great. Thanks. I'll merge.