Euphonic: Powders

[rebeccafair]

Easy and efficient simulation of powders in Euphonic in addition to single crystals is important for many users. Required tasks:

• Euphonic: Powders technical discussion #187

Priorities from above discussion:

• Euphonic: Powders - separate library #195
• Improve efficiency of powder averaging Euphonic#214

We may get powder averaging for free if the following is completed:

• Generic projections in Horace Support general coordinate transformations and generic projections Horace#34
• Trial generic projections with Euphonic #197

Worth investigating:

• Abins: Refactoring - caching, data flow #165
• Euphonic: investigate Sobol sampling for powder-averaging #164
• Investigate powder sub-shell sampling Euphonic#145

Old/outdated tasks:

• Old overall powders plan Euphonic: Powder averaging project #25
• Euphonic: implement and test effect of radial jitter in powder-averaging #163, duplicate of Euphonic 145 above
• Explore methods for powder-averaging with arbitary q-points #115
• Euphonic: Volumetric random-sampled powder-averaging #88
• Efficient spherical averaging by use of symmetry #86

[ajjackson]

pace-neutrons/Euphonic#145 is arguably a subset of #163 : the former is random jitter in the r-axis while the latter covers more systematic approaches.

[ajjackson]

Sobol sequences is still worth a look actually: Scipy offers a few quasi-random sequences and it wouldn't be much work to add them to the menu. https://docs.scipy.org/doc/scipy/reference/stats.qmc.html