possible things to implement

[RudoRoemer]

minutes of the meeting 09/07/2017:

• list of posible things to do after the code cleanup has been done:

[1] optimize structure for certain atom types only, e.g. Debye-Waller factor for oxygens only. RB wants an input table for each atom and refinement parameter.

[2] automatic creation of difference images wrt input image and along iteration.

[3] conversion test (increase number of strong beams (N) until FR has converged), time will be proportional to N^3, base convergence on difference between exp. images and simulated images. Remember that exp. images are accurate at the level of 8-bit.