Xraytrace documentation

The calculation is performed on a rectangular domain that is divided into rectangular voxels. Through these voxels the rays are passing, leaving some intensity information there to be accumulated, summed with the other rays or being somehow interpreted.

The first information that needs to be given is the voxel size. This can be in principle different in different directions, which can be useful especially to treat the different sample scale in lateral directions (xy) and z direction. Even if most of the algorithms are not related to any particular sample orientation, some sources are, so we recommend to have the sample oriented with surface normal pointing the z direction.

The following input file example sets the x,y and z voxel size to 1 nm (note that all the quantities are always in base SI units).

POOL
1e-9 1e-9 1e-9

Note that the voxel size does not need to be smaller than the incident rays wavelength, however special guidance needs to be taken if it is similar or higher (see the section on running the computation).

Then we need to setup the number of voxels in each direction to be used. This is done together with setting up the sample data, so it is a part of medium loading. If we use simple vector objects to generate the medium, we specify the voxel size in this step, like shown here for 300x300x200 voxels computational domain (followed by the vector objects file and materials table):

MEDIUM_VECTOR_SAMPLE
300 300 200 object.vector mat.table

If we provide the medium as a regular structured Ascii VTK file (where the provided numbers are directly the material indices), the medium size is loaded from the VTK file, so there is no need for providing it separately, as shown here:

MEDIUM_SAMPLE
sphere.vtk sphere.table

Note that the computational domain size is used in many places through the software, so it is e.g. size of the output planes or output VTK data.

More details about the sample geometry definition can be found in the sample geometry section.