Rahil U. (guest) 28 Aug 2018 18:00
Hello,
I have been trying to run DDSCAT for some custom shapes (FROM_FILE), with the shape.dat files generated from a MATLAB code or from the nanoHUB DDA Convert tool using Blender shapes. These are always created with the shape center at (0, 0, 0). The extinction spectra appear correct, and the E-field intensities generally make sense, but I run into issues somewhere between running ddpostprocess and using MayaVi or Paraview to visualize.
Despite the shape being centered, the E-field intensity map is always off-center and the shape outlines are thus cut off at the edges since they lay outside the computational volume (as ddpostprocess tells me when I try to select tracks from one edge of my shape to the other). I tried to visualize the target.out VTR files on the same set of axes in Paraview, and it looks like the entire shape is somehow thrown off-center since it matches up with the E-field plot. In MayaVi, however, I can only seem to visualize each separately, and while the E-field is still off-center, the target shape appears complete and centered in the outline box. This doesn't seem to change regardless of what I use to create the shape.dat file or what I use to visualize the ddpostprocess output.
While I know I can get a more complete image by changing the fractional extension of calculation volume in ddscat.par, my main concern is that the tracks I specify in ddpostprocess.par will not be directly through the center of the target as desired, since the entire shape is offset by some amount. Does anyone have any insight on how to resolve this issue? Would the output point towards some error in shape generation, in writing ddscat.par, or in writing ddpostprocess.par?
Thanks,
Rahil U.