12.08.15 meeting day - daplff/hiWi_cont GitHub Wiki

ToDo:

  • Organize some stuff to show and discuss with Roland (e.g. validation)
  • If time, continue with yesterday's work

Showcases

  • Should probably organize folders etc. so it's easier to show
  • Tryout few visualisations
  • Use diff on output files from orig. program vs. through c++ files

results from the day

  • managed to load into vtk
  • wrote c++ program to load from netcdf to ascii vtk unstructured grid
  • Augmented with multiple timesteps into different files

results from meeting

  • netcdf cool, doesn't have to be directly loadable with paraview readers
  • can build own paraview python loader, similar to vtk outputter
  • should use python-netcdf to try and convert to vtk
  • gives easy modifying to binary format etc., reasonably fast, guaranteed to give working vtks
  • Scientific.IO.NetCDF should work - depends on netcdf4
  • use one netcdf-file per timestep
  • less unlimited dimensions (bad for space)
  • can load "finished" files already
  • use finite particle dimension per timestep
  • either too big or varying size - but maybe paraview doesn't like varying
  • when building, it might make sense to build in /opt/ - cleaner for dependencies etc than /usr/local
  • should look at this for free netcdf4 build/cool visio and [this] (http://commander.users.nerdzoo.de/cgi-bin/hgwebdir.cgi/SpaceFlood/file/684e4eb4f674/tools/convertMultiBlockToVTK.py "Some stuff that it seems Roland built") for introduction to python-vtk

final remarks on day

  • note: the sphysics sim does a recount every 100 timesteps on which particles are there and not - this means number of particles is changing
  • thought i'd be able to implement that today but alas, too much complix going on
  • might do fresh start on that or just revert
  • BUT: so far, single file with no unlimited dims is MUCH faster, and 15th of size of previous output file (with 2/7 of output, so still at least halved file size?)

other stuff for tomorrow/next time

  • next on to-do list: netcdf-python
  • hack an input file manually to see how paraview copes with changing number of unstructured grid points!