Version: SMASH-3.2
Output

Output directory

Per default, the selected output files will be saved in the directory ./data/<run_id>, where <run_id> is an integer number starting from 0. At the beginning of a run SMASH checks if the ./data/0 directory exists. If it does not exist, it is created and all output files are written there. If the directory already exists, SMASH tries for ./data/1, ./data/2 and so on until it finds a free number.

The user can change output directory by a command line option, if desired:

smash -o <user_output_dir>

SMASH, by default, will create the specified folder if not existing or will use it if the specified folder exists and is empty. However, if the folder exists and is not empty SMASH will abort with an error to avoid overwriting existing files.


Output content

Output in SMASH is distinguished by content and format, where content means the physical information contained in the output (e.g. list of particles, list of interactions, thermodynamics, etc) and format (e.g. ASCII, binary or ROOT). The same content can be printed out in several formats simultaneously.

For an example of choosing specific output contents see Output configuration examples.

These are the possible contents offered by SMASH:

Attention
At the moment, the Initial_Conditions and Rivet outputs content as well as the HepMC format cannot be used with multiple parallel ensembles and SMASH will abort if the user tries to do so. The Collisions content, instead, is allowed, although in it collisions coming from different ensembles are simply printed all together in an effectively unpredictable order and it is not possible to know which one belongs to which ensemble. Therefore SMASH warns the user about this fact and this setup should only be used if in the data analysis it is not necessary to trace back which data belongs to which ensemble.

Output formats

For choosing output formats see Output configuration examples. Every output content can be printed out in several formats:

  • "ASCII" - a human-readable text-format table of values.
  • "Binary" - a binary, not human-readable list of values.
    • The binary output is faster to read and write than text outputs and all floating point numbers are printed with their full precision.
    • For "Particles" and "Collisions" contents, it is basically a binary version of the corresponding ASCII output.
      Also for binary format it is possible to customize the quantities to be printed into the file.
    • For the other contents the corresponding documentation pages about the ASCII format contain further information.
  • "Oscar2013_bin" - alias for the "Binary" format with a predefined set of quantities.
  • "Oscar1999", "Oscar2013" - aliases for the "ASCII" format with a predefined set of quantities.
  • "Root" - binary output in the format used by the ROOT software
    • Even faster to read and write, requires less disk space
    • Format description: ROOT format
  • "VTK" - text output suitable for an easy visualization using third-party software
  • "HepMC_asciiv3", "HepMC_treeroot" - HepMC3 human-readble asciiv3 or Tree ROOT format see HepMC Output for details
  • "YODA", "YODA-full" - compact ASCII text format used by the Rivet output, see Rivet output for details
Note
Output of coordinates for the "Collisions" content in the periodic box has a feature: Collision output in box modus