I have been working with the cosmic rays example which comes with the FLUKA installation. I use FLAIR, so whenever I want to activate or deactivate the dpmjet option I just click on the check square in the “Run” (see image below)
Now I want to run the generated .inp file using the command line (more specifically, I want to run it in a cluster which only has FLUKA installed). This is, I generate the .inp in FLAIR, checking the “dpmjet” box in run. Then, can I run my .inp file using the standard
/pathtofluka/bin/rfluka -M 5 file.inp?
Thank you very much!
P.S.: I have noticed that generating the .inp this way, and then opening the .inp with flair again, the dpmjet check is still marked. So does FLUKA save this information somewhere in the file?
No, you can’t, because this command invokes the standard fluka executable, while you need the flukadpm executable that embeds the DPMJET and RQMD libraries (in charge of ion-nucleus reactions). This requires to add to the above rfluka command the option -e /pathtofluka/bin/flukadpm < or just -d >
before the input file name. Note that the same flukadpm choice has to be explicitly made also in the Run/Runs tab of Flair (in the Exe field).
The dpmjet option offered by the box refers only to the corresponding Preprocessor variable in the input file, which just controls the source settings. When the variable is not set (i.e., the #define card is commented out), ions are split into single nucleons (by the PHYSICS/IONSPLIT card). When instead it is set, ions preserve their identity and so require the aforementioned flukadpm executable.
By the way, if I run a run both in flair, selecting the dpmjet exe, and the same .inp in the console, with the -d option, using the same seed, should I get the same output? I have tried this and, while the distribution is very similar, they are not exactly the same…
Solved! I wasn’t linking ldpmqmd when running via console… I guess that flair already does that.
