The efficiency curve is determined by a simultaneous fit to all experimental data. In the case that no normalisation data are given at all, then it is assumed to be equal to 1 for the first source and the other sources are fitted to this. In case multiple normalisations are given for a source, the global fit will minimise using all these values plus the constraints from the efficiency data of the other sources and their experimental normalisations. If source data given under efficiency doesn't have experimental values of normalisation constants, it's sufficient to simply type a dummy filename, i.e it doesn't have to exist. However, the ordering of the sources under -n must match those under -e.

Usage:

geff -e <eff1.dat> -n <norm1.dat> ... -e <effX.dat> -n <normX.dat> -r <E_low>:<E_upp>

effX.dat is the file containing the energy and efficiency values for source number X. The format is 4 columns:

   Energy (keV) | Error (keV) | Efficiency (arb.) | Error (arb.)

normX.dat is the file containing experimentally determined normalisation constants from arbitrary units to absolute % for source number X. The format is 2 columns:

   Normalisation (%/arb.) | Error (%/arb.)

You can also set the fitting and plot range with -r :

geff --help