confused about "fix heat"

Dear lammps-users and developers,
    I am confused about the command "fix heat".

fix ID group-ID heat N eflux.

The unit of the flux is engery/time. It appears that N has no effect on the result???

N controls how often LAMMPS actually adds the energy to the simulation. The amount of energy added will be eflux multiplied by the time elapsed since energy was last added. Say, for example, that you want to add energy at a rate of 1 energy unit per time unit (eflux = 1). If a timestep lasts for one time unit (timestep 1.0), then if you set N = 1 LAMMPS adds one unit of energy each timestep. If you set N = 2, LAMMPS adds 2 units of energy every second timestep. N = 3 --> 3 units of energy every third timestep, etc. “Heat addition/subtraction is performed every N timesteps. The parameter can be specified as a numeric constant or as a variable (see below). If it is a numeric constant or equal-style variable which evaluates to a scalar value, then the determines the change in aggregate energy of the entire group of atoms per unit time, e.g. in eV/psec for . In this case it is an “extensive” quantity, meaning its magnitude should be scaled with the number of atoms in the group. Note that since has per-time units (i.e. it is a flux), this means that a larger value of N will add/subtract a larger amount of energy each time the fix is invoked.”