Rob
Ansys Employee

With steady particles the whole trajectory is computed in one go. So each track is seen from inlet (injection) to wherever it exits the domain. The plot looks like coloured string. 

Unsteady particles are release every (particle) time step from the injection and their position is updated by the distance they can travel in one time step. If you plot this you'll see many dots/spheres in your domain. 

The result should be much the same provided you give the transient particle case time for the particles to reach the exit. A few percent different could be down to insufficient time, or lack of calculated trajectories: if one stream misses the surface how does that alter the %?