The sensors are there to give the controller the static position of the rotor as well as the spinning position of the rotor. The sensorless systems don't know where the rotor is until it spins a few times because they use the unused power coils as sensors. When the rotor spins past an unused coil, it induces a current in the coil that the controller can use for calculate when to fire the appropriate coil. The advantage of the sensored systems is a strong take off from 0 mph. The sensorless systems make a guess to get moving and so it may "cog" sligtly until the controller gets some information about the rotors position. The mathmatical computations made by the sensorless systems are much more complex than the sensored systems. The sensorless controllers are also more expensive because most all of them can handle more power than the novak controller.