The use of active suspensions has been rekindled by increasingly stringent comfort requirements in light of the diffusion of electric and safe-driving vehicles. In particular, electromagnetic solutions can guarantee intrinsic reversibility and regenerative operation, which would allow for energy harvesting and/or reuse on active quadrants. However, the use of non-ideal actuators has an impact in the regenerative and active capabilities of any active damping solution. To address this shortcoming, this work aims at validating optimal active operation. Hence, a linear quadratic regulator is used to tune a linear optimal control strategy in a quarter-car model. This plant is equipped first with an ideal actuator. Later, this device is enhanced with multiple non-ideal features: inertial and friction contributions of the moving parts, compliance of the parts that exchange forces with the suspension and the resistive load of the motor windings. Details regarding comfort, handling, are compared. Then, the realistic actuator model is compared for comfort- and handling-oriented control tasks. For both strategies, operational aspects like dynamic performance, control calibration, suspension duty cycle, damping force-speed characteristics and power absorption are discussed. It is demonstrated that a significant part of the control effort is used to mitigate non-ideal features of the realistic actuator.