Typically phase cancellation will be enabled during hitless switching to prevent a phase transient (phase hit) from immediately propagating to the outputs when switching between two frequency-locked reference inputs with a fixed phase offset. The phase cancellation persists indefinitely in the use case scenario when phase slew is not enabled commonly referred to as phase buildout. The inputs are frequency-locked when they have same exact frequency (0-ppm offset), or have frequencies that are integer-related and can each be divided to a common frequency by integers. The hitless switching specifications (t_HITLESS and f_HITLESS) are valid for reference inputs with no wander. In the case where two inputs are switched but are not frequency-locked, the output smoothly transitions to the new frequency with reduced transient.