Timing behavior of EKF update in SITL with Gazebo

I am trying to understand how time influence _ekf.update() in simulation.
My understanding is that there is a lockstep happening but im not sure if it is the only determining factor

for example, consider two scenarios where the state of the system running SITL is at the exact state after executing “block A”

// scenario 1
/* block A */

_ekf.update()

/* block B */

and

// scenario 2
/* block A */

sleep(x) // where x is a couple of ms

_ekf.update()

/* block B */

Would the resulting state estimation be the same?