Why does DataStudio compute velocity from x(t) the way it does?

damosuz

A motion sensor gives the position of an object at times separated by Δt. DataStudio then calculates velocity by this method:

v(t) = (x(t + 3Δt/2) - x(t - 3Δt/2)) / 3Δt

I wonder why it doesn't do it like this: v(t) = (x(t + Δt/2) - x(t - Δt/2)) / Δt

I know it gives the same v if acceleration is constant since in both cases it corresponds to average speed in the middle of the interval, but if acceleration is not constant (e.g. mass-spring) it must give a value for v that is not as accurate.

Thank you.

cepheid

Maybe the velocity is averaged over multiple timesteps as a smoothing (filtering) feature. An average velocity computed using every single sample might have too much high frequency jitter due to noise.