Why can't you use the formula p=mv to calculate momentum for objects in any and all inertial reference frames that are moving with respect to an observer at an arbitrary speed u? this formula doesn't account for the momentum of the inertial frame itself the quantity mv ceases to be a conserved quantity in the limit as (u/c) approaches zero the quantity mv ceases to be a conserved quantity in the limit as w approaches the momentum approaches zero in the limit as approaches