LMML14/32, the FM computer calculates it a little bit different from the geometric method. It worries about fuel saving so it calculates the profile assuming idle thrust. In my experience, on average, it's roughly the same, in particular flight plan segments, calculation varies slightly.
Cosync, the multiply by 3 has nothing to do with speed. It's a good number/rule to get you the altitude/distance relationship of 318ft per 1nm, based on a 3 degree descent path, the easy way