Go ahead and look at the game files being loaded by the client. You'll find that they are packaged into sets of 2-3 total files, sharing points of origin. These have been referred to by data-miners and XIV VFX mod developers as "stacks". I'm merely using the term that's already been used for this.
Alternatively, look at the tools by which those VFX are replaceable and how exactly their categories allow for the replacement or omission of those VFX. We can literally look at how the game is assigning and actioning these files.
Far from necessary.Let's use the following as an illustrative example.
Those use, again, only a caster and target VFX stack (or, "package" to use your term instead) each.such as the vfx where to players are tethered together like Cover or any number of boss mechanics.
What I suggested (yanking the axe to your location via the old Holmgang to-target animation) entails only a single reference to location, and that location would not be dynamic once set.it's utilizing existing points of reference that are already being tracked as opposed to generating a new one dynamically that then also has to be tracked which requires extra computational resources to have that point merely exist and is not considered efficient and so tends to not be used unless there is more going on than just a superficial visual element that lasts for 2-3 seconds.