I think it’s a testament to Asobo’s ambition to do things right, or don’t do them at all. Of course they could’ve come up with a rudimentary implementation without the need for complicated physics calculations. All they’d have needed to do was to throw in some lines of code, something like random_shake_chance and random_shake_intensity, fiddle around with a numbers a bit, badaboom, there’s your FSX-style turbulence model. It’s a simulacrum, though.
Now, obviously it doesn’t really matter on the user’s end how exactly things are done as long as the final result is somewhat authentic. From Asobo’s perspective, however, they’re working on a product that’s still in the infancy of its decade-long lifespan. They can count on more computing power being available to their customers in a couple of years, which in turn allows them to come up with more resource-hungry solutions in the long term than those are feasible today. And as fellow simmers, they remember the bloody mess of legacy code and redundant, badly-working systems and interactions FSX was at the end of its lifespan. I do too. Which is why I welcome Asobo’s commitment to doing things properly, rather than fixing blunders with gorilla code.