Posts: 3,384
Threads: 104
Joined: May 2012
Staff roles: Balance Dev
The Council Battleship was possibly a bad example. When I talk about model-, shiparch, powercore, ship package "assets" (and whatever I'm forgetting) of "clone" ships causing overhead, I am usually referring to things like the Marlin. Discovery unfortunately has a lot of "baggage" like that lying around that is easy to forget (FLStat doesn't show them without flicking on "Show blank items", for example).
I still believe it is best to discourage players from buying, for many billions of credits, what is essentially an unplayable ship. Especially if there's no guarantee that that ship will stick around or will ever even be playable.
It's easy to say "Just maintain it bro" until you're the guy who has to change the animations of the Insurgency Battlecruiser so its post-rework Battlecruiser Heavy doesn't shoot straight through the infamous Spinning Balls, then has to completely rehardpoint it - including model changes, that require cleanup aftwards, and need to also be applied to LODs that may have very funky geometry. Oh, and after that you have to adjust its hitbox and split the model into parts. This tends to break smoothing, requiring resmoothing of (large) parts of the model. Then you have to go and make damage models and fuses. And then you have to balance the collisiongroup stats, the fuse thresholds, and then the stats of the ship.
..I think people should be catching my drift by now. There was, perhaps, a time when maintaining an "unplayable asset" or two meant copy-pasting a file and swapping some texture names out. That doesn't really always apply anymore. I would expect Battleships to eventually - possibly - have similar, if not more, complexity to (reworked) Battlecruisers. That'd mean all of what I said above would have to be done for a ship like the Council Battleship.
I also forgot to mention wireframing (not an easy task for high-polycount, LODless ships, like most of Discovery's caps), and probably a few other things.