For a long time, the route architecture existed mostly in diagrams and design docs. A nice structure on paper.
Now I can finally mark a tangible milestone: the main story trunks of both versions of Actus II: Ortus—K/A-1 and M/N-1—are written.
With the shared first act and the Act II transition, the total is 258,500 words. Before calling it structurally closed, I went through 88 script files to make sure it held together. One of those rare cases where an enormous word count means there is a working dramatic machine underneath it, rather than one of those scripts that spend six textbox screens contemplating the deeper meaning of falling sakura petals or whether staring out the classroom window counts as character development. You know the type.
This does not mean "Ardengrad is finished." Not even close. Both routes still have Actus III and IV ahead, followed by the fifth and final act, Unitas. The milestone is different: both major consequences of the route split now exist in actual text instead of blueprints from the Dalmax Design Bureau.
The important part is not the word count. When I planned K/A’s "Love, Komfedmol and ‘Vesna’" and M/N’s "Dame Kätterle Liebhaber", there was one familiar landmine: the titles might end up being smarter than the actual writing. Fellow writers know the problem: you can build two beautifully opposed routes in a design doc, sit down to write them—and end up with the same story, just with the furniture rearranged.
Praise Mother Time, that is not what happened.
K/A went outward: a newspaper office, a print shop, the street, a camp, elections, public politics. It starts with a person learning to live and act with others, then asks a less comfortable question: where does solidarity end, and where does a common cause begin deciding for you what you are allowed to risk and how much of your life you are expected to give to it?
The result is a dense 100k+ word ensemble political drama with branching, gradually turning into an election thriller.
M/N went inward: a contract, an access pass, an archive, sealed files, institutional dependence. I deliberately did not write the princely House of Falkenstein as some archetypal Evil Inc. It genuinely helps people: it pays, provides medical care, gets lawyers involved, and solves problems Yura and Yesya simply couldn’t handle on their own. But the deeper Yura gets inside that system, the uglier the price of its efficiency starts to look. Behind procedures that genuinely work, other things begin to surface—things that can no longer be dismissed as simple aristocratic paternalism.
So the conflict is not "when will Yura realize everyone around him is evil?" The question is nastier: what do you do when an institution can genuinely save people, claim authority over their lives, and have very good reasons not to let you look too closely at how it actually works?
M/N became a 100k+ word institutional thriller, archival investigation, and intimate psychological drama.
In the end, both routes arrived at the contradiction I built the split around. K/A says that without ties to other people, an individual is powerless against a system—while those same ties can consume the individual. M/N insists on autonomy and authorship over your own life—and immediately runs into the fact that autonomy alone is not enough to actually change anything.
The part I am happiest with is that the script does not spend two hundred thousand words lecturing the player about which button at the route split was the "morally correct" one. Both routes eventually start attacking their own starting assumptions.
So yeah. When I came up with this ridiculous system of act titles, it was pure architecture. Now the shared first act and the first two major trunks after the split amount to an honest quarter of a million words of script.
There is still plenty of that architecture to write. First, though, I should probably turn all this writing into an actual functioning visual novel.