Adds their unique unit(s) to the roster below and applies any economy bonus they have.
Pick a unit to constantly produce, and this finds the fastest path there — automatically deciding which economy upgrades and extra Town Centers are worth it along the way. Add a stage to plan a transition instead — e.g. Scout Cavalry in Feudal, switching to Knights once Castle hits. Add a unit to train it alongside the last stage's own unit, concurrently, instead of switching to it.
Hunt Preferences (berries are always assumed; a real Arabia map has deer OR chickens, never both)
Off (default): a realistic economy — sheep, then wood, then berries/hunt up to pop 20, no Barracks before Feudal, Loom right before advancing. On: skip all of that and build everything the instant it's affordable, prioritizing the target unit above all else. A later stage REPLACES the earlier one's production once its age is reached, it doesn't run alongside it. Efficiency scales down gather AND villager-training rate together — a rough stand-in for real-play friction (idle time, imperfect pathing, reaction delay) this idealized math otherwise ignores entirely. 100% = no friction; not a researched default, meant to be tuned against real games.
Add units above to see the split.