Development covered 20 Jun 2026 to 22 Jun 2026 (commit dates).
I doubled the level ladder of my xianxia idle RPG this week, from fifty levels to one hundred, and the interesting part is that nothing about the game's total length changed. The realms, the tribulations between them, the real-world time a full climb takes: all untouched. What changed is the grain of the thing, and grain turned out to be most of what pacing is.
Fifty levels was lying about the climb
The old ladder had ten realms of five named sub-stages, early through pinnacle. Between two adjacent realms sat weeks of real time at the top end, represented on screen by one number ticking up by one. Players do not experience weeks; they experience feedback loops, and a feedback loop with a period of a week and a magnitude of one is no feedback at all.
The fix gives every named sub-stage an intermediate rung of its own, reported on screen as a half-step. Nothing takes longer; the same curve just reports itself twice as often, and the label appears everywhere the game talks to you. The psychology is trivial and the effect is not: the top of the ladder now produces a visible event at half the period, and events are what an idle game actually sells.
One currency, fewer regrets
The same window forced the skill tree's economy to a decision. It had been reachable through two currencies, which sounds like depth and behaves like a trap: new players spend the scarcer one early, learn later it gated something better, and carry the regret for the rest of the run. The Mortal World now runs its whole tree on one currency, and the second is simply not spendable here. Depth comes from what nodes do, not from which wallet they draw from. If a later realm needs a second axis of choice, it will be added where players already have context, not where they are still learning the interface.
Boards that bracket time
Leaderboards in a game where power compounds with age have a fairness problem: the list saturates with old characters and the top stops being a race. The design here brackets the boards by your cultivator's age, so you compete inside your own decade of game-time, with a separate board for newcomers. It is the same reason sports have weight classes. The interesting design residue was what to do about the very old brackets, where the population thins; the answer was to widen brackets with age rather than let boards die.
Hysteresis in a guidance card
The guidance card suggests one next action, worked out from about nineteen rules. The naive implementation flickers: as thresholds are crossed, two rules trade the card back and forth, and a suggestion that changes every few seconds trains players to ignore it. The fix was hysteresis: once a suggestion is shown, it holds until either it is completed or a materially better one appears. The card now behaves like a fellow cultivator with one opinion instead of a slot machine.
None of these changes touched the game's total length or its balance sheet. They changed how often the game tells the truth about progress, which turns out to be the actual job.
Play it at Idle Cultivation Game.
Top comments (0)