Altars and builds: Shrink, Midas Touch, manic and depressed

Altars apply a state change to the run rather than a one-shot buff. Shrink reduces the player's hit volume, Midas Touch raises the value of banked haul, and the manic and depressed states change the combat pacing in opposite directions. The state-change framing is attributed to guide sources; the existence of Shrink, Midas Touch, manic, and depressed is official Steam feature text.

Shrink reduces the player's hit volume, which means fewer incoming hits land during a run. Shrink pairs well with short, fast-weapon runs where the player wants to minimize damage taken and bank the haul quickly. The pairing is attributed to guide sources.

Midas Touch raises the value of banked haul, which means the same haul prices at a higher tier when it reaches the shop. Midas Touch pairs well with runs that plan to bank a large haul and price it on a low-demand day. The pairing is attributed to guide sources.

Manic raises swing speed and aggression at the cost of stamina stability; depressed lowers swing speed in exchange for stamina recovery. Manic fits fast-weapon, high-pressure runs; depressed fits heavy-weapon, steady runs where stamina recovery is the binding constraint. The trade-offs are attributed to guide sources; the existence of manic and depressed is official Steam combat-state text.

Divine Prayers split into safer blessings and drawback-carrying prayers. Safer blessings are the right call on a short run or when the unbanked haul is at risk. Drawback-carrying prayers are the right call on a stable run where the drawback is offset by the run's existing momentum. The framing is attributed to guide sources.

Altar effects amplify an existing build rather than replace it. A fast-weapon run pairs with Shrink or manic; a heavy-weapon run pairs with depressed or Midas Touch. The matching rule is attributed to guide sources and is the simplest way to keep the altar system from destabilising a run that was already working.

The InsertCoins review calls out Shrink + Midas Touch as a named combination rather than a flat damage bonus: Shrink keeps the player alive through a long loot-heavy descent by shrinking the player's hit volume, while Midas Touch raises the value of the haul once it is banked. The two effects compound across a single run instead of stacking the same effect twice. The combination is the right call when the planned run is loot-heavy and the player can afford to take a deeper floor; the Worldeka dungeon guide frames altar choice as the per-run build mechanic and lists at least three named combos, including this one.

Manic cycling uses the manic combat state together with a fast weapon for a short aggressive descent that banks a haul before the stamina cost catches up. The community wiki devotes a section to altar effects and cycling, and the Worldeka beginner guide treats altar choice as the per-run build mechanic. Manic cycling fits a quick bank-and-return run; it does not pair well with a long loot-heavy descent where Shrink + Midas Touch is the better composition. The cycling framing and the per-archetype trade-off are attributed to guide sources.

Not every altar effect is worth taking. The skip-an-altar rule is to pass on a drawback-carrying prayer when the run's unbanked haul is already at risk or the run's archetype does not match the prayer's downside. Pass on depressed during a fast-weapon run; pass on manic during a heavy-weapon run; pass on any drawback-carrying prayer when the next floor is the descent boundary. The skip rule keeps the altar system from destabilising a run that was already working and is the simplest way to keep altar choice aligned with the run archetype.

Each individual altar description above carries a same-archetype trade-off: Shrink pairs with fast-weapon runs but is wasted on heavy-weapon runs that already stagger enemies; Midas Touch pairs with loot-heavy descents but is wasted on short bank-and-return runs where the haul is small; manic fits fast-weapon pressure runs but burns stamina in heavy-weapon runs; depressed fits heavy-weapon steady runs but is wasted in fast-weapon runs. Read the per-effect block and the matching rule together: the effect block names the trade-off, the matching rule binds it to the weapon in hand.