nx_hairdyn_lib.nx

nx_hairdyn_lib.nx -- LIB: HAIR AS A DRIVEN CHAIN ON THE ONE SOLVER, WITH ITS PLANT MEASURED IN THE UNITS OF ITS OWN CITED BAND (2026-08-25). WHY THIS EXISTS. knowledge/gamefeel_oracle.conf has carried, since the oracle table was banked: hair_k_q10 120 420 per1024 secondary-motion-hair-2to5hz and NOTHING WAS EVER DERIVED FROM IT. The shipping hair plant is two hand-picked integers in nx_softbind (SB_K_HAIR=200, SB_C_HAIR=60) that sit inside that band by inspection, driving a SINGLE point -- so hair on /world/beach is one lagging joint offset underneath a painted skull band, and the cited row is a BAR WITH NO PLANT BEHIND IT. This lib is the plant, and its own gate measures it rather than asserting it. THE CHAIN. Hair is a strand, not a point. The game-convention lineage for this class is the SPRING-BONE CHAIN -- a bone pulled toward its rest pose RELATIVE TO ITS PARENT under a spring and a damper, with the parent's motion propagating down the chain (artellblender/springbones, https://github.com/artellblender/springbones, position and rotation over bone chains). hd_tick is exactly that, and it is why a strand reads as hair instead of as a wobbling head. That is a CONVENTION citation, not a physiological one, and it is labelled as such: it licenses the SHAPE of the model, never any number in it. ONE SOLVER, TWO CONSUMERS. Every segment is stepped by nx_softdyn's sd_step_axq -- the SAME integrator the soft-tissue lane reaches through sd_step_ax. A second spring integrator for hair would be the duplicate-ruler defect the estate has already paid for twice. K comes from sb_k_sym (nx_softbind's frequency->stiffness ruler) and C from sb_c_for_zeta (its damping ruler), so a hair constant CANNOT drift from a chest constant: both are the same two functions evaluated at different arguments. license_tier: ORIGINAL No hw writes (Rule 26).