code wiki / _hdl_build / nx_genloop_gate.nx
nx_genloop_gate.nx
buildroot/runtime/_hdl_build/nx_genloop_gate.nx
about
nx_genloop_gate.nx -- R6: CLOSE THE LOOP. One circuit, all our code, no hand-crafting:
GENERATE (t2mesh prompt -> field) -> MESH (surface nets) -> RENDER the EMITTED mesh (existing sovereign
nx_meshrender/nx_render_core pipeline, per-vertex lit -- composed, not duplicated) -> PHYSICS step driven
by the EMITTED rig (nx_autorig masses/bones -> spring-damper jiggle; mass has consequence) -> CRITIC score
(nx_critic_axes, the same rulers as the critic).
T1 the emitted-person render has content + structure. T2 the loop generalizes (dumbbell renders differently).
T3 critic scores it -- and honestly LOW (CLAY expected; a high score here would smell rubber-stamp).
T4 physics from the emitted rig: it MOVES, SETTLES, and MASS matters (light hand swings >> heavy chest).
T5 sim deterministic. T6 PNG knowledge/nx_genloop.png (eyeball the loop's own output).
Lighting = per-vertex Gouraud baked in model space (v0, disclosed). license_tier: ORIGINAL expect_exit: 0
dependencies 8 imports · 0 importers
imports: nx_syscalls.nxnx_sdfrender.nxnx_t2mesh.nxnx_meshgen.nxnx_autorig.nxnx_meshtex.nxnx_critic_axes.nxnx_png.nx
imported by: nobody (leaf or entry point)
call flow from main pre-order; caps 40 nodes / depth 6 declared; ↻ = already shown
structs
| none |
consts
| none |
functions
| 20 | func hw(s: *u8) -> i64 { var n: i64 = 0; while s[n] != (0 as u8) { n = n + 1 } sys_write(1, s, n); return 0 } |
| 21 | func pn(v: i64) -> i64 |
| 36 | func gl_normals(vb: *i64, tb: *i64, nv: i64, nf: i64, nbuf: *i64) -> i64 calls 1: ar_isqrt |
| 82 | func gl_sim(p: *i64, bones: *i64, masses: *i64, np: i64, nb: i64, root: i64, ox: *i64, vx: *i64, pk: *i64, steps: i64, J: i64) -> i64 called by 1: main |
| 118 | func main() -> i64 |