nx_paradigm_neuromorphic_gate.nx

nx_paradigm_neuromorphic_gate.nx -- honest cycle on the NEUROMORPHIC/spiking paradigm (next surfaced target). Hypothesis (testable): event-driven SPIKING computation does FEWER ops than DENSE clock-driven, via sparsity. Independent reference: a real op-count from a deterministic integrate-and-fire simulation -- dense polls every neuron every timestep (M*T ops); spiking only processes actual spike events (spikes * fan-out). Discipline (find where it LOSES too): spiking wins ONLY below the crossover firing-rate ~1/fanout; above it, per-spike overhead beats dense. Record the REGIME where the paradigm pays, honestly -- not "spiking is better". No hw writes (Rule 26). expect_exit: 0 license_tier: ORIGINAL