V91:Derivation of Standard-Model Gauge Preference via a Stability–Complexity Trade-off in Finite Update Systems

We present a transparent finite-update simulation that tests how the preferred gauge group depends on explicit design priors. Using SU(N) dynamics (N=2–5) with a shared diagnostics suite, we evaluate winners across a 4×3 cross matrix spanning (i) score modes that toggle complexity-related penalties and (ii) engine modes that control SU(2)-specific treatment and damping uniformity. We find that stability alone (no-cost scoring) can favor higher-dimensional groups such as SU(5), whereas introducing an explicit complexity cost selects SU(3) as a robust optimum with large score margins. Core diagnostics simultaneously satisfy (i) structural dominance (bare fraction ≈1.12%), (ii) an SU(2) improvement under a Higgs-like scaffold (Δ≈0.314), (iii) information localization (≈0.139), and (iv) non-abelian selection in a minimal ablation (ABL-1). A key insight from V91 is that during the update process, a "winner" is not a latent fact but an undefined quantity—it becomes meaningful only at the scoring/commit step. Likewise, questions such as "where was the particle before observation?" are ill-posed in a finite-update view, because "where" is defined only relative to an observer's present update window (Now). These results support a stability–complexity trade-off interpretation: the Standard Model's SU(3) sector emerges as an optimal balance between dynamical stability and information/implementation cost in finite-update systems. The cross matrix itself—showing which priors select which winners—constitutes the primary scientific contribution of V91.