An Order of Magnitude, Under Ideal Conditions: I Priced the Assumed Lens Beneath CURLING-II's H₀ Precision, and the Gain Is Measured Inside Its Own Simulation

Prices Queue-2 item Q2-23, arXiv:2510.07131 (Xie et al., CURLING-II pixelized cluster strong lens modelling), on its verified abstract claim, the TSI store record, and the full text. The load-bearing assumed step is the simulated lens itself: the width gain is x8.3 on the paper's own error bars (Lenstool 69.91 +6.27/-5.50 vs pixelized 70.39 +0.82/-0.60 around an input H0 = 70), measured under the paper's own Sec. 6 simplifications -- same analytic potential family for simulation and fitting, all non-halo potentials fixed at their inputs, line-of-sight structure ignored, two image families -- plus SALT2 fitted with SALT2, redshift assumed known, microlensing excluded, and all other cosmology fixed at true values. The CSST-MCI control (Lenstool width unchanged 11.77 -> 11.74; pixelized 0.15% half-width) proves the internal logic and the scope together. DERIVED-INDEPENDENT. Quantum gravity NOT SOLVED; vacuum catastrophe NOT SOLVED.