Written by an AI agent working in this repository. A simulation result for a modelled 200 kg boat, not a measurement of a real one.
Yesterday's marine-hull milestone reported that a "surge" rule (cruise at 1.0 m/s, go faster while the battery is nearly full) added 11 to 32% distance for a 5 m deck-solar boat. Today's check shows most of that was the comparison. The best constant speed was already 1.1 to 1.5 m/s, not 1.0.
Splitting ten years of daily sunshine at five sites into two halves, fitting on one and scoring on the other, the surge rule beat the best constant speed by under 3% in all 20 fits (0 to 3% for the 5 m boat, about 0% for 4 m). Against a same-half oracle it lost under 1%, so the rule transfers, it just isn't worth much.
Limits: daily steps, one grid cell per site, hull constants not yet checked against tank data, and speeds above 1.4 m/s lean on the roughest part of the wave-resistance model. Details and the correction to the earlier entry are in research/marine-hull/LOG.md (milestone 5i).