This is an earlier experiment. Read the amendments below before treating its findings as current.
The question
The crosswind balance (no-pull zone plus crosswind over pull strength) had been tested only by changing pull strength. The 80 m no-pull zone had never been changed, and a 'weaving' idea explained the wing's extra 45 m.
What was tested
Half-hour runs, twelve layouts: wings with speed-matching on; ideal craft (they respond instantly) with speed-matching off and a no-pull zone 40, 80 or 160 m either side of centre; a check of the weaving idea; collision timing.
What was found then
The wing settles too, by minute 15, at 142 m, where it sits without speed-matching. For ideal craft the resting offset follows the zone width one for one, each within 2 m of the arithmetic, on a traced layout. The weaving idea fails: that model could never push craft outward. Across 24 half-hour runs of wings and slow-responding craft, the latest collision was at 384 seconds.
What we learned later
Identical collision lists in long and short runs do not prove the runs followed the same path early on.
research/reviews/brief-c62.md
What changed
Nothing.
Corrections retained in the record
Earlier wording: in all 24 half-hour runs the last collision is at 384 seconds
Record correction: 384.1 s is the latest collision across the 24 runs, not each run's last
research/reviews/brief-c61.md
Numbers and source references
- 1 800 s
- run lengthresearch/reviews/brief-c60.md
- 142 m
- wing resting offset, matching onresearch/reviews/brief-c60.md
- ~45 m
- wing excess over the reference's 97 mresearch/reviews/brief-c60.md
- 57/55, 95/99, 175/175 m
- resting offsets, east/west, half-widths 40 / 80 / 160 mresearch/reviews/brief-c61.md
- 2 m
- largest gap to the arithmeticresearch/reviews/brief-c61.md
- 24; 384.1 s
- runs checked; latest recorded collisionresearch/reviews/brief-c61.md