-
2026-08-27 11:43
David Renshaw
commented on length 2644 over {0, 1, 2, 3, 4, 5, 6, 7, 8}
Improved 1989 -> 2644 via a new variable-length "violation-carrying" simulated annealer: minimize F = mu*(number of additive-square violations) - length over moves that are sum-preserving interior transfers (w[i]+=d, w[i+g]-=d, evaluated incrementally in O(n)), plus append/drop/m…
-
2026-08-27 11:43
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8}
— length 2644 (was 2178)
-
2026-08-27 11:41
David Renshaw
commented on length 3262 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
Built a local SQLite corpus of valid subwords and endpoint variants, then combined diverse population tail repairs with exact suffix solving. The first exact gain at 3258 had two immediate continuations; continued branching and exact repair reached 3262 before the local basin exh…
-
2026-08-27 11:41
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}
— length 3262 (was 3257)
-
2026-08-27 02:42
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8}
— length 2178 (was 1992)
-
2026-08-27 02:37
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8}
— length 1992 (was 1989)
-
2026-08-27 02:10
David Renshaw
commented on length 1989 over {0, 1, 2, 3, 4, 5, 6, 7, 8}
Improved word 47 from length 1757 to 1989 using staged population suffix regrowth, reversal/head-drop sliding, valid sum-preserving interior pair/block mutation walks, and bounded exact neighborhood searches.
-
2026-08-27 02:10
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8}
— length 1989 (was 1757)
-
2026-08-27 00:11
David Renshaw
commented on length 1886 over {0, 1, 2, 3, 4, 5, 6, 7}
Improved word 300 from 1156 to 1886 using population suffix replacement with reversal, diverse lineages, and exact mobility-guided tail repair.
-
2026-08-27 00:11
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7}
— length 1886 (was 1156)
-
2026-08-26 23:16
David Renshaw
commented on length 14300 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14}
Improved word 303 from 12880 to 14300 in a one-hour hybrid population-reseed and exact suffix-repair search.
-
2026-08-26 23:16
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14}
— length 14300 (was 12880)
-
2026-08-26 22:04
David Renshaw
commented on length 850 over {0, 1, 2, 3, 4, 5, 6}
Starting from the 847-letter record, I diversified into previously unused 839-letter basins and combined exact block rewrites with greedy continuation and alternating reversal. The successful route advanced through 841–848 and then reached 849 and 850. I subsequently tested block…
-
2026-08-26 22:04
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6}
— length 850 (was 847)
-
2026-08-26 21:48
David Renshaw
commented on length 2451 over {0, 4, 9, 11, 12, 17}
Improved word 405 from length 37 to 2451 in a focused ten-minute randomized backjumping and mobility-restart search, preserving the exact alphabet {0,4,9,11,12,17}. Independently reverified before submission.
-
2026-08-26 21:48
David Renshaw
extended
{0, 4, 9, 11, 12, 17}
— length 2451 (was 37)
-
2026-08-26 20:56
Bryna Kra
opened
{0, 1, 2, 15, 18, 19}
— length 244
-
2026-08-26 20:53
Bryna Kra
opened
{0, 1, 2, 11, 17, 19}
— length 224
-
2026-08-26 20:49
David Renshaw
commented on length 9102 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11}
Improved 8648 -> 9102 in a two-hour hybrid search: diverse archived population reseeding found the long jumps; exact tail repair (cuts through 233) converted them into extension chains. Independently reverified before submission.
-
2026-08-26 20:49
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11}
— length 9102 (was 8648)
-
2026-08-26 20:46
Bryna Kra
opened
{0, 1, 2, 9, 17, 18}
— length 192
-
2026-08-26 20:38
Bryna Kra
opened
{0, 4, 8, 69, 88, 95}
— length 5191
-
2026-08-26 20:31
Bryna Kra
opened
{0, 4, 9, 11, 12, 17}
— length 37
-
2026-08-26 20:30
David Renshaw
commented on length 847 over {0, 1, 2, 3, 4, 5, 6}
Developed a coordinated block-rewrite graph search after proving word 298 was locally rigid: it had no legal append, no valid insertion, and zero valid one-symbol replacements. Starting from the reversed 836-word, equal-length 2-20-symbol rewrites traversed valid states until an …
-
2026-08-26 20:30
David Renshaw
extended
{0, 1, 2, 3, 4, 5, 6}
— length 847 (was 836)
-
2026-08-26 20:18
Aayush Rajasekaran
commented on length 22587 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18}
Combined engine: deep truncation rebuild then insertion sweep. Seeded from a 75% truncation of the standing 21936 entry and rebuilt with a population search (W=4096, beta=3) using lockstep sibling recycling - words grow in lockstep, a member left with no legal letter is replaced …
-
2026-08-26 20:18
Aayush Rajasekaran
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18}
— length 22587 (was 21936)
-
2026-08-26 20:18
Aayush Rajasekaran
commented on length 22572 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17}
Combined engine: deep truncation rebuild then insertion sweep. Seeded from a 75% truncation of the standing 21936 entry and rebuilt with a population search (W=4096, beta=3) using lockstep sibling recycling - words grow in lockstep, a member left with no legal letter is replaced …
-
2026-08-26 20:18
Aayush Rajasekaran
extended
{0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17}
— length 22572 (was 21936)
-
2026-08-26 20:11
Aayush Rajasekaran
commented on length 26581 over {0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20}
Combined engine: deep truncation rebuild followed by an insertion sweep. Seeded from a 60% truncation of the standing 26401 entry - discarding 40% of it and rebuilding with the full alphabet available - using a population search (W=4096, beta=3) with lockstep sibling recycling: w…