Experiment 151
F1.6, Hill and trifid verdicts. 0105 solved Hill letter by letter (sizes 2-5, plain and affine, every alignment, four numberings) and 0040 solved trifid with any 27-symbol cube (periods 2-15, every offset) exactly: no fit on K4. These ciphers read the carved letters around each known letter too, so every single-letter error anywhere matters: each of the 97 carved letters changed to each other letter (2,425), each deleted (97), and a letter inserted before each place, every letter tried (2,522); the known places keep their numbering. K4's 5,044 variants first; scrambled copies only if a variant fits.
What the method does with the 24 known letters.
The Hilla Hill cipher A cipher that turns letters into numbers, takes them two, three or four at a time, and mixes each group with a fixed table of numbers. cipher (suggested by the HILL letters on the sculpture's tablethe table The right-hand half of the sculpture, 26 rows of the alphabet each shifted one place further, starting from the word KRYPTOS.) and trifidtrifid A cipher that turns each letter into three small digits, like floor, row and seat in a 27-seat cube, writes several letters' digits in rows, and regroups them, so each new letter is made from pieces of three old ones. (a three-dimensional letter cube) were ruled outdead Said of a period, grid or method when the known letters contradict it: no secret word of that kind can produce the carved letters. exactly. Both work on groups of neighbouring letters, so a single wrong, missing or extra carved letter anywhere could spoil the test. We try every such error and rerun both exact tests.
Words in the margin are defined on the glossary page.
The carved letters go in at the top; the reading comes out at the bottom.
Hill and trifid work on groups of neighbouring letters, so one wrong, missing or extra carved letter anywhere could spoil their exact verdicts.
Every place changed to every letter, deleted, or with each letter inserted before it: 5,044 versions of K4, each through both exact tests; the Hill counts compared with the same on 10 scrambled copies (0152).
No trifid cube under any error. Hill: 257 living settings with a check on K4, all 5x5 with one check, fourth of eleven against the copies.
This experiment produced no candidate: the search ruled everything out before a full text was built. Its product is the bounded negative in section 7.
Redo the method on a few letters with pencil and paper.
The same steps run by the code on the real letters.
the errors tried, for every place of the copper:
wrong: place 21 W -> each of the other 25 letters
missing: place 21 deleted, the rest moves left, a filler at the end
extra: each letter inserted before place 21, the last letter pushed out
each variant through Hill 2x2-5x5 (plain, affine, every alignment, four numberings) and trifid (periods 2-15, every offset)
Family, key space, the constraint under test, and the clues it reads.
Family G, C17, C21 (the HILL column: Hill under a slip), trifid as Sanborn's 'visual' cube (C3). Deciders unchanged: 0105's setting() (rows solved mod 2 and 13, CRT, invertibility; 'open' when too few equations), 0040's solve_setting() (union-find on coordinate variables, backtracking with the all-different rule, caps). Exact where decided; open settings counted. An error is modelled on the copper with the message numbering fixed (a deleted letter's place filled at the end; an inserted letter pushes the last one out).
Written before the run; the result decides it.
No variant fits Hill or trifid with at least one check; open settings at the rate 0105 found (5x5 affine). Zero candidates.
The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.
5044 variants: Hill alive with checks 257, Hill open settings 70218; trifid consistent 0
What was searched, as the harness recorded it.
How this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.