Experiment 185

“F4.2: trifid with a free 27-symbol cube at periods 16-97, every alignment, by constraint propagation: exact”

F4.2. 0040 decided every free trifid cube at periods 2-15 (no cube fits at any alignment). Its decider does not depend on the period: each known letter ties its three coordinates to coordinates of three carved letters in its block, whatever the block length. Here periods 16-97 (97: one block), every alignment, 4,633 settings, with the same caps; ten shuffled copies run through the same sweep as the null.

revertedstatus
experiment 40parents
6 stime, all phases
0candidates
0/24known letters
unclocked
1

The experiment method

What the method does with the 24 known letters.

Sanborn is a sculptor; a 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. code puts the alphabet in a 3 x 3 x 3 cube and mixes the three coordinates of neighbouring letters in groups. Groups of up to 15 letters were already 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.. Here every longer group, up to the whole text, with every starting point: do the known lettersthe known letters Sanborn has confirmed that positions 22 to 34 of the message read EASTNORTHEAST and positions 64 to 74 read BERLINCLOCK. allow any arrangement of the cube at all?

Words in the margin are defined on the glossary page.

1.1

The method at work

The carved letters go in at the top; the reading comes out at the bottom.

OBKRUOXOGHULBSOLIFBBWEASTNORTHEASTOTWTQSJQSSEKZZWATJKLUDIAWINFBBERLINCLOCKWGDKZXTJCDIGKUHUAUEKCARIN: THE COPPER, ALL 97 LETTERS, THE 24 KNOWN PLACES IN ACCENTtrifid, any 3 x 3 x 3 cube: each cell one group length, at every starting place; does some cube fit the 24 known letters?2-15: 004016-97: this run, 4,633 settings, none cut short21630507097no cube fits K4one setting, groups of 20 from place 1: the two groups with known letters; their ties merge the cube's 81 coordinates into 22 classes, and no cube fills themWEASTNORTHEASTOTWTQSNFBBERLINCLOCKWGDKZX21406180each cell one text: some cube fits at some setting?K410P1K4 and 10 scrambled copies: 0 of 4,633 settings each; P1, planted at groups of 30: its cube found among 288OUT: NO READING; NO TRIFID CUBE FITS K4 AT ANY GROUP LENGTHEASTNORTHEASTBERLINCLOCK
the idea

A trifid code puts the 26 letters and ? in a 3 x 3 x 3 cube and mixes the three coordinates of neighbouring letters in groups: a sculptor's cipher. Groups of up to 15 letters were already ruled out (0040).

the check

Every group length from 16 to the whole text, every starting place: each known letter ties its three coordinates to coordinates of carved letters; merging the ties and filling the cube so no two letters share a place decides each setting.

the bound

No cube fits K4, or any scrambled copy, at any setting; a planted text's cube is found. F4.2 closed.

2

Best candidate against the known letters

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.

3

Walkthrough, by hand

Redo the method on a few letters with pencil and paper.

  1. Take period 20 with blocks starting at place 1: places 21-40 form one block.
  2. Write each block letter's three coordinates in three rows, read the rows one after another, regroup in threes.
  3. Each known letter's coordinates must then equal certain coordinates of certain carved letters.
  4. Merge equal coordinates; try values 0-2 for each merged group so no two letters share an address.
  5. If nothing survives, no cube works at that period and start.
the worksheet: the first steps by hand, on squared paper
WHY A GROUP LENGTH LEAVES SO LITTLE ROOMIn a group of 20, each known letter'sthree coordinates must equal particularcoordinates of three carved letters. The24 known letters give 72 such ties among81 coordinates.Merged, they leave a handful of freegroups of coordinates; filling them sothat all 27 symbols get different placesfails every time on K4, and on scrambledcopies too.
4

Worked example, generated

The same steps run by the code on the real letters.

period 20, blocks from position 1: the block 21-40 carries EASTNORTHEAST at 22-34 against the carved FLRVQQPRNGKSS

81 coordinate variables merge into 22 classes under all the ties from the 24 known letters
backtracking over the classes with the all-different rule visited 1 nodes and found 0 cube(s)
5

Hypothesis, for a cryptanalyst

Family, key space, the constraint under test, and the clues it reads.

C3 (Sanborn's visual, three-dimensional thinking: a cube), F4; Delastelle's trifid keeps one letter per letter on 27 symbols. Model as 0040: the k-th coordinate of carved letter j in a block of n equals coordinate (3j + k) div n of plaintext letter (3j + k) mod n. 72 equalities among 81 coordinate variables, then backtracking over the classes with the 27-distinct-addresses rule and 9-per-value counts. Statistic: settings admitting a cube, K4 against 10 shuffled copies; any surviving cube decrypts and is scored.

6

Prediction

Written before the run; the result decides it.

No cube at any setting, as at 2-15; or cubes at the copies' rate whose decryptions are not English. Zero candidates.

7

Method

  1. For each period 16-97 and alignment: the coordinate equalities the 24 known letters force.
  2. Union-find, then backtracking over classes with the address and count rules (caps 3M nodes, 5,000 cubes).
  3. Count settings with cubes; decrypt and score any.
  4. The same sweep on 10 shuffled copies.
8

Result and bounds

The harness's record: candidates, gates, loss, and the negative result this experiment leaves behind.

0candidates returned
0passed every gate
0/24known letters, best
–lowest loss
nonechampion before

4633 settings; 0 admit a cube on K4; 0 capped; copies with cubes: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

What was searched, as the harness recorded it.

family
trifid, free 27-symbol cube, periods 16-97, every alignment, constraint propagation and backtracking
settings
4,633
settings with cubes
0
settings capped
0
node cap
3,000,000
solution cap
5,000
k4 settings with cubes
none
max nodes
364
best
none
plant
period
30
offset
5
cubes
288
true cube found
yes
controls
  1. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  2. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  3. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  4. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  5. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  6. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  7. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  8. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  9. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
  10. settings
    4,633
    with cubes
    0
    capped
    0
    best english
    None
seconds
6
exhaustive
yes

How this experiment sits among the others is on the lineage page; its family's open hypotheses are on the families page.

Rendered from the record on 2026-10-18 12:26 UTC; updated after every experiment.
view on GitHub