Experiment 91

“The Berlin clock as a key schedule, joined to a K3-style move before or after the table”

0079 read the key off the set-theory clock with no move. Here each of its 60,480 schedules (six readings, seven steps, every start minute) is joined to every move of the type-1 route library (27,852 distinct routes), the move either before the table or after it. A schedule plus a constant leaves no free letter, so both orders are exact tests.

revertedstatus
experiment 79parents
19 mintime, all phases
0candidates
0/24known letters
unclocked
1

The experiment method

What the method does with the 24 known letters.

Experiment 79 used the Berlin clock as the keythe secret word The word or string a cipher needs besides the method.: set it to some time, read off a number for each letter, and move the clock on by a fixed number of minutes per letter. It found nothing. Sanborn also said he moved letters around in grids before going back to the tablethe table The right-hand half of the sculpture, 26 rows of the alphabet each shifted one place further, starting from the word KRYPTOS., so here every clock key is joined to every grid move of experiments 80 to 83, with the move before or after the table. A clock key has no free letters, so each combination either explains all 24 known letters or it does not.

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 ACCENTthe clock starts at 00:00 and moves one minute per letter; its lit lamps are the key, plus one constant00:21: 5 lamps lit01:10: 3 lamps litat place 22at carved place 7160,480 schedules x 27,852 moves x 12 table rulesmove then change: deepest run of checks0710the copper: 7ten shuffled coppersthe copperlamps lit at each place (last digit)0123412345234563456745678567896789078901890129012301234123451234523456345674567856789678907890189EASTNORTHEASTBERLINCLOCKRUGHLIWFRSJKKAYMZDJCHUKCthe move: a clockwise quarter turn of a grid 8 wide; where each known letter is carvedchange then move: key needed minus lamps at the message place (last digit); accent where it matches place 22's 9922241811525095273236531move then change: minus lamps at the carved place instead; accent where it matches place 22's 11563540542896164414510398E of place 22, carved at 71 as Z, needs 14; the 3 lamps there leave 11other known places agreeing with place 22: 1 change then move, 0 move then change; all 23 must agreechange then move, every schedule and move: 0 full matches, 0 for either crib word alone, and 0 on all ten scrambled coppersOUT: NO READING; NO CLOCK SCHEDULE FITS EITHER ORDEREASTNORTHEASTBERLINCLOCK
the stack

Experiment 79 read the key off Berlin's set-theory clock and found nothing. Sanborn said he moved letters in grids before going back to the table, so here each clock schedule is joined to each K3-style move, before or after the table.

the check

A schedule plus one constant leaves no free letter: key needed minus the clock's number must be the same at all 24 known places, read at the message place or at the carved place.

the bound

Six clock readings, steps of 1 to 30 minutes, every start minute: 60,480 schedules with 27,852 moves. No full match either way, and the deepest run, 7 checks, is what every scrambled copper reaches.

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. Set the clock to 00:00 and advance it one minute per letter; the key number for a letter is the count of lit lamps.
  2. Pick a move, say a clockwise quarter turn of a grid 8 letters wide.
  3. Change then move: E of place 22 ends up at some carved place j; find the number that turns E into the carved letter there.
  4. That number minus the lamp count at place 22 must be the same for all 24 known places.
  5. Move then change: the clock runs along the carved text instead, so compare the needed number with the lamp count at place j.
  6. Repeat for every start minute, step, clock reading, move and table rule.
the worksheet: the first steps by hand, on squared paper
CLOCK FROM 00:00, ONE MINUTE PER LETTER, LAMPS LIT; QUARTER TURN OF A GRID 8 WIDE; SCULPTURE'S ALPHABET, VIGENEREplaceknownmoved tocarvedkey neededlamps at placedifferencelamps at carveddifference22E71ZZ-E = 145931123A83CC-A = 262272124S95CC-S = 3722111825T10HH-T = 10825564B90UU-B = 1349103Change then move reads the clock at the message place, move then change at the carved place. Either way the difference must be one number.Change then move gives 9, 22, 22, 2, 9; move then change 11, 21, 18, 5, 3. Neither is constant: dead.
4

Worked example, generated

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

clock from 00:00, one minute per letter, lit lamps; move: clockwise quarter turn of a grid 8 wide;
sculpture's alphabet, Vigenere

change then move: needed number minus lamps at the message place must be constant
place  known  moved to  carved  needed  lamps  difference
   22      E        71       Z      14      5           9
   23      A        83       C       2      6          22
   24      S        95       C       3      7          22
   25      T        10       H      10      8           2
   64      B        90       U      13      4           9

move then change: the lamps are read at the carved place instead
place  known  moved to  carved  needed  lamps  difference
   22      E        71       Z      14      3          11
   23      A        83       C       2      7          21
   24      S        95       C       3     11          18
   25      T        10       H      10      5           5
   64      B        90       U      13     10           3
5

Hypothesis, for a cryptanalyst

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

Type 6 of directions.md: 'the same schedules combined with a moving stage (type 1)'. Scheidt: the system is 'simple, can be remembered' (C5); Sanborn pointed at the Mengenlehreuhr in 2014 and said BERLINCLOCK is a clue (C14, C11a). Schedules as in 0079: lit lamps, hour, minute, hour plus minute, the five-minute row and the time's digits, the clock advancing 1, 2, 3, 5, 10, 15 or 30 minutes per letter from every start minute, key value mod 26 plus one free constant. Moves as in 0090 (K3's rule, one or two quarter turns, plain and true-keyword columnar, forward and inverse). Change then move: the 23 key differences forced at the crib places must equal the schedule's, a dictionary lookup. Move then change: the clock runs over carved places, so the forced differences sit at the moved crib places; candidates come from a pair index and are filtered check by check (the time-digit reading only at one minute per letter here, where it is a cyclic text). 12 conventions (Vigenere, Beaufort, variant Beaufort; plain and cipher numbered on standard or KRYPTOS). The Carter and dedication dates of 0038 are periodic keys of period at most 10 and are covered exactly by 0080 and 0083. Ten shuffled coppers.

6

Prediction

Written before the run; the result decides it.

No combination explains all 24 known letters in either order. Change then move: no schedule matches either whole crib stretch beyond the scrambled-copper count. Move then change: the deepest partial match is about 6-8 checks, inside the scrambled-copper range. Zero candidates.

7

Method

  1. Build the 60,480 clock schedules and the deduplicated route library.
  2. Change then move: for each route and convention, undo the move at the 24 crib places, compute the forced key numbers and their differences from the first, and look them up in an index of every schedule's differences at the crib places (all 24, and each crib stretch alone).
  3. Move then change: for each clock reading and step, find the start minutes that agree at the first two moved crib places through a pair index, then filter by the other places, recording the deepest agreement.
  4. K4 first, then ten shuffled coppers as time allows.
  5. Decrypt and return any full match.
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

27,852 routes x 12 conventions x 60,480 clock schedules. Change then move: full K4 0, controls [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]; EASTNORTHEAST stretch 0 (controls [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]); BERLINCLOCK stretch 0 (controls [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]). Move then change: full 0, deepest K4 7, controls [7, 7, 7, 7, 7, 7, 7, 7, 7, 7] (10 completed).

What was searched, as the harness recorded it.

routes distinct
27,852
conventions
12
schedules
60,480
readings
lit lamps, hour, minute, hour plus minute, five-minute row, time digits
steps minutes
1, 2, 3, 5, 10, 15, 30
start minutes
1,440
change then move
k4 full
0
controls full
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 stretch13
0
controls stretch13
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 stretch11
0
controls stretch11
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 stretch hits
none
move then change
clock families
36
time digits steps
1
k4 full
0
k4 best depth
7
controls best depth
7, 7, 7, 7, 7, 7, 7, 7, 7, 7
controls completed
10
k4 depth histogram
1
10,513,339
2
1,352,764
3
145,672
4
19,110
5
1,123
6
54
7
2
k4 top
none
carter dates
periodic keys of period at most 10: covered exactly by 0080 and 0083
controls
10 shuffled coppers (seed 91), identical search

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