Experiment 166

“F6.12, F6.13, F6.15: the Czech EVA double triangle and Chiffre XIII's five-row mask with its period-5 table, and the YAR '1-2-3' orderings, then the table”

Three items, one run, a verdict each. F6.12 EVA (Czech exile, WWII, type TT): the text written into a triangle growing by two cells a row (ten rows under 19 columns for 97 letters, three cells empty), copied out by the 19-letter key's numbering, written into a second such triangle and copied out again under the same numbering; keys: 19-letter windows of the K1-K3 plaintexts, 19-letter dictionary words, pairs of puzzle words totalling 19; then a repeating key 1-26 on the standard or KRYPTOS table. F6.13 Chiffre XIII (type TS): five rows as wide as the key (24-31 letters: K1-K3 windows and puzzle phrases), rows 1, 3, 5 full, row 2 left-aligned and row 4 at every cyclic shift, their lengths every split of what is left of 97, columns read in key order; then the published substitution: plain A-Z over one unknown mixed alphabet, rows shifted by one and used in turn (position mod 5), either shift direction, decided exactly (the unknown alphabet's cells are forced one by one). F6.15: Sanborn's reported '1-2-3' spacing of the raised YAR as orderings: letters taken by counting 1, 2, 3 places in turn around the remaining text (every order of 1, 2, 3), and the same with 4, 5, 7 (Y, A, R's places in ENDYAHR), then the table. Ten scrambled coppers.

revertedstatus
experiment 164parents
2 mintime, all phases
0candidates
0/24known letters
unclocked
1

The experiment method

What the method does with the 24 known letters.

Two ciphers of the Czech exiles in London: EVA wrote the messagethe message The 97 letters of English that Jim Sanborn wrote and enciphered. into a triangle twice over, and Chiffre XIII wrote it into five rows of a keyword's width, two of them short, then enciphered it with five shifted rows of one scrambled alphabet used in turn. Both are hand ciphers keyed by a sentence. And a third idea: Sanborn's raised letters YAR, reportedly spaced '1-2-3', as a way of counting off letters. Each is tested against the 24 known lettersthe known letters Sanborn has confirmed that positions 22 to 34 of the message read EASTNORTHEAST and positions 64 to 74 read BERLINCLOCK. and scrambled copies.

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 ACCENTEVA's two triangles (3,840 routes), move first: does a key of this length fit K4?1102026no setting fitssome dono length fits K4 or any copyEVA's two triangles (3,840 routes), key first: does a key of this length fit K4?110202625: K4 1, copies 0-4; 26: K4 1,123, copies 1,072-1,375YAR's '1-2-3' as counting-off orders (24 routes), move first: does a key of this length fit K4?1102026no length fits K4 or any copyYAR's '1-2-3' as counting-off orders (24 routes), key first: does a key of this length fit K4?110202626: K4 10, copies 4-10Chiffre XIII, 2,248,512 five-row masks then the shifted alphabet: does any mask pass three checks?K410K4 and the ten copies: 0 masks eachOUT: NO READING; EVA, CHIFFRE XIII AND THE YAR ORDERS FIT NEITHER K4 NOR ITS COPIESEASTNORTHEASTBERLINCLOCK
the idea

The Czech exiles in London used EVA (a message written into a triangle twice) and Chiffre XIII (five rows of a keyword's width, then five shifted rows of one scrambled alphabet in turn). Sanborn reportedly spaced YAR '1-2-3'.

the check

Every key from the earlier messages' text and the dictionary, every open choice of XIII's short rows enumerated, and the YAR spacing as counting-off orders, each against the known letters and scrambled copies.

the bound

Nothing fits better than on the copies; under Chiffre XIII's substitution no mask fits K4 or any copy at all. 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. EVA: write the 97 letters into rows of 1, 3, 5, ... cells, centred under 19 columns; copy out by the key's numbering; write into a second triangle; copy out again.
  2. XIII: write into five rows of the key's width, rows 2 and 4 short (row 4 shifted), copy out by the key's numbering; then encipher letter 1 with row 1 of the shifted alphabet, letter 2 with row 2, ..., letter 6 with row 1 again.
  3. Each known letter then forces one cell of the unknown alphabet; two known letters forcing different letters into one cell, or one letter into two cells, rule the setting out.
  4. YAR: count 1, 2, 3, 1, 2, 3 around the text, taking each letter counted to, until all are taken.
  5. Repeat for every key and choice, and on scrambled copies.
the worksheet: the first steps by hand, on squared paper
CHIFFRE XIII'S SUBSTITUTION: PLAIN A-Z OVER FIVE ROWS OF ONE UNKNOWN ALPHABET, EACH ROW SHIFTED ONE PLACE, USED IN TURNtransposed placerow usedplain lettercell of the unknown alphabet11EE + 022AA + 133SS + 244TT + 355NN + 461OO + 0Each known letter, wherever the maskputs it, names one cell of the unknownalphabet and the carved letter fills it.Two letters for one cell, or one letterin two cells, rule the mask out.
4

Worked example, generated

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

triangle: ten rows of 1, 3, ..., 19 cells, centred; key KRYPTOS repeated to 19 letters and numbered; columns read in that order

first 20 message places read: 82 5 11 19 29 41 55 71 89 36 48 62 78 96 17 27 39 53 69 87

place  letter  lands at  carved
   22       E        35       O
   23       A        77       D
   24       S        22       F
   64       B        96       A
   65       E        42       Q
   74       K        39       Q
5

Hypothesis, for a cryptanalyst

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

C1(a) ('double matrixes'), C22(d) (YAR's '1-2-3' spacing, clue register), C5; sources: foreign_agent_hand_systems.md section 4 (scz.bplaced.net ยง10.1: EVA, Chiffre XIII; the XIII substitution checked there on the published example), reports/Scheidt patents and field ciphers.md. Where the sources leave a rule open (XIII's partial-row lengths and row 4's shift; EVA's blank cells) every choice is enumerated; this is stated as a parametric family, not the documented one. Periodic-key items: orders A and B, 8 classes, 0134's statistic. XIII: a route is consistent if the forced cells M[(p + r) mod 26] = c (or p - r) are one-to-one; statistic: consistent routes with at least 4 checks, K4 against the coppers' range.

6

Prediction

Written before the run; the result decides it.

EVA and YAR: nothing below key length 20 beyond the coppers' rate. XIII: consistent routes at the coppers' rate. Zero candidates.

7

Method

  1. Keys: K1-K3 windows of 19 and 24-31 letters, 19-letter dictionary words, puzzle-word pairs.
  2. EVA and YAR routes through the compiled periodic-key test (orders A, B, 8 classes).
  3. XIII: each mask route; the 24 forced cells of the unknown alphabet checked for one-to-one, both shift directions; counts by checks.
  4. K4 against the coppers per item.
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

EVA: 3,840 routes, shortest live A None B 25; YAR: 24 routes, shortest live A None B 26; XIII: 2,248,512 masks, consistent with >=4 checks K4 0 (coppers 0-0)

What was searched, as the harness recorded it.

family
EVA double triangle and YAR orderings then or before a periodic key; Chiffre XIII five-row masks then its period-5 shifted unknown alphabet
eva keys
1,920
xiii keys
6,296
items
F6.12 EVA
routes
3,840
k4 live by period
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1,123
controls live by period min
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1,072
controls live by period max
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4, 1,375
k4 shortest live
A
None
B
25
coppers shortest live
A
99
B
25
k4 settings live below 20
A
0
B
0
controls settings live below 20
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 live below 20
none
cells k4 above every copper
none
null cells above rest per copper
0, 1, 0, 0, 0, 0, 0, 0, 1, 0
stop queue
no
F6.15 YAR
routes
24
k4 live by period
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10
controls live by period min
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 4
controls live by period max
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 10
k4 shortest live
A
None
B
26
coppers shortest live
A
99
B
26
k4 settings live below 20
A
0
B
0
controls settings live below 20
A
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
B
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 live below 20
none
cells k4 above every copper
none
null cells above rest per copper
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
stop queue
no
F6.13 Chiffre XIII
routes
2,248,512
keys
6,296
consistent ge
3
k4
0
coppers
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
4
k4
0
coppers
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
5
k4
0
coppers
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
6
k4
0
coppers
0, 0, 0, 0, 0, 0, 0, 0, 0, 0
k4 top
none
seconds
36
controls
10 shuffled coppers (seed 1612)
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