Zodiackillersite.com

I understand what you are saying about 55 plines ( parallel lines ). We worked on that a long time ago. You made a picture showing the start points of the 55 plines, and asked me to make a message. You were able to recover a few contextual words from the plaintext, I was impressed, you didn't seem impressed, then we moved on to something else.

I am glad that we covered so much territory, but only at minimal depth in a lot of areas. It is funny, I find myself returning to ideas previously worked on.

Are you thinking about hill climbing a reorganization of the plines? What about the idea where either a special symbol or the end of a row tells the program to look up scores for 1 grams, then 2 grams, then 3 grams, etc. moving away from the special symbol or end of row. Like at the end of the message. The program has a one dimensional array 0-339 that holds the message, and another array 0-339 that holds the scores? What about another array in between where each cell holds a value that designates what n gram table to look up the score in. Find some versatile way to use that, maybe that would have more uses than hill climbing a reorganization of fragments. But I don't know exactly what you do...

The scrolling test was with only two messages. If you want, I could perform more thorough scrolling / similar experiments and delve into your transposition scheme a little more before you do anything. Let me know.

I am still working on the 340 over here a little bit at a time. Sometimes I don't know what to do. It is very hard.

I tried a simple multiple transcription idea, something that I have been thinking about for quite a while. Two 17 x 10 transcription rectangles, top half and bottom half, vertical transcription, periods 19 and 21. Expanded the + and tried all untransposed possibilities. No solve.

smokie treats wrote:I understand what you are saying about 55 plines ( parallel lines ). We worked on that a long time ago. You made a picture showing the start points of the 55 plines, and asked me to make a message. You were able to recover a few contextual words from the plaintext, I was impressed, you didn't seem impressed, then we moved on to something else.

Yes, I remember.

smokie treats wrote:I am glad that we covered so much territory, but only at minimal depth in a lot of areas. It is funny, I find myself returning to ideas previously worked on.

Indeed.

smokie treats wrote:Are you thinking about hill climbing a reorganization of the plines?

Yes.

smokie treats wrote:What about the idea where either a special symbol or the end of a row tells the program to look up scores for 1 grams, then 2 grams, then 3 grams, etc.

It is on my to do list, I briefly looked at it and it will require some optimization and thought.

I decided to merge the pline rearrangement with my transposition solver (somehow).

My current thoughts are that the correlation between the 340 and the magic square reading rule is significant. And the period operation we have been using (columnar transposition) is not very able to get a solve on different pline orders.

A few questions,

Would any kind of misalignment within a pline order such as the magic square reading rule (go 2 squares left and 1 up) be more likely to cause a period shift from 38 to 39, or 30 to 29? In other words, is the periodic bigram peak at 39 caused by the pivots or has the period shift caused the pivots. The chicken or the egg.

And would any random pline order create a bigram peak at period column 2 such as the 340 and the magic square reading rule inspired matrix that I made? Or is it only that periodic pline orders cause these?

Those are some good questions. Let's say I write the message at an angle, over two down one. When I get to column 17 and want to continue to the left side of the message, I move 19 spaces, which skips row 10 ( top picture ). That's period 19 transposition and creates no misalignments.

Here is you matrix ( middle picture ). If I go right two spaces, but stay in row 9, then down one, I really only move two spaces. Period 2 becomes period 21, and period 3 becomes period 40. So that seems to make the matrix less viable.

But what if I start a new parallel line on row ten ( bottom )? . Period 1 becomes period 1, period 2 becomes period 20, and period 3 would become period 39. Maybe the pivots were created by some type of improper diagonal transposition, but the message was re-drafted into a different shape thus putting the pivots in the center of the message.

The only thing is, there are fewer period 3 repeats to work with even before homophonic diffusion.

smokie treats wrote:What about the idea where either a special symbol or the end of a row tells the program to look up scores for 1 grams, then 2 grams, then 3 grams, etc.

It is in the works. The row solver currently uses 3,4,5 and 6-grams. The smaller ngram sizes are used to capture the information at the end of each row.

So I tried it on a 340 character piece of the 408 with a period 19 transposition and then extracted the 55 plines. It solved but the balance is very tedious with 55 plines.

Code: Select all

solution:

Adjusted multiplicity: 0.28219
Score: 16671.52 Ioc: 0.06533

E
F
NK
TH
MAN
TUE
ILLS
ETHE
EXPER
ETTER
OURROC
RLTHEB
IATHAEW
LBEREBO
CEANDALL
LLEDWILL
LAVESIWIL
OUMYNAME
OUWILLTRY
INGPEOPL
TISSOMUCH
REFUNTHA
ILDGAMEIN
TBECAUSE
OATDANGER
OFALLTOK
INGGIVESM
HRILLING
ITISEVENB
GETTINGY
FFWITHAGI
PARTOFIT
NIDIEIWIL
INPARADI
HEIHAVEKI
ECOMEMYS
LNOTGIVEY
BECAUSEY
ILIKEKILL
BECAUSEI
UNITIAMO
ILLINGW
EFORRES
ISTHEM
ANAMAL
OMETH
MOATT
ENCE
THAN
KSO
EST
HE
RN
T
B

cipher:

W
J
(/
LM
qGD
IYE
P#B@
pI)W
+j=6\
WI5+t
TYr\de
t%L)NV
98LMlNA
BVW\+VT
e67DzG%%
%%ZfAP#B
#8c+@9A9B
dYq_^SqW
TYA9%#Lt_
kOR=pX=B
HP@K!qYeM
tNQYD5)S
U%fRlqEk^
HVWe8Y@+
X85zS(RNt
TQS#BLd/
U^RRkcZKq
Mr9#BPDR
UHkFZcpOV
R6HI9DR_
JQAP5M8RU
=GrI!Jk5
(PzUpkA9#
P^=SrlfU
MNk)ScE/9
peXqWq_F
%OT5RUc+_
VZeGYKE_
9%P/Z/UB%
V+eGYF69
Y^UIk7qT
9#BPORA
ZJdr\pF
9KI)6q
lO8qGB
XqEHM
q!85L
N(eE
L)l^
/@X
EKH
)Z
tO
I
V


I just checked p15 and 19 for the 340 and the best result came from p15 but does not seem to be a solve comparing it to 16671 score of the 408 piece. Double checking the solution would not be a bad idea though. Anyway, this row solver was a really great idea smokie. I am sure that it will be useful in many ways, do you want to be credited in any way?

Code: Select all

solution:

Adjusted multiplicity: 0.32923
Score: 14863.03 Ioc: 0.05941

F
U
ID
BC
DMH
VAH
OPYU
ODFM
LADSW
IRDTO
OWITHI
OWRAYW
NAINTUO
EFILLED
SIMISSYK
NETAGODD
HTTOOLINM
NEOFFITH
EWOMANYPE
RAREEACR
ENYAMEREA
NTOUSESM
LTSSTEPTS
KOWASABE
SSOFUNNYS
OMMSANTO
NTCITEDST
GHTHATIT
FUNOWHERT
ITTOBEHE
ITINGSIGU
PMRAGEII
FOGOODMOD
YWHENITD
ESSTHECON
TOFUNDED
MERGLORIE
PLCKNOWL
UKOHASBET
PEWITHAS
YSAMPLEM
IVELAAO
ONYPLAC
ELVESO
RDDREA
ETYRI
OBLAY
ERPL
SDRO
TFN
TOE
SH
OW
T
E

cipher:

d
2
G)
Tf
LKJ
1*H
|<z2
k.#K
PYLR/
VW)+k
^D(+4(
l%WO&D
pOGp+2|
>#Z3P>L
R(UVFFz9
EB+*5k.L
H+M8|CV@K
pclddG+4
N:^j*Xz6-
y7t-cYAy
BpzOUNyBO
p+l2_cFK
SMF;+B<MF
9^%OF7TB
_Rq#2pb&R
8KjROp+8
p+fZ+B.;+
5J+JYM(+
#2b^D4ct+
G++|TB4-
(MVE5FV52
<Ut*5cZG
dl5||.UqL
z/JNbVM)
-RR+4>f|p
+l#2E.B)
Ucy5C^W(c
<Sf9pl/C
29^4OFT-+
6N:(+H*;
zF*K<SBK
G1BCOO|
lXz6PYA
B31c_8
y.LWBO
c+ztZ
|TC7z
cW<S
R)Wk
+dp
+kN
FH
|D
M
>


This is really great, Jarlve. I am so happy that you did that. Even though the solve above isn't perfect, it is still a pretty good solve. You can read a lot of it. I say when you are ready we test it on various different types of transpositions, a variety of messages, plines and transpositions with various disturbances. Compare with untranspositions and the regular solver, solve independent chunks of messages. Before attacking the 340. Thanks.

We work together. I want to solve the 340, have a few beers together and high five. If not, then move on to retirement or semi retirement in peace. You can credit me with the idea if you want to.

Here is smokie39 again. Inscription rectangle 34 columns x 10 rows LRTB, reading TBLR making it period 10, transcription is LRTB into a 17 x 20, except that rows 3 and 18 are transcribed RL. The 7 symbol is a polyphone. You should be able to expand the 7, then untranspose rows 4-17 at period 10, use the row solver and get a solve.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
12 18 19 20 15 21 22 23 24 25 1 26 27 10 17 28 25
29 7 30 4 31 32 33 34 35 20 36 13 20 37 38 39 6
40 2 27 41 12 7 11 5 12 33 9 14 41 42 43 40 25
23 15 8 44 45 41 36 16 40 46 47 48 49 1 4 17 18
24 19 50 21 51 7 52 37 28 6 44 27 34 53 54 41 20
33 55 54 7 50 7 14 13 38 2 7 56 53 9 23 57 58
35 5 52 30 59 39 8 60 20 24 2 42 7 61 11 14 48
21 2 40 7 11 44 15 27 23 2 28 32 29 53 38 49 16
37 31 7 39 44 14 54 28 6 25 13 37 55 15 14 18 32
19 41 57 56 8 62 7 26 27 10 37 49 32 33 35 40 7
7 12 24 27 6 63 20 47 14 57 49 25 30 2 3 6 38
48 35 27 7 15 43 53 13 37 49 25 15 23 37 16 32 42
38 34 7 45 20 63 52 53 58 25 54 4 30 55 42 56 16
59 28 60 61 17 7 21 7 3 20 42 35 7 26 6 12 5
61 9 14 7 8 24 20 51 43 19 4 30 4 25 27 39 63
63 15 15 36 56 21 62 34 55 14 16 42 35 7 18 20 7
44 50 28 25 15 40 41 30 2 38 5 1 38 21 34 7 37
44 32 55 19 41 36 7 57 59 41 11 56 14 20 60 62 32
7 22 27 25 3 23 61 36 28 43 43 1 30 60 40 61 38