Yes, period 19. I apologize for being slightly ambiguous about the purpose for smokie16a. I am going to work on smokie16b soon. I want to use the row / column shuffle analysis to compare messages that have high count period 19 repeats both transposed and not transposed. The purpose is to find evidence of what the 340 really is.
I have a fairly demanding work schedule right now and it is getting on in the week. I hope to take a look at the column re-arrangement soon, as well as show heatmaps for smokie12 and color my bigram repeats. And update my shuffle spreadsheet maybe to analyse different portions of messages that are parts of contiguous rows and columns. And probably some other stuff that I can’t think of right now.
I’ll ask a question now. Right shifting the columns/rows by 1 causes radical and strange changes/shifts in the 340’s bigrams at various periods, is that explicable?
Basicly, performing the operation causes period 15 dominance to shift toward period 19 while increasing the bigram and trigram count even further. What sort of transposition could cause such response?
Very interesting. I have no idea what could account for that. I ran my repeating fragment detector on them to compare the patterns. It detects repeating ngrams and repeating fragments, and uses a probability score similar to smokie’s to rank the results. Here are the top 30 patterns for each:
Horizontal flip -> Period 15 Right shift by 1 -> Period 19 ====================================================== ================================================= Length Probability Count Pattern Positions Length Probability Count Pattern Positions 5 3.225456187417913E-14 2 9^?OF [189, 198] 6 2.260051480803209E-17 2 29^?OF [236, 315] 2 6.211985937661232E-14 5 p+ [111, 142, 241, 258, 277] 7 2.351357560627659E-13 2 V??+?#2 [24, 144] 8 1.6328871948803185E-13 2 )+k????2 [134, 209] 3 1.8876175972816484E-12 2 Xz6 [246, 328] 8 7.903174023220746E-13 2 +k????2p [135, 210] 2 4.4607284962504756E-12 4 (+ [16, 128, 173, 301] 3 1.8876175972816484E-12 2 Xz6 [76, 290] 8 4.644656909881795E-12 2 +????+|T [129, 169] 7 7.550470389126594E-12 2 P?L???^ [184, 300] 2 8.94870564637226E-12 3 <S [160, 284, 332] 2 8.94870564637226E-12 3 <S [67, 174, 248] 6 1.027702914075564E-11 2 P?L??5 [76, 162] 3 1.0114549025214595E-11 3 )?k [89, 134, 209] 2 2.7200968727031756E-11 4 p+ [93, 184, 207, 272] 6 1.98245160894206E-11 2 M??)?k [86, 206] 2 3.020188155650637E-11 3 29 [95, 236, 315] 2 5.898804991505152E-11 3 #2 [129, 215, 303] 2 5.898804991505152E-11 3 #2 [29, 149, 222] 8 1.1418921267506267E-10 2 l??|???L [57, 179] 3 1.0193135025320899E-10 3 2?^ [150, 236, 315] 5 8.615977441638801E-10 3 |???p [110, 138, 273] 6 2.5692572851889094E-10 2 K2???O [235, 287] 5 9.47757518580268E-10 2 |p??p [110, 273] 3 2.923243844481604E-10 2 (+H [16, 301] 6 1.6241900748165652E-9 2 p+???* [111, 241] 5 4.719043993204121E-10 3 2???O [236, 288, 315] 2 1.9329204196164076E-9 3 +4 [54, 106, 233] 8 5.965803764248174E-10 2 +?????-R [38, 133] 2 3.0694060367056848E-9 3 (+ [224, 232, 287] 4 6.57729865008361E-10 2 ^?(+ [14, 126] 2 3.0694060367056848E-9 3 +l [127, 143, 288] 2 1.9329204196164076E-9 3 G+ [6, 104, 132] 6 4.511639096712681E-9 2 |???p+ [138, 273] 2 1.9329204196164076E-9 3 +4 [7, 47, 129] 8 4.511639096712681E-9 2 O??p???+ [255, 271] 4 1.9329204196164076E-9 3 V??+ [24, 120, 144] 2 4.7892146885214484E-9 2 YA [100, 294] 4 3.0694060367056848E-9 3 +??M [17, 169, 258] 3 6.523606416205375E-9 3 2?+ [17, 23, 141] 8 3.0694060367056848E-9 3 5??????+ [12, 40, 167] 2 7.483147950814765E-9 2 N: [71, 222] 8 4.581737290942597E-9 3 +??????R [38, 47, 133] 6 8.948705646372261E-9 3 F????+ [14, 322, 323] 2 4.7892146885214484E-9 2 YA [250, 312] 2 1.0775733049173263E-8 2 PY [184, 293] 2 7.483147950814765E-9 2 N: [299, 323] 8 1.191072721532148E-8 3 +??????p [112, 135, 210] 7 8.948705646372261E-9 3 +?????| [129, 169, 192] 2 1.546336335693126E-8 3 +B [169, 262, 311] 2 1.0775733049173263E-8 2 PY [162, 249] 5 1.6837082889333224E-8 2 _???U [146, 316] 5 1.191072721532148E-8 3 p???+ [90, 125, 184] 6 1.6837082889333224E-8 2 E????k [1, 131] 2 1.546336335693126E-8 3 +B [188, 227, 258] 7 2.4245399360639836E-8 2 j?????z [74, 238] 7 2.993259180325906E-8 2 N?????D [65, 107] 6 2.993259180325906E-8 2 9????T [198, 325] 4 3.300068246309311E-8 2 _??K [232, 276]
Original 340 for reference:
Length Probability Count Pattern Positions 7 4.6446569098817964E-12 2 +???FBc [141, 210] 5 1.1279097741781705E-11 2 J??p7 [50, 160] 6 2.5893245504549364E-11 2 #O???Y [22, 125] 5 4.1108116563022564E-11 2 5?4?. [195, 227] 7 4.511639096712682E-11 2 H???p?^ [0, 49] 5 5.825980238523606E-11 2 O?*?C [91, 317] 2 1.0193135025320899E-10 3 G2 [14, 120, 270] 7 1.9581766912815457E-10 2 Np????O [17, 327] 6 2.5692572851889094E-10 2 M??2?c [184, 268] 3 3.171922574307296E-10 2 |5F [76, 212] 8 3.355764617389598E-10 2 +?????)L [39, 281] 3 4.719043993204121E-10 3 2?c [109, 187, 271] 7 5.965803764248174E-10 2 FB????R [166, 214] 6 8.615977441638801E-10 3 p????O [18, 181, 328] 6 1.0872677360342291E-9 2 G2???+ [14, 270] 2 1.1185882057965327E-9 3 FB [145, 166, 214] 4 1.1185882057965327E-9 3 B??O [20, 260, 314] 5 1.9329204196164076E-9 3 V???+ [59, 157, 233] 8 1.9329204196164076E-9 3 +??????L [39, 127, 281] 6 1.932920419616408E-9 2 4??+?B [197, 234] 2 3.0694060367056848E-9 3 M+ [38, 70, 138] 6 6.061349840159959E-9 2 P????/ [79, 297] 5 6.523606416205375E-9 3 2???+ [15, 67, 271] 5 6.523606416205375E-9 3 +???z [127, 200, 290] 7 8.948705646372261E-9 3 +?????c [141, 210, 290] 8 8.948705646372261E-9 3 F??????+ [97, 203, 282] 8 1.0523677840133776E-8 2 +?????M+ [64, 132] 3 1.9156858754085794E-8 2 >?D [98, 323] 2 2.993259180325906E-8 2 UZ [40, 250] 3 2.993259180325906E-8 2 C?> [246, 321]
Another step.. Rotate each column so it comes back on the bottom and joins up. Not sure what I’m achieving but it is still interesting my aim is to find a way that the 19 bigrams stay together in a simplistic approach.
the other thing I can do with this is slot the even rows between the odd rows and they work back on themselves in a normal manner. I will do this tonight if im not to busy.. Its T.G.I.F. for me
Can someone rotate it to the vertical
edit.. I have got a bit wrong but you should get the picture I will work on it later. Actually got a lot wrong in real time cut and paste(lol) see my next post that puts it in perspective.
UPDATE (by doranchak): Rotated the attachment
I’m done with the analysis for now. I’ll cut it down. In my test ciphers a large section of increased bigram response can be found around the actual randomized row. Some examples, look for the last number in each row, it’s the percentual difference from the average where higher is more indicative of a random fragment.
From my 1st row cipher: 170: 170 to 186: 47 110.2 171: 171 to 187: 47 110.5 172: 172 to 188: 48 112.5 173: 173 to 189: 47 111.7 174: 174 to 190: 47 111.7 175: 175 to 191: 47 111.7 176: 176 to 192: 49 115.1 177: 177 to 193: 49 115.8 178: 178 to 194: 49 115.9 179: 179 to 195: 49 115.9 180: 180 to 196: 48 114.4 181: 181 to 197: 48 114.4 182: 182 to 198: 48 113.4 183: 183 to 199: 48 113.0 184: 184 to 200: 48 113.5 185: 185 to 201: 48 113.1 186: 186 to 202: 47 112.0 187: 187 to 203: 49 115.4 188: 188 to 204: 48 114.1 189: 189 to 205: 49 114.6 190: 190 to 206: 48 114.3 191: 191 to 207: 48 114.0 192: 192 to 208: 48 113.9 193: 193 to 209: 48 114.0 194: 194 to 210: 47 111.8 195: 195 to 211: 48 112.7 196: 196 to 212: 48 113.3 197: 197 to 213: 48 112.7 198: 198 to 214: 48 112.9 199: 199 to 215: 48 113.5 From my 2nd row cipher: 265: 265 to 281: 39 114.7 266: 266 to 282: 38 114.1 267: 267 to 283: 37 110.9 268: 268 to 284: 38 114.3 269: 269 to 285: 38 113.4 270: 270 to 286: 38 114.2 271: 271 to 287: 38 114.2 272: 272 to 288: 38 114.0 273: 273 to 289: 39 117.3 274: 274 to 290: 39 116.8 275: 275 to 291: 39 117.1 276: 276 to 292: 40 118.3 277: 277 to 293: 40 118.3 278: 278 to 294: 40 117.4 279: 279 to 295: 40 117.7 280: 280 to 296: 40 117.7 281: 281 to 297: 40 119.4 282: 282 to 298: 39 116.7 283: 283 to 299: 39 116.2 284: 284 to 300: 40 119.2 285: 285 to 301: 39 117.2 286: 286 to 302: 40 117.4 287: 287 to 303: 40 118.4 288: 288 to 304: 40 118.6 289: 289 to 305: 40 118.6 290: 290 to 306: 40 118.7 291: 291 to 307: 40 118.8 292: 292 to 308: 39 116.0 293: 293 to 309: 39 115.8 294: 294 to 310: 39 115.8 295: 295 to 311: 39 117.0 296: 296 to 312: 38 113.9 297: 297 to 313: 37 110.1 From the 340, longest contiguous 110+ section: 31: 31 to 47: 44 110.4 32: 32 to 48: 45 111.2 33: 33 to 49: 45 111.1 34: 34 to 50: 45 111.8 35: 35 to 51: 45 112.3 36: 36 to 52: 45 111.4
No such long contiguous "high scoring" sections exist in both the 340 and the smokie16a for period 1 or 19. So it does not seem that any row could be strongly identified as random in the 340 with the test.
This is a good visual for direction that puts period 19 bigrams together.. They all fit using this angle odds and evens .. Not columns or rows its angled..is it perfect? so far they all align on the same plane.
Jarlve can you run that angle through. and disregard my odds evens top left a few wrong .
UPDATE (by doranchak): Rotated the attachment.
.
Jarlve, I am still working on smokie 16a-c.
Thanks for your work so far. I will draw some row shuffle comparisons between ciphers which are and are not transpositions and which have different numbers of symbols. Preliminary findings are that the number of symbols in a message are a factor in whether random shuffling of a row or column result in higher counts of period 19 repeats.
Randomly shuffling a row or column is basically the same thing with a transposition cipher because it disrupts the period 19 diagonal rows.
With fewer symbols in a message, a random shuffle of row or column creates more repeats partly because there is a lower count of different symbols adjacent to the row or column that is shuffled. With fewer symbols, there will be more repeats created.
With more symbols in a message, a random shuffle of row or column creates fewer new repeats partly because there are a higher count of different symbols adjacent to the row or column that is shuffled. With more symbols, there will be more repeats created.
Still working on it.
.
@Mr lowe,
Thanks for yet another original interpretation, since you didn’t complete the whole image and specified any specific order I improvised. 38 bigrams, no solve. Welcome to the sea of transposition.
Resulting cipher: 2dG)TfLKJ1*H|<z2k .#KPYLR/VW)+k^D(+ 4(l%WO&DpOGp+2|># Z3P>LR(UVFFz9EB+* 5k.LH+M8|CV@Kpcld dG+4N:^j*Xz6-y7t- cYAyBpzOUNyBOp+l2 _cFKSMF;+B<MF9^%O F7TB_Rq#2pb&R8KjR Op+8p+fZ+B.;+5J+J YM(+#2b^D4ct+G++| TB4-(MVE5FV52<Ut* 5cZGdl5||.UqLz/JN bVM)-RR+4>f|p+l#2 E.B)Ucy5C^W(c<Sf9 pl/C29^4OFT-+6N:( +H*;zF*K<SBKG1BCO O|lXz6PYAB31c_8y. LWBOc+ztZ|TC7zcW< SR)Wk+dp+kNFH|D>M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 16 17 18 8 19 20 7 21 22 23 24 4 25 16 26 27 28 25 29 28 30 31 24 32 33 27 34 32 3 34 25 1 13 35 18 36 37 19 35 7 21 28 38 23 39 39 15 40 41 42 25 11 43 16 17 7 12 25 44 45 13 46 23 47 8 34 48 30 2 2 3 25 29 49 50 26 51 11 52 15 53 54 55 56 57 54 48 20 58 55 42 34 15 32 38 49 55 42 32 34 25 30 1 59 48 39 8 60 44 39 61 25 42 14 44 39 40 26 31 32 39 56 5 42 59 21 62 18 1 34 63 33 21 45 8 51 21 32 34 25 45 34 25 6 36 25 42 17 61 25 43 9 25 9 20 44 28 25 18 1 63 26 27 29 48 57 25 3 25 25 13 5 42 29 54 28 44 23 41 43 39 23 43 1 14 38 57 11 43 48 36 3 2 30 43 13 13 17 38 62 7 15 22 9 49 63 23 44 4 54 21 21 25 29 35 6 13 34 25 30 18 1 41 17 42 4 38 48 55 43 46 26 24 28 48 14 60 6 40 34 30 22 46 1 40 26 29 32 39 5 54 25 53 49 50 28 25 12 11 61 15 39 11 8 14 60 42 8 3 10 42 46 32 32 13 30 52 15 53 19 20 58 42 37 10 48 59 45 55 17 7 24 42 32 48 25 15 57 36 13 5 46 56 15 48 24 14 60 21 4 24 16 25 2 34 25 16 49 39 12 13 27 35 44 Transposition matrix: 73 65 57 50 43 37 31 26 21 17 13 10 7 5 3 1 2 90 82 74 66 58 51 44 38 32 27 22 18 14 11 8 6 4 107 99 91 83 75 67 59 52 45 39 33 28 23 19 15 12 9 124 116 108 100 92 84 76 68 60 53 46 40 34 29 24 20 16 141 133 125 117 109 101 93 85 77 69 61 54 47 41 35 30 25 158 150 142 134 126 118 110 102 94 86 78 70 62 55 48 42 36 175 167 159 151 143 135 127 119 111 103 95 87 79 71 63 56 49 192 184 176 168 160 152 144 136 128 120 112 104 96 88 80 72 64 209 201 193 185 177 169 161 153 145 137 129 121 113 105 97 89 81 226 218 210 202 194 186 178 170 162 154 146 138 130 122 114 106 98 243 235 227 219 211 203 195 187 179 171 163 155 147 139 131 123 115 260 252 244 236 228 220 212 204 196 188 180 172 164 156 148 140 132 277 269 261 253 245 237 229 221 213 205 197 189 181 173 165 157 149 292 285 278 270 262 254 246 238 230 222 214 206 198 190 182 174 166 305 299 293 286 279 271 263 255 247 239 231 223 215 207 199 191 183 316 311 306 300 294 287 280 272 264 256 248 240 232 224 216 208 200 325 321 317 312 307 301 295 288 281 273 265 257 249 241 233 225 217 332 329 326 322 318 313 308 302 296 289 282 274 266 258 250 242 234 337 335 333 330 327 323 319 314 309 303 297 290 283 275 267 259 251 339 340 338 336 334 331 328 324 320 315 310 304 298 291 284 276 268
Thanks for your work so far. I will draw some row shuffle comparisons between ciphers which are and are not transpositions and which have different numbers of symbols. Preliminary findings are that the number of symbols in a message are a factor in whether random shuffling of a row or column result in higher counts of period 19 repeats.
Yes, but it’s not just the number of symbols. The number of instances per symbol also matters. A good measure I found (glurk pointed out it’s Friedman Index of coincidence used in another way) is to sum the frequency of each symbol using Friedman’s "c*(c-1)". So in the 340 the "+" symbol is counted 24 times and this gives 24*(24-1)=552. Now sum these for all symbols in the cipher and you get the following numbers.
408 (capped 340): 2108
340: 2236
smokie16a: 3286
The resulting number is a score (higher is more) which seem to correlate with the number of repeating fragments you can expect in randomizations of these ciphers (probably not linear). By these numbers you can expect the 340 to have more repeating fragments than the 408 per character length even though it has 9 more symbols. Mainly thanks to the frequency of the "+" symbol.
That’s why I had to switch to using percentages to analyze the smokie16a for random fragments versus the other ciphers, and that works.
I’ve done the fragment randomization bigram response test for the 340 regarding columns.
Fragment randomization bigram response test for: 340.txt ----------------- Period: 19. Fragment size: 20. Step: 20. Top-to-bottom, left-to-right. ----------------- Fragment number, fragment range, average bigrams, distance average. ----------------- 1: 1 to 20: 37 94.0 2: 21 to 40: 41 103.0 3: 41 to 60: 36 91.7 4: 61 to 80: 40 100.1 5: 81 to 100: 43 108.2 6: 101 to 120: 43 107.1 7: 121 to 140: 42 104.6 8: 141 to 160: 39 96.8 9: 161 to 180: 36 91.1 10: 181 to 200: 41 104.0 11: 201 to 220: 35 86.9 12: 221 to 240: 41 102.6 13: 241 to 260: 38 95.0 14: 261 to 280: 36 89.7 15: 281 to 300: 44 110.3 16: 301 to 320: 38 96.0 17: 321 to 340: 47 118.0 <---
118 percent higher than average is quite high. I’ve ran the same on my own row ciphers and smokie16a and the highest recorded out of these 3 is 112. Although when stepping by one the 110+ section is only 6 broad and it did not correlate with period 1 numbers.
316: 316 to 335: 44 111.2 317: 317 to 336: 46 116.0 318: 318 to 337: 46 114.9 319: 319 to 338: 46 115.9 320: 320 to 339: 46 116.3 321: 321 to 340: 47 118.2
I think it is related to the column/row right shift by 1 bigram period shift/increase I showed earlier. Sigh…
I’ve created another column cipher (column 17 is random before encoding) to compare to the 340.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 2 19 20 21 22 23 24 25 26 27 28 29 30 31 32 9 14 33 21 5 34 35 36 18 8 37 1 10 24 35 38 17 27 25 4 13 35 8 35 12 39 40 6 41 29 26 11 16 20 28 23 42 34 21 36 17 37 7 32 43 1 25 15 33 38 14 30 4 44 23 39 45 42 29 9 46 3 12 31 24 44 5 26 43 6 14 40 10 47 41 34 7 38 13 36 30 1 20 9 47 27 8 17 28 15 4 11 6 16 32 41 48 26 48 33 19 34 36 1 37 31 23 46 4 24 40 11 27 6 39 5 49 10 49 13 16 15 22 2 37 7 44 18 20 34 29 23 47 45 12 30 8 17 36 1 32 45 33 45 19 4 22 48 21 38 6 34 36 18 19 27 3 25 31 23 46 35 39 22 48 1 5 44 46 40 24 35 18 7 43 47 44 30 3 38 39 28 11 10 21 25 49 41 4 19 13 22 48 9 6 42 8 28 26 43 46 17 31 44 2 27 15 24 2 7 20 32 49 18 33 9 16 5 30 10 12 31 13 20 3 19 37 32 21 45 24 25 22 8 43 14 34 33 2 36 5 50 40 41 11 29 15 45 47 1 38 16 42 35 39 30 26 17 18 37 42 40 41 14 7 10 38 4 47 9 15 19 23 28 11 39 3 22 29 13 20 27 32 16 2 18 19 6 33 49 5 23 12 14 21 31 24 10 41 15 35 2 47 8 22 18 45 [)H'S&%.7$9RGC=XA N)#KJ4UZ*,F@P:"+7 C1JSDQYN./[$ZQ5AF *'GQ.QR6T&WP,9XK@ U<DJYA/%+O[*=15C: '(U6M<P72HR"Z(S,O &CT$3WD%5GY:[K73F .A@='9&X+WI,I1#DY [/"U2'ZT9F&6SL$LG X=4)/%(NKDPU3MR:. AY[+M1M#'4IJ5&DYN #FH*"U2Q64I[S(2TZ QN%O3(:H56@9$J*LW '#G4I7&<.@,O2A"() F=Z)%K+LN17XS:$R" GKH#/+JMZ*4.OCD1) YS>TW9P=M3[5X<Q6: ,AN/<TWC%$5'37=#U @96H4PGKF+X)N#&1L SURCJ"Z$W=Q)3.4NM
Results:
Fragment randomization bigram response test for: jarlve_col2.txt ----------------- Period: 19. Fragment size: 20. Step: 20. Top-to-bottom, left-to-right. ----------------- Fragment number, fragment range, average bigrams, distance average. ----------------- 1: 1 to 20: 46 105.9 2: 21 to 40: 40 93.6 3: 41 to 60: 42 96.9 4: 61 to 80: 40 92.8 5: 81 to 100: 40 93.2 6: 101 to 120: 45 104.4 7: 121 to 140: 46 105.7 8: 141 to 160: 42 98.7 9: 161 to 180: 45 105.6 10: 181 to 200: 44 101.5 11: 201 to 220: 40 92.8 12: 221 to 240: 46 105.8 13: 241 to 260: 38 87.3 14: 261 to 280: 45 104.0 15: 281 to 300: 42 98.2 16: 301 to 320: 43 98.9 17: 321 to 340: 49 113.7 <---
Column 17 correctly identified.
306: 306 to 325: 48 112.8 307: 307 to 326: 48 113.0 308: 308 to 327: 48 112.9 309: 309 to 328: 48 112.7 310: 310 to 329: 47 111.0 311: 311 to 330: 48 111.3 312: 312 to 331: 47 110.9 313: 313 to 332: 48 113.3 314: 314 to 333: 47 110.6 315: 315 to 334: 47 111.1 316: 316 to 335: 47 111.2 317: 317 to 336: 49 113.9 318: 318 to 337: 50 116.8 319: 319 to 338: 49 114.3 320: 320 to 339: 49 114.6 321: 321 to 340: 49 114.6
110+ section is broad enough, taking into consideration that it is the last column.
Jarlve. The 19 bigram scenario is Odds / Evens on that bias that I posted up last. It carries all trigrams and bigams perfectly wrapping around and around candy cane style two rows together. not sure how to explain it.
Another strange observation, add one random column to the 340 at the end and suddenly bigram repeats period 5 pop up with counts around 40, tried various randomizations of the last column and they all had the same effect. I haven’t much time left but seems inconsistent with the magic square scheme and regular period schemes. Seems to be somewhat consistent with the transposition matrix I put up for Mr lowe…
340, 18 by 20 example: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 5 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 21 36 37 38 39 20 40 41 15 27 22 34 13 23 42 1 43 14 44 5 5 45 7 6 18 31 8 46 5 24 20 20 3 32 16 38 47 48 39 20 42 49 50 17 11 51 35 9 20 52 53 10 54 20 5 18 3 7 35 6 24 55 31 17 56 10 35 4 16 26 22 38 23 51 20 32 57 25 58 16 40 38 59 15 8 29 42 13 11 5 22 15 16 43 33 50 23 24 20 47 19 28 42 20 60 13 48 50 17 30 39 20 61 20 41 3 16 35 21 38 36 62 63 53 32 54 55 42 6 40 8 20 7 43 20 24 5 45 30 35 21 36 55 28 40 20 3 54 51 49 2 11 26 28 21 5 61 14 39 32 24 47 16 30 38 6 3 43 11 31 51 14 53 39 29 20 52 21 35 29 42 63 48 44 36 23 20 19 11 51 35 21 38 22 58 18 3 36 6 15 35 19 7 33 51 16 53 61 29 38 8 53 49 20 20 38 36 21 59 12 31 37 53 48 56 2 4 8 40 41 51 55 20 39 11 38 29 46 42 21 32 22 24 5 7 29 33 39 57 15 16 24 3 38 14 20 13 12 63 56 30 20 35 6 27 21 11 34 13 7 20 20 34 27 56 42 27 38 9 24 44 1 14 54 22 34 5 47 11 35 10 17 27 30 45 49 21 47 28 24 21 31 55 56 38 20 4 39 26 1 19 5 10 44 42 41 24 18 62 11 32 58 20 26 HER>pl^VPk|1LTG2d8 Np+B(#O%DWY.<*Kf)F By:cM+UZGW()L#zHJT Spp7^l8*V3pO++RK2c _9M+ztjd|5FP+&4k/+ p8R^FlO-*dCkF>2D(c #5+Kq%;2UcXGV.zL|p (G2Jfj#O+_NYz+@L9j d<M+b+ZR2FBcyA64K/ -zlUV+^J+Op7<FBy-Y U+R/5tE|DYBpbTMKO_ 2<clRJ|*5T4M.+&BF. z69Sy#+N|5FBc(;8Ry lGFN^f524b.cV4t++c yBX1*:49CE>VUZ5-+M |c.3zBK(Op^.fMqG2O RcT+L16C<+FlWB|)L^ ++)WCzWcPOSHT/()p_ |FkdW<7tB_YOB*-Cc+ >MDHNpkSzZO8A|K;+D Mr lowe's inspired transposition matrix cipher: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 8 22 23 24 25 26 27 28 29 30 31 32 32 33 34 12 9 35 5 18 36 37 13 6 15 33 38 7 39 3 10 20 40 41 42 21 43 1 2 24 14 44 9 45 40 19 11 28 30 17 27 36 21 4 6 37 16 46 32 20 8 47 42 29 1 22 26 39 45 31 18 24 47 40 14 30 12 13 33 42 20 38 15 48 43 22 2 23 3 16 39 19 28 45 34 10 17 18 49 24 29 44 11 8 1 12 25 4 5 14 25 5 26 47 41 2 35 7 39 34 13 19 31 25 50 28 23 27 37 35 36 38 15 40 16 20 6 29 44 42 43 3 7 1 4 32 10 23 45 46 22 26 34 37 20 43 44 14 19 28 8 40 42 47 12 35 21 7 13 29 15 11 17 46 37 4 43 50 30 18 26 5 16 50 31 41 37 3 24 45 27 34 8 23 12 44 1 9 2 4 35 46 14 10 13 30 17 15 26 43 22 34 9 18 47 24 19 16 36 25 8 32 44 9 12 38 41 22 4 28 47 32 39 21 7 2 31 50 29 3 13 22 1 30 14 15 27 23 38 36 40 19 21 33 16 10 28 5 17 31 29 27 11 37 1 26 20 14 35 42 9 25 5 50 18 45 49 38 3 8 24 36 19 12 46 10 13 6 28 21 29 49 1 7 15 40 14 19 47 41 43 38 37 49 34 17 46 49 33 38 30 49 44 46 31 9 41 27 36 ,/"=(YXHC?9ZG-OB 'AIDX5>4N6U!@&8TT E)9H%='12Z(-ELYM" CIP*<D7,/4GVH.PA? !&BU1D(2OKTIX:<@ ,56M.8'4:PG&9ZE<I L-Q75/>"OMA!.)CB' 34@V?X,9N=GN=6:* /%YM)ZA8N$!>U2%1L -POI(@V<7"Y,TC>. K56)2I7VGA!XP<:9% DYZ@-?BK27$&'6=O $8*2"4.U)X>9V,H/ %KGCZ&B-675)H':4A O1NXTVH9L*5!:TMD Y/8$@"Z5,&G-U>L1P ADEOC!=B8@U?2,6IG %<HN=$'.3L"X41A9K CZ(!D@3,Y-PGA:*7L 23)BK3EL&3VK8H*U1 Previous cipher but 18 by 20 to show period 5 (stronger mirrored): 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 18 21 8 22 23 24 25 26 27 28 29 30 31 32 32 33 34 35 12 9 36 5 19 37 38 13 6 15 34 39 7 40 3 17 10 18 41 33 42 21 43 1 2 24 14 44 9 45 41 20 11 7 28 30 17 27 37 21 4 6 38 16 46 32 18 8 47 42 29 7 1 22 26 40 45 31 19 24 47 41 14 30 12 13 34 42 18 24 39 15 48 43 22 2 23 3 16 40 20 28 45 35 10 17 19 22 49 24 29 44 11 8 1 12 25 4 5 14 25 5 26 47 33 42 2 36 7 40 35 13 20 31 25 50 28 23 27 38 36 37 39 19 15 41 16 18 6 29 44 42 43 3 7 1 4 32 10 23 45 40 46 22 26 35 38 18 43 44 14 20 28 8 41 42 47 12 36 30 21 7 13 29 15 11 17 46 38 4 43 50 30 19 26 5 16 2 50 31 33 38 3 24 45 27 35 8 23 12 44 1 9 2 4 16 36 46 14 10 13 30 17 15 26 43 22 35 9 19 47 24 20 21 16 37 25 8 32 44 9 12 39 33 22 4 28 47 32 40 21 2 7 2 31 50 29 3 13 22 1 30 14 15 27 23 39 37 41 8 20 21 34 16 10 28 5 17 31 29 27 11 38 1 26 18 14 32 36 42 9 25 5 50 19 45 49 39 3 8 24 37 20 12 46 30 10 13 6 28 21 29 49 1 7 15 41 14 20 47 33 43 39 12 38 49 35 17 46 49 34 39 30 49 44 46 31 9 33 27 37 26 ,/"=(YXHC?9ZG-OBI 'AIDX5>4N6U!@&8TT* E)9H%='12Z(-ELYM"B CIP*<D7,/4GVH.PA?Y !&BU1D(2OKTIX:<@Y ,56M.8'4:PG&9ZE<I4 L-Q75/>"OMA!.)CB'5 34@V?X,9N=GN=6:*< /%YM)ZA8N$!>U2%1L' -POI(@V<7"Y,TC>.M K56)2I7VGA!XP<:9%& DYZ@-?BK27$&'6=O/ $8*2"4.U)X>9V,H/O %KGCZ&B-675)H':4AD O1NXTVH9L*5!:TMD/ Y/8$@"Z5,&G-U>L1PX ADEOC!=B8@U?2,6IGT %<HN=$'.3L"X41A9K& CZ(!D@3,Y-PGA:*7L9 23)BK3EL&3VK8H*U16
Jarlve. The 19 bigram scenario is Odds / Evens on that bias that I posted up last. It carries all trigrams and bigams perfectly wrapping around and around candy cane style two rows together. not sure how to explain it.
Please make a genuine effort, it seems very interesting right now. I’ll be back later. Can you write numbers (transposition order) next to the symbols?
Another strange observation, add one random column to the 340 at the end and suddenly bigram repeats period 5 pop up with counts around 40, tried various randomizations of the last column and they all had the same effect. I haven’t much time left but seems inconsistent with the magic square scheme and regular period schemes.
Interesting. I did my own counts, treating the added column as nulls so they don’t accidentally make new ngrams, and got 35 repeating bigrams at period 5 and 34 at period 20. The extra column might be shifting most of the period 19 bigrams to period 20. Not sure why period 5 would start peaking though (by comparison, the original 340 already has 30 bigrams at period 5).
Jarlve. The 19 bigram scenario is Odds / Evens on that bias that I posted up last. It carries all trigrams and bigams perfectly wrapping around and around candy cane style two rows together. not sure how to explain it.
Please make a genuine effort, it seems very interesting right now. I’ll be back later. Can you write numbers (transposition order) next to the symbols?
I’ll spend some time on the weekend jarlve.. I don’t run a computer as such at home anymore and only run iPad so lots of programs don’t load. So it’s manual work but that is why I think I see what he has done, the visual aspect helped.
Think I’ll go get myself a good laptop..
