What if I get a score over 22222, but it isn’t really a solution? Should it count?
What if I get a score over 22222, but it isn’t really a solution? Should it count?
I have been noticing the same thing. Here is a new version with a new settings:
(Substitution + period + nulls & skips) Over : 22500
https://drive.google.com/open?id=1zQSmu … De7JuAct8c
Using 23000 just to be sure but 22500 should do also for smokie 5/5.
If I make 5 new symbols at the end, multiplicity and score will go up. I am sorry about smokie 5/5. Do you want to work with a new message? I will try the new solver. What about being able to enter the minimum solve score. That way we could solve an untransposed message, know what the real score is, then deduct whatever you judge appropriate, and enter.
If I make 5 new symbols at the end, multiplicity and score will go up. I am sorry about smokie 5/5. Do you want to work with a new message?
No, a cipher like smokie 5/5 is exactly what is needed to better the program. I am getting answers but it will take time.
What about being able to enter the minimum solve score.
What do you mean? Is that not what the "(Substitution + period + nulls & skips) Over : 22500" setting does?
Yes, I see that it was it does. Here is smokie 5/5 repaired, with the last five symbols what they should have been and map to the five letters, A B A S B, the ones you did not have in your solution.
5 62 26 40 27 41 28 36 49 25 53 1 47 60 48 23 29
13 47 42 14 20 61 37 7 35 60 9 43 26 18 50 15 2
62 53 32 33 35 9 63 40 3 21 41 63 17 4 49 19 38
27 63 42 52 51 61 1 6 27 29 13 50 43 51 26 57 53
52 40 50 21 47 39 14 36 27 5 58 15 56 53 16 11 59
24 32 28 2 57 41 61 50 17 37 40 43 54 59 25 22 13
55 14 3 12 34 61 56 35 4 53 43 54 55 15 58 10 16
17 29 34 23 47 11 38 13 2 51 14 33 26 18 32 2 3
56 27 28 39 15 4 62 5 41 28 30 29 42 31 53 54 62
36 60 56 12 56 44 21 9 53 56 16 55 39 17 10 38 28
24 24 9 12 1 23 39 61 47 52 49 37 28 13 54 53 37
14 15 38 43 16 17 30 15 40 2 14 23 45 29 48 39 15
24 15 34 43 20 58 34 8 17 16 6 54 51 51 40 14 21
15 3 33 16 9 62 47 4 20 17 26 7 27 13 46 55 13
43 8 22 10 52 32 38 5 56 49 6 22 37 11 1 53 2
58 23 38 31 28 54 24 33 5 29 62 15 40 40 53 16 35
22 48 26 17 56 53 2 18 42 13 57 21 50 39 34 58 27
57 63 12 14 43 56 55 5 32 22 33 41 60 15 56 23 41
56 16 54 55 54 28 59 63 57 47 15 13 62 14 32 47 33
45 51 53 57 1 54 1 61 45 54 15 20 2 6 3 52 5
EDIT: Score 25154
Sorry if this breaks up the momentum of this thread, but this post on Klaus’ site reminds me of the various transposition schemes we’ve talked about here:
http://scienceblogs.de/klausis-krypto-k … a-dog-tag/
If you follow the numbers around, they seem to progress in a periodic fashion:
22 47 16 41 10 35 04 05 23 48 17 42 11 29 30 06 24 49 18 36 12 13 31 07 25 43 19 37 38 14 32 01 26 44 20 21 39 08 33 02 27 45 46 15 40 09 34 03 28
Once a number sequence runs into itself, it jumps below two lines (and wraps around if necessary). Very curious dog tag.
That just got me working on z340 again. I’m totally psyched right now. Sorry when I brag: The first two comments in the link shown by David are from me since I recognized a magic square.
The link shown there by Norbert is extremely interesting. A connection of the Zodiac signs (planets) with own Magic Squares. I have researched a lot since I read Klaus’ blog post, but so far I have only loose ideas whether something could be in the thing. I really need to see that I can find some time to work on it, even though I actually wanted to stop with z340.
Jarlve was the first who found the connection between p19 and the magic square on the FBI-Files I think.
Translated with http://www.DeepL.com/Translator
That just got me working on z340 again.
Funny how little things like that can trigger enthusiasm. Happens to me too.
It looks like a transposition matrix. Interesting because it is made of brass and looks a little worn. There would have to be a companion.
http://radionerds.com/images/b/bb/TM_11-485.pdf
Here is the gridmaker web gadget showing the construction of the magic square: http://zodiackillerciphers.com/gridmaker-dogtag/
I made a little modification to my matrix spreadsheet so that I can look at null skip positions next to corrective skip and wildcard locations after 1000 iterations. Right now set to 5/5 and 15% shift div 2, but other settings do this as well. Some of the corrective skips and wildcards glue themselves together, cancelling each other out. Iterations moving one away from the other don’t seem to score better than keeping them glued together, and the spreadsheet is only really working on a fewer number of corrective skips and wildcards. I do not know if AZD is doing this or not.
I understand that a null and a skip in the same line cancel each other out. But what do you mean by "keeping them glued together"? You made me realize a good point, not allowing the hill climber to stack nulls and skips in the same line may turn out to be a very good optimization.
They seem to find each other and not break away from each other because later iterations of moving one does not give a better score. Sometimes they find each other in an area of the message that doesn’t have an actual null or skip and just sit there together cancelling each other out while the computer moves the other ones around instead. The spreadsheet puts them next to each other because you can’t have more than one item in a cell, but I don’t know if the program does or not, or can put them in the same position. I suppose the only way to deal with it is to move both at the same time where both make an improvement in the score. Or, make them move away from each other regardless of score. I have not confirmed this with the program, just wondering because I think maybe it could.
Here is the gridmaker web gadget showing the construction of the magic square:
The tool is very useful, thank you for sharing it! When I find the time, I will build it into peek-a-boo in a similar way and connect it directly to a 2d transposition.
Funny how little things like that can trigger enthusiasm. Happens to me too.
Yes, indeed. Unfortunately, just not a good timing for me. As always: Many ideas, little time. So that I don’t disturb this thread, I will open a new one soon (magic squares and 2D-Transposition).
Back to the topic:
I followed your work with zeroes/skips, but did not follow everything in detail. Sorry, if the answer to my question is already in the thread:
Have I understood correctly that you are also investigating how nulls/skips affect PC cycles? That it is also a possibility that the interrupted cycles in z340 are a consequence of adding nulls or skips? If yes: Does the "Merge Homophones" mode work similar in AzDecrypt? Let’s say you have an untransposed cipher (P1) that is not 100% cyclic. These interruptions could come from nulls or creative cycling. If you let merge symbols over a hillclimber and evaluate them according to the cycle-score, couldn’t you reduce the search space and "smoothen" the cipher? This could possibly lead to mistakes, but it would perhaps point in the right direction. Or am I barking up the wrong tree right now?
Translated with http://www.DeepL.com/Translator
They seem to find each other and not break away from each other because later iterations of moving one does not give a better score. Sometimes they find each other in an area of the message that doesn’t have an actual null or skip and just sit there together cancelling each other out while the computer moves the other ones around instead. The spreadsheet puts them next to each other because you can’t have more than one item in a cell, but I don’t know if the program does or not, or can put them in the same position. I suppose the only way to deal with it is to move both at the same time where both make an improvement in the score. Or, make them move away from each other regardless of score. I have not confirmed this with the program, just wondering because I think maybe it could.
EDIT: Possibly a more equal division with a lot of nulls and skips may cause this to occur more often, as compared with an unequal division of more nulls or more skips.
I followed your work with zeroes/skips, but did not follow everything in detail. Sorry, if the answer to my question is already in the thread:
Have I understood correctly that you are also investigating how nulls/skips affect PC cycles? That it is also a possibility that the interrupted cycles in z340 are a consequence of adding nulls or skips? If yes: Does the "Merge Homophones" mode work similar in AzDecrypt? Let’s say you have an untransposed cipher (P1) that is not 100% cyclic. These interruptions could come from nulls or creative cycling. If you let merge symbols over a hillclimber and evaluate them according to the cycle-score, couldn’t you reduce the search space and "smoothen" the cipher? This could possibly lead to mistakes, but it would perhaps point in the right direction. Or am I barking up the wrong tree right now?
Jarlve has added an option to his program to hill climb the positions of nulls and skips caused by maybe interrupted route transposition or transcription errors, or some combination of that, while at the same time hill climbing the homophonic key when the message is untransposed and the nulls and skips are corrected for. The idea is to keep testing different methods and, once we are confident in the null skip hill climber’s accuracy and efficiency, attack the 340.
