You will talk about the 340? It would be worthwhile to know the opinions of the ACA people on the 340 and perhaps our observations.
Yes, I will try to focus primarily on Z340. I’m eager to hear what they think. They spend a lot of time creating and solving all sorts of cryptograms in the magazine they publish every two months. The members seem very competitive, and track the scores of top solvers. Several have even written their own hillclimbers for cracking the codes. I’m hoping that giving them an overview of all of our observations will fuel their curiosity and motivation for exploring Z340.
Several ACA members also post their own homemade cryptograms in this Google Plus group: https://plus.google.com/u/0/communities … 1588941207
It is good, the 340 feels more than genuine enough to receive more attention.
Made this image for the main page, it looks neat.
Working on a new solver in the midst of things, substitution + sequential homophones. Use the sequential homophones aka cycles as leverage. I have something up and running and it works unbelievably well. It runs about 7 times slower due to the added cycle calculations though ciphers with cycles solve quicker and better. Here are a few examples of what it did so far:
Solve on a 5 row 408 with 6-grams, AZdecrypt’s previous best was 6 rows with 7-grams:
Score: 24955.60 Ioc: 0.05714 Ngrams: 55 PC-cycles: 461 ILIKEKILLINGPEOPL EBECAUSEITISSOMUC HFUNITISMANYOUHAV EAKIDLINGWILDGAME INTHEFORBESTBECAU
Solve on a 6 row 408 with 5-grams, a tie with AZdecrypt’s previous best which had to use 7-grams to accomplish this:
Score: 25098.68 Ioc: 0.05862 Ngrams: 156 PC-cycles: 865 IDIWEWILDINGPEOPL EBECAUSEITISSOMUC HFUNITISMANYOURTH ERWILLINGKIDSGAME INTHEFORBESTBECAU SEMARISTHEMOATVER
Solve on smokie’s purple haze cipher with 5-grams, the one with wildcards, previously impossible for AZdecrypt without expansion of said wildcards:
Score: 19226.57 Ioc: 0.05396 Ngrams: 643 PC-cycles: 7820 PULPLEHAVEALLINRY BLAIDLATELYTHINGS OMUTSEERTHESAREAS TINFUNNYBUTIMCUMK NOWWHYSSUSEREWHIL EIKISSSHESKYODLOL EHAVEADLAOCUSMMBN TOUMWIOIRSCRONGUP CLMCWNARIHAPPYBLI NOIDELYWOATESELIT ISTSATGILLPUTASPE LDDUREHESPREHESPR ECHNCCHYMAHPUDPLE HAVOALLINROEYESMO NDKUMWIFITSMAYCLN IGHTYCUSOGCTREBLB WINGBLMWISGRORIUM ISITDCDCSLCWBMJUS TTHEENMMFTIROHELP REOEAHPUOPLSHAOEX
Solve on jrob’s cipher with 5-grams, previously impossible for AZdecrypt without using the reddit 5-grams or 6-grams:
Score: 22785.22 Ioc: 0.06461 Ngrams: 2088 PC-cycles: 42290 ILIKEEATINGICEC REAMBECGUSEITIS SODELICUNIDISMO REDELICIOUSTHAN EATINGSHNRBETOF THEFROZENFOODAI SLEBECAUSEICECR EAMINTHEMOSTDEL ICIOUSGELATOOFA LLTONGDSOMETHIN GSWEETGIVESMETV EMONTPLEASIGNEX PERENCEITISHVEN MOREPLEASINGTHA FSLURPINGDOWNAW EGDYSNHAKETHEBE STPARDOIITISTHA TWHENIURINGTEAL LTVICECREAMIHAV HEATENWILLBEREB ORMASLEMONAUEIW ILLNOTGIVEYOUTH EGAMEOFMYGROCER BECGUSHYOUWILLS LOWDOWNORNDOPMY EATINGOFICECREA MHAANCZRTESTDIA
Solve on the fake Zodiac 169 cipher with 5-grams, previously impossible for AZdecrypt without 6-grams:
Score: 26152.14 Ioc: 0.05571 Ngrams: 1600 PC-cycles: 2948 SHAUNAIFYOUWI SHTOKNOWMYIDE NTITYYOUWILLH AVETOPUTSOMEE FFORTINTOITDO NOTFIDDLEANDF ARTAROUNDHERE ISACLUEASKMIS SUSTAYLORIFSH EREMEMBERSWHO DRAWSTHESYMBO LTHENYOUWILLH AVEMEGOODLUCK
Solve on a 5 row 408 with 6-grams, AZdecrypt’s previous best was 6 rows with 7-grams:
Wow! Nice job! That’s particularly impressive since the first 5 rows have a multiplicity of 0.588. More than half of the symbols are different!
Looks really great Jalve!! Great work! I wish that I had the knowledge and ability to use these great programs you all develop.
The Best Mystery Is An Unsolved Mystery….
Solve on a 5 row 408 with 6-grams, AZdecrypt’s previous best was 6 rows with 7-grams:
Wow! Nice job! That’s particularly impressive since the first 5 rows have a multiplicity of 0.588. More than half of the symbols are different!
Yeah I am loving it too.
Looks really great Jalve!! Great work! I wish that I had the knowledge and ability to use these great programs you all develop.
Thanks bmichelle. Basic AZdecrypt use is really simple, check out this 2 step how to:
If there are specific things you want to do with the program you can always ask me here.
SpiderBasic appears to be a really cool new thing – https://www.spiderbasic.com/ – that could allow to me to port AZdecrypt to the web. I saw a speed comparison on the FreeBASIC forum that showed that it generates extremely fast code, though that remains to be seen. I will be looking into it for sure.
SpiderBasic is new web client-side programming language based on established BASIC rules. Its allows development of very complex, windowed based web applications, including mobile app for iOS and Android. It provides a large commandset to handle complex and reactive GUI, 2D games, and many more in a coherent manner. Every library and commands behave the same and have similar syntax for easier and faster learning. The full documentation can be found here.
SpiderBasic is a compiler which generates optimized JavaScript, which needs a modern browser to run (at least HTML5 compliant). As a compiler, it has strong typing and various checks that JavaScript doesn’t provide, allowing robust code construction. SpiderBasic is not object oriented because we do believe that object oriented programs are hard to maintain, but we provide flexible namespace support and many other features.
SpiderBasic is based on the PureBasic syntax, and is compatible to some degree with it. If you are already familiar with PureBasic, using SpiderBasic should be an easy task. It’s even possible to create a program which runs on the web, mobile and desktop using the same source code.
Lower is faster:
Getting things done with SpiderBasic and here is a proof of concept screenshot of my solver running in a web browser. This is not how it will eventually look. The idea is to have this as a simple gadget, AZdecrypt based substitution solver, more or less with nothing but a few buttons to click. With the free version of SpiderBasic the programs cannot have more than 800 lines so that will help to keep it simple!
Wow, that’s great! A platform-independent solver would indeed be a really, really helpful thing! It’s really motivating to watch your progress!
Agreed, Largo! Nice work, Jarlve. A simplified web-based version of your solver will be very useful and will likely increase the number of people working towards a solution for Z340.
Thanks Largo and doranchak!
Apparently, when a SpiderBasic procedure is running, the window, browser, freezes. There is no straightforward way to issue this and SpiderBasic does not support threading. I solved it by letting the solver run periodically.
Another issue is that SpiderBasic does not seem to allow loading of ngrams automatically. Instead, every file requires user input. So it will seem that I have to ask the user to download the ngrams from a location first and load them with every new instance of the gadget.
What about the idea of running the solver as a stand-alone program in server mode? I did this once with my monoalphabetic solver. It runs as a simple console application and waits for network packets. If you send it an encrypted text, it decrypts it and sends it back. The ngrams only need to be loaded once. You can run as many solvers as you like, as long as each listens to a different port or has a different IP. The items per second are thus only limited by the number of solver PCs. This also creates platform independence. The required library is very easy to use:
There are bindings for all possible languages, also for FreeBasic. As soon as I have finished my magic square test, I can clean up my source code and put it online. But as I said, the connection to the linked library is super easy and done in 10 minutes.
Translated with www.DeepL.com/Translator
Hey Largo,
Yes I remember that. Thanks again for the mention and link though I am not sure if it is something that I would like to do.
