« Winnners for Challenge 6 | Main | Winners for Challenge 7A »
December 09, 2005
FIRST HINT FOR CHALLENGE 8B
[For clarification: this is NOT a change to the text of Challenge 8B, it
is extra information about it. You still need to submit your decryption
of the ORIGINAL 8B text.]
I just got this from signals intelligence. Apparently immediately after
we received the encrypted transmission yesterday from the Weather
Station we picked up a ghost transmission from the same source on a
different frequency. It was very faint but we were close enough to get
all of it I think. Here it is:
MDY5H YI2Y1 D8EV4 H08WU O1E7K 2502D CG8DV Z5X4T DT5IO GW2KI G7HJA GUYZW
SI68B I5OSY HAS03 DVZMK MNARO 7O8I9 R1YR4 839DH ORMUR SO4VJ THO2P ZR7PZ
SJZHO T3WPH CK3NL V9MM9 IY5VZ B8B4Z 02VNM W4M3L D3XRO OX2WZ 5ULZM KXKY5
0H14O VJH4N OZ6U7 RF1IX 8VQLN K3W5U GO4Y6 EMWFN NY47G A9IYK FL06C MYYME
R7AR7 90RVT GB0T3 CSA3C CYS9F NWH0V R4YBL Y5HE1 MBXI9 UGG1U ECI5H SKUGA
PR7CG BLED4 KN7FU QLZMC 839EV ACSVL 9U5II QN8V1 029ZD BIVTO DLMC9 XXOP0
T60FA U2LUH ZMSW4 PNY3B 5NG8I 7H773 KEQG1 IYWNR U4RPK Z6BJO ZZJSV P27WY
ZKV4P NFMTH TQ5OW EKZQD VDFM1 9XQDO PGRW8 3L4DD HFPOF 7121M TLCW3 HXZ75
XOK7M SM8C0 SC2KS D1OYR TLLW6 O715Z WVWEJ XN4VU 3MP2T U5Q1Q F4MYF XFMQK
OPKLU 3GK94 2MZHM YOE46 AYGJ2 IS2L2 QAL30 TF9O6 2QDMJ 4X5JC SWJ1W IIEQ1
NAL1T S0J06 C2XMX NKWDL XFMHG E6EQI Z3UX9 PAJAK JSP5S WPPK4 WBM8P 89BZW
UY6WL FN4PW 3BXBZ 92X5N GNF4Q KY8HR XSUVC WTXWW HDZFR M5ZKG FVAGP DGIDR
POJXA 03D8D 3MTG4 1IKVQ T60OT AKE2L 0CWEF YEY8Z 9Q4DJ IEN8J JF0KC PLM36
RD72V RHSJ9 9VPWR OWNI0 D1PED 6ZMNK FNIYH EWJF4 ID4XI EFU76 3MYDA 4C8OT
AFLQ9 PHKXD FNWSS NZQCX SOB3P BKBMO 9WOD8 WQIBJ NIS5O T5UKS C84JQ L1WOB
XK0UX MJ6IF END1X H07XM WM2RZ QVVC5 2MFS2 2CYBW N6MKT 8JMQD TCJEH 4HQFF
UFKLM YWF5C YMK18 0X68X Z79R1 O3081 XHPJT 4URB0 IR0QX 8W6E1 GLQP2 D2TR8
C7DCS WFSS7 BLXCU OWY0R 027IL YNYRM 7GBAP IHEO5 N2ZUN P5V3S X94RO PR5SP
6PJMQ PDOKQ 4PFWF VPXBJ XXDYO V8EEO YJG6I CU66R YKNZY 6SFH8 KPRJ6 XYCH7
EFXWP GQII1 LFQR2 T3Y35 86W7V NO7YN 993NO GH4ML XPPNE OO3P8 SV7ZJ SC0FA
3Q08Z PNMWH 2L7BL NEGKJ MPF77 DVYF2 9BF4L 8JO4L 6JKFK QXIDH VP9O8 HZVUX
6FWEC NDAAQ 62OLJ X6R7Q KHOWF Z0YM7 WZCDC I6DYI GP8Z0 SZMGA UH087 NS0NF
YS60L 13JOG V8KO6 Z3XDF XJW84 TZH8G H7UQ3 259H1 1ZTUQ ISKIC 9PW3H 6LHZQ
N1VPI A215S G8JB5 NKIBN OIODE 0K1FP GKXA4 P91OR 72TDW XHAJC IH2WG EN55R
JM0XH LZWZ1 MAJ09 9OIYA U42KN ENUP5 TFGCG GY1SW QJ3YO 0I1QL 5WA03 78P2Y
CK51C GLO46 6HJG6 XWRFJ AVKH7 1L3BG 7QHD6 BJSKB EXLRN TZR7Y 9BSQI TU4SI
Z8S8Y 4BHF5 ONVHR 1Z0CZ GONKP K6VH0 N50J9 HVBTC VULD2 TPA9Z XXQXX 6NZN7
1E0YF UKM2U NQTC6 84IO1 0SAYM H0KWZ A4XG2 9W3UI EOBJU ZBSTN 1ALA1 9WZKI
BV9YK HYE8I LICF3 YXBUT ACLJY D0KRW GPSBT J9E5K SNFM2 J9AS5 3MMK4 BZYSP
WN3H9 HUK3K UNOC0 XUH97 SA7AD FSE46 SL6M7 THS3K WHM8C 5GUD3 52Z1X WMMJN
76F5G 1NNW0 RH0W3 G1JQJ 3OJ4A PMTRG 7K6UT P71KZ 00DZ1 G5JXP U8U14 DX00T
TIKBQ B2A9D G11IO J6IYE MNB9Z J4J04 IFE8Y F263T L45C7 DEWZZ J0MHO 399QD
GH0DD PN915 07JGE YK905 SW87U Q595M NXCLJ EKY97 PWQXE RTBX3 RQZQD 1X8WZ
5IC00 Z0USF LVTX4 VYS6F 5SH4Q 6PZET IDGHF 9BPVN DQA7P JFLQC X196U WX04J
136YK V186A EUV1H XZZ4R CVWHG G0YVF 9AZ75 UOT14 DSACA NF1XX IDDCV Y0V6E
JA02G NLH2S JN9KW I1QO6 N19YI EZSKW 4R9JZ WPGOY IVSG8 2PA6V JRWTC KIP9G
I6PSL YPVMD HOGWF U2PIX 1ADX6 Z9JU5 9YUPL DM1OR 0YIA6 1M6T7 J2P41 9LBBK
4B9WC 820DQ G1Q7R 81QC1 CX2MU IDPPK 06RQ4 0BQ1B AKKMH CP2BF YSGO6 S6V9Q
ALXCW OLD40 T5GLN 0QJMS 1GSGV 3L80Z W48VG AXPRS HJFS1 NMMU9 U5DVG HUJ6F
2MCKN OSGEO ATQTG JOGE7 LPGY7 9R5V9 P0M3S XCSO6 90JOP F6RIQ SKXH3 BX3VH
S91V7 DCVGO NGMQZ TLQU
As you can see it starts off the same as the transmission we received
but has a whole chunk extra at the end. I asked the radio guys if they
have any idea what was happening and one of them suggested that it might
be a secure long range transmission using a directional mast. That's why
we found it hard to pick up. But he made an interesting remark. If it
was long range then the main part of the message (after any headers,
weather reports etc) would normally be sent twice in case of
transmission errors. If my hunch is right this extra text will be the
report from the captain of the Leninsky Komsomov. Now looking at the text we can see they haven't just transmitted the encrypted text twice, so I figure they have a made a classic mistake and instead have encrypte part of the message twice!. If my hunch is right the extra text is just the captain's report and I think we all might be able
to guess how that starts.
Now it is almost certainly a coincidence that the initial message had
1296 characters in it but it is highly significant for us. It means that
at the end of encrypting the original whole message the rotors were back at their
initial position. If the clerk then re-entered the captain's report he
would have been doing so from the SAME INITIAL SETTINGS as the machine
started with.
If we can guess how the captain's report starts then we might be able to
reduce the number of possible wirings of the rotors.
I had a look at my notes, and have a couple more observations to make on
the machine. The rotors are driven using a similar mechanism to the
Enigma. The first rotor turns fast and every complete revolution of it
drives the second rotor one notch in the opposite direction. The catch
engaging the second rotor is adjacent to the Z on the wheel, but of
course if the first rotor is inserted with the Z at the top then the
second rotor could turn when the first character is entered. By putting the
first rotor in at different positions you can get the first turn of the
second rotor at any of the first 36 characters of the ciphertext.
Presumably the Soviet cipherclerk had been given initial settings for
the rotors. Maybe they were written down and I didn't notice.
Our photolab guys tell me that the other rotor image should be back from
processing at 3.30pm on Monday.
More later.
Posted by Harry at December 9, 2005 03:03 PM
Comments
If you wish to leave a comment please act responsibly. No personal or innapropriate comments should be made. Concerns about a comment should be emailed directly to us here. Hints about the challenges may be blocked or held back at our discretion. We reserve the right to refuse to publish comments without entering into a discussion about why, and to censor comments as we see fit. Please be patient if it takes a while for your comments to be approved.
how many solvers so far?
[None so far. Harry]
Posted by: Luke at December 11, 2005 07:09 PM
Erm Harry, I'm sorry but how exactly are you expecting us to draw absolute and definite rotor wirings fron the diagrams. There are so many that are very very debatable and some that seem completely impossible. My monitor isn't very awful, in fact I think it's quite good. AArrrggghh.
[Who said I expected you to get definitive connections? The drawings cut down the possibilities considerablyt though. Even if you don't trace any of the wires you there are just a few pins of each colour so this cuts the number of possibilities considerably. As for the reflector, on my monitor I could trace all but two of the wires. As someone said try viewing the image in a graphics program and varying the hue and saturation if the colours are too close. I will be giving more info about the wirings as the challenge progresses. Harry]
Posted by: sean at December 11, 2005 06:18 PM
Hi can anyone tell me how i am supposed to make an enigma/fialka machine out of a PRINGLE!! tube of all things, this will be hard for me as me and my maths class as we are S1 and scottish!!
thanks....
Posted by: Cameron at December 10, 2005 07:24 PM
Well, I've got the reflector down to 9 possibilities, only just started looking at the rotors, but how does the punchcard have rows and columns?! It's just random spots on the page! Can anyone let on how the punchcard works, and how I should be looking at it? How many rows/columns are there?
Posted by: Will at December 10, 2005 05:16 PM
It seems that this is a lot more complex than I anticipated, and my original thoughts haven't worked. The hints are just too unrelated! However, I think I have made some useful observations and I might be able to find a way to make a program severely cut down on the possiblities. Of course, I will not elaborate, but I think the method may be similar to last year's cipher. At least, my method. The number of unknowns may hinder success though. I will go over my design and then get to work with the programming, if I can see it working. Design is a very important stage here, as brute force will not be successful using every setting, at this stage. I think, when the image is done for Monday, it may become possible for brute force to break the cipher, but who knows?
[You will have a lot more information once the other rotor image is cleaned up. Harry.]
Posted by: Jason at December 10, 2005 04:35 PM
help!!!!!!!!!!! i'm sooooo confused i looked up the machines but that confused me more! I have no idea what everyone is going on about, rotors, reflectors?? and with loads of people talking about writing programs i feel really left behind, help me harry please!!!!!!!!!
Posted by: becky at December 10, 2005 04:33 PM
im intrequed to know how robert lasenby is getting on lol!
Posted by: martyn compton at December 10, 2005 12:16 PM
just hoping everyone has found that the morse is twice encoded, oh and kati the five part morse chars are actually numbers, hope that helps
i cant believe im even trying this by hand!!
Posted by: martyn compton at December 10, 2005 12:06 PM
Hold on...which bit's incrypted twice and where does it say "COMRADES"?
is "6DYIGP8Z" "COMRADES"?
[That would be my guess. Harry]
Posted by: Captain Cool at December 10, 2005 11:39 AM
Actually you really don't have to guess the wiring diagrams, assuming you got most of the first rotor, and almost all of the reflector wirings right, and at least some of the second rotor wirings correct, there will last blocks of text that makes sense when its deciphered (as rotor2 only changes every 36, if it gets on one of the few right choices, it will be right for 36 characters). Guessing the wiring diagrams is futile, but you can trial and error them when you have some plaintext (I havn't managed to get any plaintext yet thought :( )
Posted by: Captain Cool at December 9, 2005 08:58 PM
Good luck with it Jason, for me at the moment it isn't a weakness by your definition :(
As for the reflector wiring, I've got a good guess at what it is, but we'll see. I also supposedly have the rotor 1 possibilities down to 16 :D
Posted by: Ian at December 9, 2005 08:36 PM
Of course this is EXTREMELY useful. Encrypting twice is a CARDINAL SIN in the world of cryptography. However, they say in the world of chess that a weakness is not a weakness if you cannot exploit it. Don't worry about that though - I'll find a way!
I really needed that rotor turning info. Now I can really get going! I was trying to work out how a program could efficiently test all combinations, but I doubt that that will be so complicated any longer because of this clue. I didn't expect it so soon, but OK. Also, the number of possibilities has been severely decreased. I intend to have the cipher completed this weekend, but is that too optimistic?
Also, you said that you could trace ALL the wires in the reflector. But two of the pairs get mixed up with each other. Have another look Harry!
[Good luck! Any of the rest of you who want to beat Jason had better get a move on. As for the reflector wiring, maybe I have a better monitor, or maybe I just know where the wires are going. Harry]
Posted by: Jason at December 9, 2005 07:22 PM
Are you saying that the ways of putting the rotors in matters more than the combinations of the wires? Because the number of possibilities you have at the moment is 3 869 835 264, but even if you know as much as half of Rotor 2 wirings that's still over 6 000 000 000 000 000 combinations (about 1.5 million times as much).
Posted by: Ian at December 9, 2005 07:09 PM
Harry
If the fialka machine can be used for ciphering and deciphering, and as this prototype uses lamps, does that mean that like Engima it can't encipher a letter as itself? Or is it really only a program that you've written?
[I think it is like the Enigma in that respect. The genuine Fialka should also be hampered like this but I think it is believed to have a special circuit to allow it. This pseudo Fialka is much simpler. Harry]
Posted by: Sir Solvalot at December 9, 2005 06:57 PM
ok got written a program to check everything, only problem is there are too many unknowns!, 36 ways of putting reflector in, 36 ways of putting rotor1, 36 ways or putting rotor2, 8 ways of positioning rotors, 8 ways of putting in punchcard. thats:
36*36*36*8*8 ways my poor program has to try, on top of the some of rotor1 and most of rotor2 are unknown, meaning it has to have room for error, so I'm making it check for the number of 3 letter words each time it's deciphered, then outputting all results with more than 4 three letter words. This means for its checking 1296*36*36*36*8*8 characters...I can't use a crib, even though i'm pretty sure i have one, because the odds are even if I get the correct settings, the wiring won't be right, so the program wont know its found the crib :/
Posted by: Captain Cool at December 9, 2005 06:45 PM
The problem OcelotIIX had reading "these men" is really the fault of these men doing Soviet coding. Better for them if they introduced a space character between words. What if a weather report was sent as "Winds W" and read as "Wind SW"? At Leyte Gulf, Admiral Halsey is said to have mistaken some padding for part of the message, with adverse results on his judgement.
Posted by: John at December 9, 2005 05:48 PM
harry you are evil, that is what the boy called mat says anyway so i think i will stand by him, he is in 19
Posted by: william at December 9, 2005 05:29 PM
Harry, and everyone else, i might be being incredibly dim here, but what on earth are rotors and reflectors? what do the diagrams mean? and possibly most important of all, what on earth is this hint about??????????????? please tell me something. *looks hopeful*
[Have you looked up the Enigma machine, or the Fialka machine in a book or online? Harry]
Posted by: lara at December 9, 2005 05:13 PM
Ooo... How many people have cracked 8B so far?
[None of course! Harry]
Posted by: Ian at December 9, 2005 04:45 PM
I got the morse off the RSS feed and it was totally different. I've been wasting my time I think. Ish.
[The Morse as published yesterday was a string of zeroes, ones and twos. These as usual represented the dots, dashes and slashes of a Morse message. Run through the Morse decoder a second time you get the text as just published. Harry]
Posted by: Ben at December 9, 2005 03:44 PM
Harry,
Could you, or anyone else, confirm the operation of the machine is... Ciphertext 'C'' entered; i
Plugboard, leaves as 'C1',
Rotor1: C1 to C2
Rotor2: C2 to C3
Reflector: C3 to C4
Back to Rotor2: C4 to C5
Rotor1: C5 to C6
Plugboard: C6 to plaintext!!
Good luck to anybody attempting this by hand if this is correct!
[That's what it looked like to me. Harry]
Posted by: Mike at December 9, 2005 03:41 PM