So what exactly are we doing?

Message boards : Number crunching : So what exactly are we doing?

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Benjamin Rietveld

Send message
Joined: 25 Oct 08
Posts: 13
Credit: 14,389
RAC: 0
Message 708 - Posted: 22 Nov 2008, 15:11:05 UTC
Last modified: 22 Nov 2008, 15:20:37 UTC

I've been reading most of the stuff about the methods used in this project and I get most of it, but I still don't get what exactly are WE doing?
For example, what does one workunit consist of? I guess not just testing one combination, then checking what the outcome is and generating a score, cause that shouldn't take this long. So are we doing a batch of configurations for the enigma machine, then returning the one with the best score? If so, why are the workunits different sizes? Different amounts of configurations?

Just curious :) I love the project!

EDIT: Also, would it be possible to make the graphics like this program for the M4 project?
Graphical Version
ID: 708 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 267,994,998
RAC: 0
Message 709 - Posted: 24 Nov 2008, 2:17:41 UTC - in response to Message 708.  
Last modified: 24 Nov 2008, 2:21:23 UTC


For example, what does one workunit consist of? I guess not just testing one combination, then checking what the outcome is and generating a score, cause that shouldn't take this long. So are we doing a batch of configurations for the enigma machine, then returning the one with the best score?


A single workunit is just a small piece of all possible rotor, rings and start position combinations. For both awgly and hceyz workunits, the wu 'length' is 13 right ring steps.


If so, why are the workunits different sizes? Different amounts of configurations?

That's because rotors VI-VIII have two notches, if one of them is used as right/middle rotor, it causes rotors on it's left side to step more often, and this affects computation speed. That's why workunits with two VI-VIII rotors (type '3' workunits) are the longest.

EDIT: Also, would it be possible to make the graphics like this program for the M4 project?
Graphical Version


Perhaps you could ask the person who wrote this monitor to add configurable path to the BOINC data folder and scanning through slot folders to find all resume files. In it's current state, the program works, but you have to put it inside the slot folder, so BOINC will try to remove it each time after WU is finished.
M4 Project homepage
M4 Project wiki
ID: 709 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Benjamin Rietveld

Send message
Joined: 25 Oct 08
Posts: 13
Credit: 14,389
RAC: 0
Message 710 - Posted: 24 Nov 2008, 9:54:59 UTC - in response to Message 709.  

Perhaps you could ask the person who wrote this monitor to add configurable path to the BOINC data folder and scanning through slot folders to find all resume files. In it's current state, the program works, but you have to put it inside the slot folder, so BOINC will try to remove it each time after WU is finished.

Thanks for the explanation and I'll go do that. I tried to get it to work but I thought I had to put it in the project folder to get it to work :P
ID: 710 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 267,994,998
RAC: 0
Message 721 - Posted: 9 Dec 2008, 1:21:53 UTC - in response to Message 710.  

ID: 721 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Benjamin Rietveld

Send message
Joined: 25 Oct 08
Posts: 13
Credit: 14,389
RAC: 0
Message 722 - Posted: 9 Dec 2008, 6:53:15 UTC - in response to Message 721.  

Enigma Monitor 2.00 is available:

http://hajenius.ifastnet.com/software/EnigmaMonitor/EnigmaMonitor.htm

I was just going to post that :P

I haven't checked it out yet, but it looks promising
ID: 722 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile SUNY-GT Enig

Send message
Joined: 1 Jan 09
Posts: 1
Credit: 11,717
RAC: 0
Message 777 - Posted: 4 Feb 2009, 2:31:09 UTC - in response to Message 722.  

I've had it for a while and its pretty cool. :) My only complaint is that you can't launch it directly from the BOINC Manager, but that may not something that's doable if it can't be BOINC-exclusive. Maybe the next version could auto-detect BOINC and search its slot folders when launched if BOINC is running as well?
ID: 777 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 267,994,998
RAC: 0
Message 778 - Posted: 4 Feb 2009, 12:13:12 UTC - in response to Message 777.  

I don't know if the author visits this forum, but you can find him on the M4 mailing list and ask about additional features :-)

M4 Project homepage
M4 Project wiki
ID: 778 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
LostInTennessee

Send message
Joined: 21 Sep 07
Posts: 5
Credit: 73,360,716
RAC: 0
Message 828 - Posted: 21 Mar 2009, 9:31:38 UTC - in response to Message 721.  

Enigma Monitor 2.00 is available:

http://hajenius.ifastnet.com/software/EnigmaMonitor/EnigmaMonitor.htm



Link not working
ID: 828 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Naesbye

Send message
Joined: 20 Jul 08
Posts: 5
Credit: 4,355
RAC: 0
Message 863 - Posted: 1 May 2009, 6:56:13 UTC

So, when we're done with this run of hceyz (ETA: 13/07/09) and awgly (ETA: 27/3/2010) then the code will have been broken? And can it happen before this?
ID: 863 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile mdoerner
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 30 Jul 08
Posts: 202
Credit: 6,998,388
RAC: 0
Message 864 - Posted: 1 May 2009, 8:42:07 UTC - in response to Message 863.  

Should be, unless it's corrupted (i.e. wrong letters or incomplete transmission) or if it's double-encyphered. Check out one of the threads in CAFE , I think, where this goes into more detail....

Mike Doerner
ID: 864 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Drudge

Send message
Joined: 11 Apr 09
Posts: 18
Credit: 568,771
RAC: 0
Message 872 - Posted: 2 May 2009, 11:28:33 UTC - in response to Message 828.  

Enigma Monitor 2.00 is available:

http://hajenius.ifastnet.com/software/EnigmaMonitor/EnigmaMonitor.htm



Link not working

Link still not working.
"Verloren ist nur, wer sich selbst aufgibt."
ID: 872 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile mdoerner
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 30 Jul 08
Posts: 202
Credit: 6,998,388
RAC: 0
Message 873 - Posted: 2 May 2009, 18:58:07 UTC - in response to Message 872.  

The guy must have fallen off the face of the earth. Anyone have a ZIP file of his app?

Mike Doerner
ID: 873 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [KWSN] PhastPhred

Send message
Joined: 30 Jun 09
Posts: 1
Credit: 155,376
RAC: 0
Message 1100 - Posted: 5 Jul 2009, 0:26:54 UTC - in response to Message 863.  

So, when we're done with this run of hceyz (ETA: 13/07/09) and awgly (ETA: 27/3/2010) then the code will have been broken? And can it happen before this?


Well, it's been a couple of months since you asked...

Did you ever get a response to the question, maybe by PM?

We can't see anything posted...

Enquiring Minds Want To Know!
ID: 1100 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
fitz

Send message
Joined: 15 Apr 09
Posts: 31
Credit: 147,954
RAC: 0
Message 1101 - Posted: 5 Jul 2009, 9:24:05 UTC

I cannot say with certainty but i think i can shed some light.

It is not guarenteed that the message will be broken after both runs, as this last message is known to be part garbled there may not be enough correct transmission data to decode. that is why it has been split in two, the two different runs are working on two different parts of the message to increase the chances of it being broken, as it increases the chance that one run is on a set of correct text. After both runs are complete i believe there is a reasonable chance that it will be broken, Or that the scope will have been significally narrowed for another run?

It can be broken before the end time of the run, and each workunit runs a set of combinations and as soon as that is reported and analysed if one is correct the message would be revealed.

I'm guessing that we would probably have known by now if the heczy run had cracked the message..but hopefully with the increasing amount of computing power being attributed to the project, and maybe people joining because it is close the end of its runs, current at least. we should know within a few months if we have been successful!
ID: 1101 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile mdoerner
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 30 Jul 08
Posts: 202
Credit: 6,998,388
RAC: 0
Message 1102 - Posted: 5 Jul 2009, 13:11:05 UTC - in response to Message 1101.  

Also, looking at the M4 project, I thought the program needed a minimum of 100 characters to reliably decipher a message. Now since the 1st 2 messages broke so easily, it may have been a calculated risk to use 72 characters to get the result more quickly. As we all know now, either the 72 characters weren't enough information to reliably decipher the message or the 1st segment of the message is garbled. So the 2nd message snippet is 100 characters long.

Also, the message may not be garbled, but double-enciphered, in which case I wouldn't hold my breath on this thing just yet.....;-)

Mike D
ID: 1102 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 267,994,998
RAC: 0
Message 1119 - Posted: 8 Jul 2009, 16:49:37 UTC - in response to Message 1102.  

Also, looking at the M4 project, I thought the program needed a minimum of 100 characters to reliably decipher a message. Now since the 1st 2 messages broke so easily, it may have been a calculated risk to use 72 characters to get the result more quickly. As we all know now, either the 72 characters weren't enough information to reliably decipher the message or the 1st segment of the message is garbled. So the 2nd message snippet is 100 characters long.


I believe that minimum for the hillclimb algorithm to be quite reliable (high success rate) is somewhere around 50 characters. Adding more chars will make the statistic part of algorithm (ic-score) work better, but at the same time it increases the chance of encountering missing/garbled letters. And of course the longer the text, the longer it takes to process full workunit.

M4 Project homepage
M4 Project wiki
ID: 1119 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
entrum

Send message
Joined: 4 Apr 09
Posts: 1
Credit: 53,202
RAC: 0
Message 1180 - Posted: 7 Aug 2009, 9:23:56 UTC - in response to Message 873.  

I've been able to obtain Enigma Monitor v1.4b from a kind guy who had it in his archive.

http://nic-nac-project.de/~milo/EnigmaMonitor_1-4b.zip
http://ifile.it/cu39a05 (mirror, just in case)

SHA-1 checksum: ad31a912a669c23725a2225f0c5aa248b210210a

bye
ID: 1180 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile mdoerner
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 30 Jul 08
Posts: 202
Credit: 6,998,388
RAC: 0
Message 1181 - Posted: 9 Aug 2009, 3:24:28 UTC - in response to Message 1180.  

Cool! I wondered if it had disappeared forever or not. Too bad we can't contact the author.

Mike D
ID: 1181 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile TJM
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 25 Aug 07
Posts: 843
Credit: 267,994,998
RAC: 0
Message 1192 - Posted: 1 Sep 2009, 20:57:24 UTC
Last modified: 1 Sep 2009, 20:57:48 UTC

It took quite a while, I finally moved all the hceyz72 results to another database, so they can be viewed directly with a simple interface:

http://www.enigmaathome.net/results_hceyz72.php?page=1

That's 68110 pages, over 1,36M unique results. Unfortunately, the message remains unbroken for now. It would be nice if someone who knows German better than me could review the first few pages and check if there's anything that makes sense.
M4 Project homepage
M4 Project wiki
ID: 1192 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
quel

Send message
Joined: 19 May 09
Posts: 34
Credit: 32,923,471
RAC: 0
Message 1226 - Posted: 9 Sep 2009, 18:17:46 UTC - in response to Message 1192.  

I passed it to a friend of mine and got this back.

"I've started and most of it looks like gibberish with the
occasional word or *possible word* thrown in. To understand some of the
problems I'm facing, I recommend:
http://en.wikipedia.org/wiki/History_of_German there was a spelling reform
back in the 90s (and there is still an on going discussion about spelling).

I also read http://en.wikipedia.org/wiki/Cryptanalysis_of_the_Enigma to get
an idea of what I should be looking for. It outlines more problems
(abbreviations, numbers) and also provides potential solutions.

I've attached a pdf of what I've got so far (I picked several at random),
most of it is guesses, some more educated than others. I've included links
to a decent English/German dictionary, b/c several of the words depend
largely on the context in which they are used.

If I get the chance I'll see if I can get a hold of a friend of mine living
over in Germany (her german is much *much* better than mine) and see if she
can make any progress on it. No promises - she can be a bit flakey.

If I make any more progress I'll send it your way."

pdf is available from http://insomnia.quelrod.net/docs/enigma_round_1.pdf
ID: 1226 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
1 · 2 · Next

Message boards : Number crunching : So what exactly are we doing?




Copyright © 2024 TJM