Hi list, one side note: the success rate of the A5/1 rainbow tables is much lower when looking up random data compared to the success rate of looking up actual GSM key stream. This is due to the collapsing state space that shrinks with every round of A5/1. After the 100 pre-ciphering rounds of GSM, the state space collapsed to somewhere around 2^61. The states stored in our rainbow tables are all part of this set.
Hope this clarifies why finding hits from random numbers takes many tries. Cheers, -Karsten On 06-Nov-10 01:28, Georg Hofstetter wrote: > > Hello 'moongaboonga', > > AFAIK this might be because not all A5 the states are present in the > tables. Try some other bursts. > > With ~700 tried bursts i achieved a success rate of 17.5%. > Not sure how many are because of wrongly guessed bits. > I know, this should be no deal to check this. But didnt have time for > this yet :) > > > Tyke some of my samples: > > Bits__110111110100100100111100100001000100100011100001010110010001011110110101000011101100111010101000111010111000000010 > > COUNT_1979178 > Kc____8E7B6C78C031995 > > Bits__100010111001001010000001111101011100001110000111111011111000011111101011000010010101110010001111100101001010011011 > > COUNT_1987569 > Kc____B26C3B4D2CF691C > > Bits__000110001111011111011110111011101010111011111000111010001001001001011010101101110001000010110111010110010101101101 > COUNT_2005811 > Kc____D9AEE14845BB05DD > > Bits__011010001011010011101110110110011100111010111100001111100101111101001110011101100111001001000100100010111101111001 > COUNT_2013271 > Kc____FDFE856BB9EB5C00 > > > BR, > Georg > > > Am 05.11.2010 15:38, schrieb moongaboonga moongaboonga: >> Hi! >> >> I downloaded 40 tables via torrents, put them on 2 1TB disk, then use 2 >> another one 1TB disks to put tables using Behemoth.py (all extern usb, >> 20 tables on each). >> >> Then compiled kraken, older and now newer version. All goes without problem. >> >> But when try to crack bit strings different then test or one of streams >> from mail list >> >> kraken didnt crack them, just print how time it took. >> >> I tried *many* combination and nothing. No errors just no results. >> >> Then I try to change bits from begining and end of test stream: >> >> When put *first 32 bits or 11 last bits to 0 or 1* the kraken return the >> same result, from table *412*, like for test stream. >> >> Why cant be cracked anything else except test stream and another one >> what was like example in the mailing list? >> >> Thank you in advance! >> >> >> >> >> _______________________________________________ >> A51 mailing list >> [email protected] >> http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51 > > _______________________________________________ > A51 mailing list > [email protected] > http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51 _______________________________________________ A51 mailing list [email protected] http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51
