Author | Post | |||
Towley![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
what about bytes 260-263 ? |
|||
![]() |
|
|||
Phas(retired)![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
It seems to be a 14 bit crc/hash/checksum at bytes 104-106 ((little endian && 0x03FFF0) >> 4): [clear] 0000 2 0040 a1 007c a0 01bc aa 03fc b 0400 c 0680 cccccccc 0e86 bbbbbccc 120f bbbbbbcc 126f abcdef 191a bbbbcccc 1a0f 01234567 24a2 bbbbbbbb 30ec bbbbbbbc 326c ac 397d ab 3bfd 0 3e41 1 3f81 a2 3fbd a 3c01 ba 3c05 Extracting the function from the values is not that hard, but requires a lot of time. I'd try with passwords like 0, 1, 2, 3, 4, 5, 6, 7, 8 and 9. Having only 3 consecutive plain texts makes it harder. |
|||
![]() |
|
|||
unknown user |
it's under that assuption I formulated most of my hypothesis, .. there are just too many things that refuse to make sence to me ![]() i'll provide the additional dumps, over time (takes a bit of effort) tnx for looking at it. additionally this is a diff of two dumps with no password set, across a few boots, for those convinced it are other bytes.. but then again other things might be at play here. ![]() 1c1 < 0000000 48 --- > 0000000 22 3c3 < 0000002 02 --- > 0000002 08 5c5 < 0000004 15 --- > 0000004 00 7,8c7,8 < 0000006 02 < 0000007 11 --- > 0000006 01 > 0000007 08 55c55 < 0000066 01 --- > 0000066 80 177,180c177,180 < 0000260 02 < 0000261 32 < 0000262 33 < 0000263 a0 --- > 0000260 74 > 0000261 21 > 0000262 32 > 0000263 ac 257c257 < 0000400 48 --- > 0000400 22 259c259 < 0000402 02 --- > 0000402 08 261c261 < 0000404 15 --- > 0000404 00 263,264c263,264 < 0000406 02 < 0000407 11 --- > 0000406 01 > 0000407 08 311c311 < 0000466 01 --- > 0000466 80 the addresses are octal as usual. that prevents me from typing "-Ad" ![]() i could also upload the raw files, if that would help anybodies analysis. |
|||
11.06.2007 16:55:33 |
|
|||
quangntenemy![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
That eliminates the other bytes ![]() Now we can focus on the main part. I suspect it's somehow similar to the other checksum algo. Just need to work a little more with my CryptoBox ![]() |
|||
![]() |
|