[solved] Unknown LF card: relation between written number and content.

marshmellow · 2016-07-29 16:17:22

i would expect the highest value for the digit after FFFF would be 7 so highest possible number = 7999999999 otherwise it would be difficult to determine the preamble (start of stream) so preamble is likely something like

11111111111111110

? but might be necessary to test to be sure.

could try:
FFFF9999999999C7
FFFF8999999999D7
FFFF7999999999E7
FFFF000029447FD6 what would this output?

to answer the few remaining questions regarding this format.

Last edited by marshmellow (2016-07-29 16:18:02)

iceman · 2016-07-29 16:30:42

yup, I added that the MSB the digit after FFFF is not set in clone/sim,
and for the preamble detection it looks for 17bits, to get the zero bit also.

But OP should test it out against his valid reader, as I've asked him to do.

hexa3e8 · 2016-07-29 18:35:37

great work on the jablotron fork update! I am testing it and will post my findings in the other post.
test results:

FFFF9999999999C7 -->  no reaction
FFFF8999999999D7 -->  no reaction
FFFF7999999999E7 -->   370-5032-703   (strange result from reader)
FFFF000029447FD6 -->  14100000294485

iceman · 2016-07-29 18:42:34

The higher

----+-------------ch
FFFFA000011001

The 'A' hex symbol is interesting to try 0-F, against your reader. The output will tell us more about the preamble.

marshmellow · 2016-07-29 18:45:07

so 11111111111111110 preamble is confirmed.

not sure why 7999999999 produced 370-5032-703.... odd
and 29447F resulting in 294485 is odd as well. it converted the F to decimal and added it to the rest of the number in decimal...? probably just not expected by the reader so it is an exception...?

marshmellow · 2016-07-29 18:46:39

iceman wrote:

The higher
----+-------------ch
FFFFA000011001
The 'A' hex symbol is interesting to try 0-F, against your reader. The output will tell us more about the preamble.

9 and 8 will have the highest bit set, 7 sets all the other bits - and 7 worked (ish). i doubt higher will work but as long as testing is fun...

Last edited by marshmellow (2016-07-29 18:47:48)

iceman · 2016-07-29 18:49:59

Yep, agree about preample.

The hex only allows values between 0-9 and not A-F, there might be BCD involved

'7999999999' - 70-5032-703, is strange indeed

marshmellow · 2016-07-29 18:56:03

iceman wrote:

The hex only allows values between 0-9 and not A-F, there might be BCD involved

that is what i was attempting to test but i expected a failure or error or something else, not 85 from 7F
meaning it did 70+15 (F) = 85 - odd.

it does mean that there might be more than one way to get 60% (0-5) of the possible numbers output from the reader...
294485 == 29447F. (not that it matters much... )

Last edited by marshmellow (2016-07-29 19:00:39)

hexa3e8 · 2016-07-29 19:21:48

You guys are fast in replying. just finished the other post.
testing is fun.......yep depends on the situation
So far ....a nice party against testing jablotron........ the party can wait a few more moments.
what to test?
I still test with lf simask.

iceman · 2016-07-29 19:33:41

If it is 70 + 15 then 0x7E == 84?

marshmellow · 2016-07-29 19:56:01

iceman wrote:

If it is 70 + 15 then 0x7E == 84?

correct

this is further proven by:
lf sim jab 18CFE = 19364

so:

E=14     == 4
F=15 +1  == 64
C=12 +1  == 364
18 +1    == 19364

Last edited by marshmellow (2016-07-29 19:57:05)

hexa3e8 · 2016-07-29 20:08:14

good thinking!

marshmellow · 2016-07-29 20:10:37

what i can't figure is WHY...

iceman · 2016-07-29 20:15:47

So, it takes hex, converts it to decimal and then adds it to gether when digit is > 9..

marshmellow · 2016-07-29 20:34:30

iceman wrote:

So, it takes hex, converts it to decimal and then adds it to gether when digit is > 9..

that is what it is doing... kinda...

it must convert each digit from hex to decimal and then multiply by digit (10s 100s 1000s etc.) then add it all up.
instead of doing it byte by byte it does it per 4 bits.

for 18CFE

1 =  1 * 10000 = 10000
8 =  8 *  1000 =  8000
C = 12 *   100 =  1200
F = 15 *    10 =   150
E = 14 *     1 =    14

10000 + 8000 + 1200 + 150 + 14 = 19364

this process must be what is used to convert the hex to decimal. odd.

Last edited by marshmellow (2016-07-29 20:40:54)

iceman · 2016-07-29 20:37:53

Its checking if the hex is not a BCD, for each nibble ie hex symbol. Look like its a custom built BDC implementation

And shouldn't it become: 19364 (above shows 19464)

the C should be 12?

marshmellow · 2016-07-29 20:40:36

correct

iceman · 2016-07-29 22:29:15

So I pushed some changes, if now prints the card id as @OP's reader output but the odd 7999999999 behavior I can't reproduce.

ref http://www.proxmark.org/forum/viewtopic … 389#p23389

Proxmark3 community

Announcement

#101 2016-07-29 16:17:22

Re: [solved] Unknown LF card: relation between written number and content.

#102 2016-07-29 16:30:42

Re: [solved] Unknown LF card: relation between written number and content.

#103 2016-07-29 18:35:37

Re: [solved] Unknown LF card: relation between written number and content.

#104 2016-07-29 18:42:34

Re: [solved] Unknown LF card: relation between written number and content.

#105 2016-07-29 18:45:07

Re: [solved] Unknown LF card: relation between written number and content.

#106 2016-07-29 18:46:39

Re: [solved] Unknown LF card: relation between written number and content.

#107 2016-07-29 18:49:59

Re: [solved] Unknown LF card: relation between written number and content.

#108 2016-07-29 18:56:03

Re: [solved] Unknown LF card: relation between written number and content.

#109 2016-07-29 19:21:48

Re: [solved] Unknown LF card: relation between written number and content.

#110 2016-07-29 19:33:41

Re: [solved] Unknown LF card: relation between written number and content.

#111 2016-07-29 19:56:01

Re: [solved] Unknown LF card: relation between written number and content.

#112 2016-07-29 20:08:14

Re: [solved] Unknown LF card: relation between written number and content.

#113 2016-07-29 20:10:37

Re: [solved] Unknown LF card: relation between written number and content.

#114 2016-07-29 20:15:47

Re: [solved] Unknown LF card: relation between written number and content.

#115 2016-07-29 20:34:30

Re: [solved] Unknown LF card: relation between written number and content.

#116 2016-07-29 20:37:53

Re: [solved] Unknown LF card: relation between written number and content.

#117 2016-07-29 20:40:36

Re: [solved] Unknown LF card: relation between written number and content.

#118 2016-07-29 22:29:15

Re: [solved] Unknown LF card: relation between written number and content.

Board footer