decrypt and win $100

[Dio]

i knew that would get your attention.


using my picture encryption thing i scrambled a picture that had text in it.

here is the picture:




i'm even giving you people the source code and i still think you can't do it. and putting 100 dollars on the line.*

* i may or may not send you 100 (hundred) dollars. you must post (PM) your source to qualify for receiving the 100 dollars. however even if you do meet the criteria to win the $100, i may not send you $100. based on how cool your entry is. my judgment is arbitrary and final. contest never ends unless otherwise noted my me. by posting an entry, you acknowledge the rules above. Please ignore spelling errors and what not. look, just don't sue me!

[Anonymous]

Ur code crashes out of the box. the one where u say "yay success i got it"

edit: needed "option dynamic" at top, those arrays on top are too large to fit into the stack (internal compiler changes in newer versions of fb.)


You realize this is an irreversible encryption, right =)

edit2: program still crashes.

edit3: making it dim shared, removing option dynamic and removing those other 2 arrays makes it work. looks klike that's an fb bug, i'll report that on fb forums asap.

and btw i saw the randomize timer, but now i see it in fact is reversible. i'll try cracking it...

[Anonymous]

btw, if you cut this image out of a larger image you encrypted using that algo it's irreversible, so, did you? =p (brutefocre cracking prog running in background)

[Anonymous]

Code:

option explicit
#include "crt.bi"
const as integer sz = 161 * 11 * 4


'sub dio_DeCrypt( srcImg as any ptr, dstPtr as any ptr, seed as integer )
#macro dio_DeCrypt( srcImg, dstPtr, seed )

'  if srcImg = 0 then exit sub
    
'  static as integer imgWidth, imgHeight, i, x, y, pa, pb, rx, ry, xo, yo
'  static as integer ptr pix, imgData2
  
'  imgWidth = cptr( ushort ptr, srcImg )[0] shr 3
'  imgHeight = cptr( ushort ptr, srcImg )[1]

  
  memcpy( dstPtr + 4, srcImg + 4, ( sz ) )
'  memcpy( dstPtr + 4, srcImg + 4, ( imgWidth * imgHeight ) * len( integer ) )

  imgData2 = cptr( any ptr, VarPtr( cptr( ushort ptr, dstPtr )[2] ) )
  
  randomize seed
  
'  dim as integer pic( sz * 8 )
  
  i = 0
  for y = 0 to 11 - 1

    for x = 0 to 161 - 1

      pic( i ) = int( rnd * ( 161 - 1 ) )
      i = i + 1

      pic( i ) = int( rnd * ( 11 - 1 ) )
      i = i + 1

    next  

  next
  
  pix = varptr( imgData2[(11 * 161) - 1] )
  for y = 11 - 1 to 0 step -1
    
'    yo = (y * imgWidth)
    for x = 161 - 1 to 0 step -1
  
      i = i - 1
      ry = pic(i)

      i = i - 1
      rx = pic(i)
      
      xo = (ry * 161)+ rx
      
      
'      pa = imgData2[yo + x]
      pa = *pix
      pb = imgData2[xo]

      imgData2[xo] = pa
      *pix = pb
'      imgData2[yo + x] = pb


      pix -= 1

    next    
    
  next
  
#endmacro
'end sub


#macro check_topline( __COLOR__ )
  if j then
  for l = 0 to 160
    if lc[l] <> __COLOR__ then
      j = 0
      
      exit for
    end if
  next
end if
#endmacro

#macro check_rightline( __COLOR__ )
  if j then
  for l = 0 to 10
    if lc[m + 160] <> __COLOR__ then
      j = 0
      
      exit for
    end if
    m += 161
  next
end if
#endmacro

#macro check_bottomline( __COLOR__ )
  if j then
  for l = 0 to 160
    if lc[l + 1600] <> __COLOR__ then
      j = 0
      
      exit for
    end if
  next
  end if
#endmacro

#macro check_leftline( __COLOR__ )
  if j then
  for l = 0 to 10
    if lc[m] <> __COLOR__ then
      j = 0
      
      exit for
    end if
    m += 161
  next
end if
#endmacro

#macro check_alllines( __COLOR__ )


  check_topline( __COLOR__ )
  check_rightline( __COLOR__ )
  check_bottomline( __COLOR__ )
  check_leftline( __COLOR__ )
  

#endmacro

screen 14, 32

dim as any ptr img1, img2
dim as integer ptr lc

img1 = imagecreate( 161, 11 )
img2 = imagecreate( 161, 11 )

bload "10dollarch.bmp", img1
'bload "newimg.bmp", img1

lc = cptr( any ptr, varptr( cptr( ushort ptr, img2 )[2] ) )
dim as uinteger iii, j, k, l, m, ddd

dim as integer imgWidth, imgHeight, i, x, y, pa, pb, rx, ry, xo, yo
dim as integer ptr pix, imgData2

dim as integer pic( sz * 8 )



if command( 1 ) <> "" then
ddd = val( command( 1 ) )

end if
  


'for iii = ( 2 ^ 32 ) - 1 to 0 step -1
'for iii = 170000 to ( 2 ^ 32 ) - 1
for iii = ddd to ( 2 ^ 32 ) - 1

  dio_DeCrypt( img1, img2, iii )
  
  j = -1
  k = lc[0]
  m = 0
  
'  check_alllines( rgb( 255, 255, 255 ) )
  check_topline( k )
  k = lc[160]
  m = 0
  check_rightline( k )
  k = lc[1600]
  m = 0
  check_bottomline( k )
  k = lc[0]
  m = 0
  check_leftline( k )


'  rgb( 255, 255, 255 )
'  for l = 0 to 10
'    if lc[m] <> rgb( 255, 255, 255 ) then
'      j = 0
'      
'      exit for
'    end if
'    m += 161
'  next
  skip:
  if j = -1 then
    ? "heh: " & iii
    bsave "dioresult.bmp", img2
    sleep
    
  end if
  
  if multikey( &h2F ) then
    put( 0, 0 ), img2, pset
    locate 20
    ? iii
    
  end if
  if multikey( 1 ) then
    end
    
  end if
  
    
  
next

verdict: after about 100 million iterations between a few computers, i think it probrbaly isn't brute force crackable =|

[Dio]

Ur code crashes out of the box. the one where u say "yay success i got it"

edit: needed "option dynamic" at top, those arrays on top are too large to fit into the stack (internal compiler changes in newer versions of fb.)


edit2: program still crashes.

edit3: making it dim shared, removing option dynamic and removing those other 2 arrays makes it work. looks klike that's an fb bug, i'll report that on fb forums asap.

ok. i admit. i suck at programming. Besides, i improved on the code since then.

btw, if you cut this image out of a larger image you encrypted using that algo it's irreversible, so, did you? =p (brutefocre cracking prog running in background)

don't worry. i did good :^_^:

verdict: after about 100 million iterations between a few computers, i think it probrbaly isn't brute force crackable =|

and that image is 11X161. with plain white text. I figured you could make a program to search for letters. But can you imagine 1000X700 images that could be anything from B&W satellite photos to other shit you don’t want grandma to see. And could you imagine 100 million iterations at 1000X700? And what if the FBI used the source code to try and decrypt it but I kept re-encrypting the same image a thousand times or a million times? Should they have their computer (super in orientation no doubt) go thru the same image and the same pass re-decrypting a million times to be sure? What if I changed the pass 500,000 iterations into it?


How ever, it may only work for BMPs though. JPEG compresses and screws with the pixels

[NovaProgramming]

So does that mean that you can decrypt that image into whatever it's supposed to be, but you're just waiting to see if we can do it? Because that's pretty sweet :wtnod2:

[Dio]

i guess.

[NovaProgramming]

hey all I know is that I couldn't program that cuz I stink at that kind of stuff. That's why I didn't try to decrypt it. my "skillz" are n00bish at best. But I can still outprogram my brother who is a Computer Science major .

[Dio]

my "skillz" are n00bish at best. But I can still outprogram my brother who is a Computer Science major .

i would have an aneurysm trying to program "Ballsbreaker" :wow: :wow: :wow:

[NovaProgramming]

lol Ballsbreaker *2* should be so much better if it ever cools down outside. I use my laptop for programming and the best atmosphere I find is either in class, while I'm supposed to be listening, or I'm in my bed with the TV on, using my lap for the laptop (ingenius, I know). But lately, as you all know (all of you in the U.S, at least), it has been extremely hot. It was a record-breaker here, something like 93 with a heat index of around 106... and it's northern minnesota so it's supposed to be cold