Reversing bit order in delphi ? (unmoderated)
in [ASM]
|
Prev: Apple II Debugger
Next: TMA Assembler?
From: jukka on 25 Aug 2006 01:54 Skybuck wrote: > 00408F62 8BC2 mov eax,edx > 00408F64 8BD0 mov edx,eax > > Very weird ! Who can EXXXXPLAAAAIN THIS ? Crappy compiler? =) > YOU BAD COMPILER, I HATE YOU NOW, FOEI, FOEI, FOEI ! Just kidding > ofcourse, but it remains nasty ?!?! Holy **** I almost crapped my pants reading this, a nice analysis but the end really caught me.. luckily didn't blow coffee on the keyboard! Christ dude, keep it down will ya' ? :) :)
From: jukka on 25 Aug 2006 02:20 Skybuck wrote: > Thank you for participating, you are free to try again and this time in > clearity, and then you can come again =D lol, have a nice day ! =D You are a meanie.. :( Awwww... but I SHALL HAVE MY REVENGE MUAHAHAHAHA!!! v = ((v * 0x0802 & 0x22110) | (v * 0x8020 & 0x88440)) * 0x10101 >> 16;
From: f0dder on 25 Aug 2006 05:33 Skybuck wrote: *snip* You use some pretty nonstandard way of measuring performance. The normal is to decide a number of times to call each routine, then measure how long it takes (usually in clock cycles, though miliseconds is okay for a large number of calls).
From: ?a/b on 25 Aug 2006 11:42 On Wed, 23 Aug 2006 20:54:17 -0700, Jamie wrote: >Skybuck wrote: > i have copied over piles of source code that was originated on Big >Endian platforms over to little Endian that contain all kinds of bit >operations etc.. never had a problem and never had to change anything >other than byte order coding .. there is only one *good* way to rapresent numbers in a computer 0 0 1 1 2 01 3 11 4 001 5 101 etc
From: Allen Egerton on 25 Aug 2006 12:21
?a\/b wrote: > On Wed, 23 Aug 2006 20:54:17 -0700, Jamie wrote: >> Skybuck wrote: >> i have copied over piles of source code that was originated on Big >> Endian platforms over to little Endian that contain all kinds of bit >> operations etc.. never had a problem and never had to change anything >> other than byte order coding .. > bits are NOT bytes. > there is only one *good* way to rapresent numbers in a computer > 0 0 > 1 1 > 2 01 > 3 11 > 4 001 > 5 101 > etc The decimal (assumed) values on the left do NOT match the binary (assumed) values on the right in your example. Unless you're trying to reverse the bits for Skybuck :). The leftmost columns in binary notation contain greater values than the rightmost, just as in every other numeric based scheme. When counting, you increase the column to its max value, then set it to zero and increment the column to its left by one. Here's a very small example: Decimal Binary Octal 0 0 0 1 1 1 2 10 2 3 11 3 4 100 4 5 101 5 6 110 6 7 111 7 8 1000 10 9 1001 11 10 1010 12 11 1011 13 And the neat thing about the relationship between octal and binary is that each grouping of three binary bits represent an octal column. -- Allen |