On Feb 18, 8:40=A0pm, Roy <roy.van.r...@[EMAIL PROTECTED]
> wrote:
> bomb =A0 =A0dat =A0 =A0 >-86 =A0 =A0 =A0 =A0 =A0 =A0, <2667
>
> =A0 =A0 =A0 =A0 mov.i =A0 bomb =A0 =A0 =A0 =A0 =A0 =A0, @[EMAIL PROTECTED]
> att =A0 =A0 mov.i =A0 bomb =A0 =A0 =A0 =A0 =A0 =A0, *comp-att
> =A0 =A0 =A0 =A0 sub =A0 =A0 #-7 =A0 =A0 =A0 =A0 =A0 =A0 , att
> =A0 =A0 =A0 =A0 djn.b =A0 -3 =A0 =A0 =A0 =A0 =A0 =A0 =A0, #20
> =A0 =A0 =A0 =A0 jmp =A0 =A0 #0
I think you will find that 'smart' bombing is better :-)
bomb dat 0,0
att mov.i bomb,*comp-att
mov.i bomb ,@[EMAIL PROTECTED]
sub.ab #-5 ,att
djn.b att,#25
In other words, use the internal a-operands of the enemy as pointers.
If you look at a qscan you will see lots of self-pointers that are
'look-ahead's, so you direct bombs at code that is logically
downstream of the code you are targeting.
P. Kline
|
