From news-rocq!univ-lyon1.fr!in2p3.fr!oleane.net!oleane!news.maxwell.syr.edu!xfer13.netnews.com!netnews.com!newspeer1.nac.net!news.ttsg.com!jkw@koth.org Fri Nov 17 21:54:58 2000
Article: 11944 of rec.games.corewar
Path: news-rocq!univ-lyon1.fr!in2p3.fr!oleane.net!oleane!news.maxwell.syr.edu!xfer13.netnews.com!netnews.com!newspeer1.nac.net!news.ttsg.com!jkw@koth.org
From: jkw@koth.org
Newsgroups: rec.games.corewar
Subject: Re: kofacoto quarterfinal results
Date: 17 Nov 2000 15:00:00 -0500
Organization: TTSG
Lines: 57
Distribution: world
Message-ID: <4.1.20001117133257.00abf780@pop-server>
Reply-To: jkw@koth.org
NNTP-Posting-Host: odin.ttsg.com
X-Trace: odin.ttsg.com 974491200 140 216.231.111.29 (17 Nov 2000 20:00:00 GMT)
X-Complaints-To: abuse@news.ttsg.com
NNTP-Posting-Date: 17 Nov 2000 20:00:00 GMT
Originator: corewar-l@koth.org
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.1 
Xref: news-rocq rec.games.corewar:11944

>>Moore submitted what I think is a one-shot scanner, which
>>he called a two-shot scanner.  Odd. :)  Close battles...
>>that little G2 program was cagey...
>
>Its a two shot scanner.  The first scan modifies the scan loop to do 
>blur-style bombing in the loop.  The second scan jumps to the coreclear.

Hmm I guess you're right...  I looked at it moore :) closely this time
and see that it does have a brief period of blur/mirage-ish scanning.

If you don't look closely, you'll miss that
	djn.f *-4, #0
being converted to
	djn.f *-5, #0
...causing the switch.

That's very interesting...  I wonder if this could have a good chance
against imp/stones if it used a d-clear instead of the spl/dat clear.
I'll have to test that out. :)

;redcode-94
;name G2
;author David Moore
;assert CORESIZE==8000
;strategy two-shot scanner

ival equ 10
step equ 191
x equ (4+step)

      dat 0,0
ptr   dat 0,0
      dat 0,0
sw    dat 1, 14
g2    spl #ptr+100, 14
      dat 0,0
      dat 0,0
clr   spl  #step, step
      mov   *sw, >ptr
      mov   *sw, >ptr
      djn.f  -2, }g2
      dat 0,0
      dat 0,0
      dat 0,0
      dat 0,0
      dat 0,0
      dat 0,0
      dat 0,0
      mov   clr, >ptr+2
      add   *-1,  1
main  sne     x, x+ival
      djn.f  *2, *-1
      mov.b main, <-4
      djn.f *-4, #0

end main