From news-rocq.inria.fr!doligez Thu Aug 13 20:03:03 1998 Article: 9482 of rec.games.corewar Path: news-rocq.inria.fr!doligez From: Planar Newsgroups: rec.games.corewar Subject: Re: newbie question Date: 13 Aug 1998 09:28:21 GMT Organization: "" Lines: 56 Sender: doligez@inria.fr (Damien Doligez) Message-ID: <6qubjl\$dvb@news-rocq.inria.fr> References: <6qtntr\$6dq\$1@msunews.cl.msu.edu> NNTP-Posting-Host: volnay.inria.fr Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit >From: mooredav@cps.msu.edu (David Matthew Moore) > >The solution is to copy imps from different places, >such as "MOV aaa, bbb+1". However, this only works if >the imps at aaa and bbb are the same. By induction, >you could conclude that all of the imps on the entire ring >would have to be the same. You could conclude this, but your conclusion would be wrong :-) ;redcode ;name strange rings ;author Planar ;types imp ;hills 88 94 lp big tiny nop ;assert CORESIZE % 2 == 0 ;; a MOV b, d+1 ;; b MOV a, c+1 ;; c MOV d, b+1 ;; d MOV c, a+1 ;; constraints: b-a = d-c ;; a-b = c-d ;; d-a = b-c ;; c-b = a-d ;; these constraints are equivalent to: ;; a-c = b-d = CORESIZE/2 or a-c = b-d = 0 a equ 1234 ;; change these b equ 5678 ;; as you like c equ a+CORESIZE/2 d equ b+CORESIZE/2 z: mov imp1, z+a mov imp1, z+c mov imp2, z+b mov imp2, z+d spl l2 l1: spl l12 l11: jmp z+a l12: jmp z+b l2: spl l22 l21: jmp z+c l22: jmp z+d imp1: mov b-a, d+1-a imp2: mov a-b, c+1-b end -- Planar Remove ".gov" from my address.