From news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!blackbush.xlink.net!news.uni-mainz.de!news.th-darmstadt.de!uni-erlangen.de!lrz-muenchen.de!news Wed May 15 13:44:54 1996
Article: 5122 of rec.games.corewar
Path: news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!blackbush.xlink.net!news.uni-mainz.de!news.th-darmstadt.de!uni-erlangen.de!lrz-muenchen.de!news
From: guenzel@extern.lrz-muenchen.de (Bjoern Guenzel)
Newsgroups: rec.games.corewar
Subject: coal
Date: Tue, 14 May 1996 22:09:45 GMT
Organization: Tichy Works
Lines: 68
Distribution: world
Message-ID: <4nasoo$f3c@sparcserver.lrz-muenchen.de>
NNTP-Posting-Host: dial043.ppp.lrz-muenchen.de
X-Newsreader: Forte Free Agent 1.0.82

Here is coal, I think I will not investigate in it any further. It was
a little inspired by discussions with David van Damn and Paul Kline.
Initially I wanted to make a bomber that would colour the core mod 4
and mod 5, to slow down all kinds of scanners (=smoke). Stupid enough
I didn't realise at first that mov <x,y only colours mod 5 ... Well,
perhaps it was good against dodgers....

questions, comments, suggestions? (Seemingly this could as well be a
rhetoric question... my poor warriors are probably too strange, so
nobody can add something)

Bjoern


P.S.: As one feature I planned to add imps, but I never managed to. I
am curious for Beppe Bezzis next Core Warrior hint! In the meantime I
found a method, but it is very slow :-(



;redcode-94
;name coal 3.22b 
;author bjoern guenzel
;strategy burn and smoke - stone without extras, but improved against 
;strategy djn-streams
;assert CORESIZE==8000
;release 29.4.96 - got 53 challenges old 
;kill coal
;p-class stone,clear

step4		equ (4*19)		;steps more or less randomly chosen...
step5		equ (5*19)
gate		equ clptr+5000
magic		equ (2000-1)		;a field has to point to bptr exactly when
					;b field points to ptr
		
		dat 0,0			;dec,hit
		dat 0,0

		
dec		djn.f -1,{clptr-ptr
clptr		dat -1,100	

		dat 0,0				;hit

b1		spl #b2-bptr,bptr-clptr+1 	;dec 	 
step		spl #-step5,#step4	;first to be executed
		mov.i *bptr,>clptr
ptr		mov.i <bptr+magic*step5,ptr+step4 	;last hit -  by the
mov,
							;which moves a djn -1 there. So I have 
		add.f step,ptr				;a 0.5c clear+djn-stream
		djn.f ptr,<gate				;dec - mutilated djn stream :-(

		dat 0,0
		dat 0,0			;hit

b3		dat #b3-bptr,switch-clptr+1
b2		spl #b3-bptr,bptr-clptr	
bptr		dat #b1,switch+2	;dec

		dat 0,0			;hit
	
switch		mov.i dec-ptr+step5,>1-step4
		end step