From news-rocq.inria.fr!univ-lyon1.fr!oleane!dish.news.pipex.net!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!newsrelay.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s Fri Jul 7 13:55:40 1995 Article: 1934 of rec.games.corewar Path: news-rocq.inria.fr!univ-lyon1.fr!oleane!dish.news.pipex.net!pipex!howland.reston.ans.net!vixen.cso.uiuc.edu!newsrelay.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s From: pk6811s@acad.drake.edu Newsgroups: rec.games.corewar Subject: Withershins Thrice Date: 6 Jul 95 14:30:59 CST Organization: Drake University, Des Moines, Iowa Lines: 57 Distribution: world Message-ID: <1995Jul6.143059@acad.drake.edu> NNTP-Posting-Host: acad.drake.edu Currently scoring 160-170 on the '94 Hill, my only regret is that in tweaking for a high score I couldn't also make it beat Agony. This is the same style bomb/scan now in Rude Wind II, and makes use of the multi-pass core-wiper I posted earlier. It's an old strategy, make one fast scan then start wiping core, but works pretty well. Only three programs outscored it, Agony, Duel, and Quiz; the multi-pass clear using two spl-passes with repeating dat-passes works very well against Silk-style replicators. (Those 73% wins against B-Panama are especially gratifying :-) FYI Withershins beats Rude Wind II 63%, Keystone 63%, and Taking Names 65%, most other programs around 50%. I still can't make up my mind about those step sizes... and I really _hate_ giving up 64% losses to Agony... ;redcode-94m quiet ;name Withershins Thrice ;kill Withershins Thrice ;author P.Kline ;strategy once-thru 80% bomb/scan with multi-pass core-clear ;macro step1 equ 11 step2 equ 12 step equ (step1+step2) dat #gate-10,clr-gate+5 ; just in case clr is decremented gate dat #gate-10,scan-(3*step2)-5 wipe4 dat #gate-10,<2667 wipe3 dat #gate-10,clr-gate+5 wipe2 spl #4000 ,clr-gate+5 ; redundant wipers wipe1 spl #5000 ,clr-gate+5 for 4 dat 0,0 rof bomb dat -step,-step dat 0,0 next add inc,@2 ; scanner bombs two and scans two locations mov bomb,@scan ; in a five-instruction loop mov bomb,*scan scan sne.x *150-step1-step2,*150-step1 djn next,#7900/(2*step) add.f scan,gate ; position for wiping inc spl #2*step,#2*step clr mov @1,>gate djn.b clr,{wipe1 ; self-decrementing pointer for 52 dat 0,0 rof for 11 dat 1,1 ; small decoy rof end next+1 Paul Kline pk6811s@acad.drake.edu