From news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!akemi Wed Jan 24 11:22:44 1996 Article: 3629 of rec.games.corewar Newsgroups: rec.games.corewar Path: news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!ix.netcom.com!netcom.com!akemi From: akemi@netcom.com (David Boeren) Subject: Time Lapse (source code) Message-ID: Organization: NETCOM On-line Communication Services (408 261-4700 guest) X-Newsreader: TIN [version 1.2 PL1] Date: Wed, 24 Jan 1996 00:00:06 GMT Lines: 60 Sender: akemi@netcom12.netcom.com OK, Time Lapse v0.1 has been pushed off the 94 hill, so here's the code for you all to enjoy: ;redcode-94 verbose ;name Time Lapse v0.1 ;author David Boeren ;assert CORESIZE == 8000 ;strategy Bomber -> SPL/DAT coreclear ;strategy v0.1 Fast bomber, does not use stun bombs step equ 184 stepb equ (step*2) stepc equ (step*3) gate equ loop-7 org boot ptr JMP.B #step, #2 DAT.F #16, #0 SPL.B #16, #1 loop MOV.I ptr, *targ ; Tornado bombing with JMP 0's MOV.I ptr, @targ targ MOV.I step, *stepb ADD.F inc, targ DJN.B loop, #333 inc SPL.B #stepc, #stepc clear MOV.I @ptr, }ptr ; Coreclear modified from Phosphorus DJN.F clear, 100, >101 ; die in a core-coloring way for (MAXLENGTH-CURLINE) / 4 DAT.F loop, clear ; big decoy to confuse scanners DAT.F targ, boot ; No two are alike! DAT.F boot, targ DAT.F clear, loop rof END Basically, it's like Tornado, but using JMP 0 bombs instead of DATs and refitted with a SPL/DAT coreclear. I wanted something that could do reasonably well against papers, which Tornado gets clobbered by. My theory was that by using JMP 0's for bombs then at least the enemy would not get his processor time back when a process got hit. That, plus the SPL coreclear pass helps me get a reasonable score versus papers. Although I do get more points against papers, it often happens that my coreclear will get hit by bombers and scanners, making me unable to finish off these single-process warriors who are just sitting there helpless on my JMP 0. Obviously that means that I get more ties when I would have gotten a win if I'd used a DAT. Anyway, although it was not a wild success, it performs reasonably well for a beginner's program.