From news-rocq!jussieu.fr!oleane!news.maxwell.syr.edu!howland.erols.net!portc02.blue.aol.com!audrey01.news.aol.com!not-for-mail Fri Aug 6 14:45:09 1999 Article: 10504 of rec.games.corewar Path: news-rocq!jussieu.fr!oleane!news.maxwell.syr.edu!howland.erols.net!portc02.blue.aol.com!audrey01.news.aol.com!not-for-mail From: happycan@aol.comjnkkil (Marshall) Newsgroups: rec.games.corewar Subject: Self Repair? Lines: 42 NNTP-Posting-Host: ladder07.news.aol.com X-Admin: news@aol.com Date: 06 Aug 1999 01:04:48 GMT Organization: AOL http://www.aol.com Message-ID: <19990805210448.09545.00001352@ng-bj1.aol.com> Xref: news-rocq rec.games.corewar:10504 Ive made a warrior combining a imp-ring and a silk. It makes 3 silks and sends them through the core. Here is the code: ;name Silk Spiral ;author Marshall ;assert MAXPROCESSES > 11 kick1 EQU 2666 kick2 EQU 5332 step EQU 504 ;this covers the core most efficiently, 50% w/ DAT and 50% with silks (which kill anything with < 4 processes) bomb EQU step+1 ORG splkick boot1 DAT silk, kick1 boot2 DAT silk, kick2 splkick SPL 1 ;\ SPL 1 ;4 processes kick MOV }boot1, >boot1 ;\ MOV }boot2, >boot2 ;makes 2 copies of the first silk SPL kick1-7 ;split to those copies SPL kick2-7 ;/ silk SPL @0, step ;the silk MOV }silk, >silk MOV @kill, }kill kill DAT #bomb, #0 ;if one more process is added, a JMP -1 can be stuck here to make eack silk continue to carpet-bomb it occurred to me that it might be possible make this warrior self repair by scanning where another copy should be with SEQ and if not finding an exact copy laying a DAT carpet over the area (to kill useless stun processes) and writing a new copy there then jumping to the copy. Would this be worth trying, or would having the bigger and slower self repairing silks make the warrior too vulnerable to scanners than if it were small and fast? Small and Fast vs. Big Slow and Tough again. =P. maybe ive just answered my own question... Marshall