From news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!in2.uu.net!01-newsfeed.univie.ac.at!newsfeed.ACO.net!CARNet.hr!news Fri Jun 7 11:34:15 1996 Article: 5370 of rec.games.corewar Path: news-rocq.inria.fr!jussieu.fr!math.ohio-state.edu!howland.reston.ans.net!newsfeed.internetmci.com!in2.uu.net!01-newsfeed.univie.ac.at!newsfeed.ACO.net!CARNet.hr!news From: Edgar Newsgroups: rec.games.corewar Subject: (no subject) Date: 7 Jun 1996 08:23:51 GMT Organization: CARNet, CROATIA Lines: 51 Distribution: world Message-ID: <4p8oun$ccf@bagan.srce.hr> NNTP-Posting-Host: pc20.mef.hr Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Mailer: Mozilla 1.1N (X11; I; Linux 1.2.13 i486) X-URL: news://bagan.srce.hr/rec.games.corewar Hello. (since this is just my second post to this newsgroup please don't be too harsh with the flames. Thank you.) Being stuck with papers for a while, I developed one that did rather well,namely Toxin. It floated somewhere in the top 10 of the b-hill for most of its lifetime , and IMHO it was doing rather good,but since the last version I haven't been able to improve its scores. I tried to make a paper that would be aggressive instead of one that was only surviving , so after each copy does its bombing (3*8 locations) it dies, thus releasing the processes and making the copies that are still active execute faster. The improvements I managed to include increased the overall score of the warrior somewhat, but the wins were also traded for draws. The constants are also not optimal , as I used the first ones that proved to cover the core well. I am thankful for any suggestions. ;redcode-b ;name Toxin IVa ;author Edgar ;strategy silk ... shorter boot, new constants ;assert CORESIZE==8000 step1 equ 3217 step2 equ 2701 org boot false for 25 cmp -false,-false*3 rof boot spl 1,}2000 spl 1,}4000 spl 1,}6000 mov -1,}-1 spl step2,0 mov >-1,}-1 mov bom,}2000 mov bom,}4000 mov bom,}6000 bom dat <1,<1 decoy for (MAXLENGTH-CURLINE) nop decoy*2,decoy*3 rof