From news-rocq.inria.fr!jussieu.fr!univ-lyon1.fr!howland.erols.net!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!news.maxwell.syr.edu!news.idt.net!nntp.farm.idt.net!rotten.com!franz Fri Mar 7 18:41:50 1997 Article: 7217 of rec.games.corewar Path: news-rocq.inria.fr!jussieu.fr!univ-lyon1.fr!howland.erols.net!cam-news-hub1.bbnplanet.com!news.bbnplanet.com!news.maxwell.syr.edu!news.idt.net!nntp.farm.idt.net!rotten.com!franz From: George Newsgroups: rec.games.corewar Subject: Evolution, it works Date: Fri, 7 Mar 1997 02:43:01 -0700 Organization: IDT Lines: 96 Message-ID: NNTP-Posting-Host: 199.245.121.176 Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Sender: franz@rotten.com after 52 generations i now have a warrior that does twice as good its generation brother ... it can now do 80 points on the hill so maybe the 100th or so generation migth get on the hill so here's the little wonder, it's not too good with scanners as it's too long .. but otherwise it's scores are fairly balanced against all warriors ... this one scored above 1000 in my evolution algorithm we'll see if this improves by morning, the main difference on this warrior is that it does not die ... watch the djn instruction which catches all processes ;redcode-94x ;name Warrior #638 / Generation #52 ;author Evolution ;strategy ??? Random ??? ;assert CORESIZE==800 JMZ.AB {755,$13 NOP.AB {391,$15 JMN.F #450,>780 SPL.AB #616,$664 SEQ.AB {789,$685 DJN.BA $261,$799 MOV.AB <794,{404 JMZ.AB @294,>16 SLT.AB <799,#3 SNE.A }14,@80 SPL.A *18,$4 SNE.A }14,@80 SPL.BA }168,<0 NOP.X #797,@18 MUL.I *782,}786 SPL.AB #616,$664 SPL.AB #616,$664 SPL.AB #616,$664 MOV.I >7,}4 DJN.F #795,{505 END 9 well although mov djn and spl are the main instructions producing the success of this warrior the per instruction stats don't seem that way .... btw mov popularity has dropped considerably (from about 2500 to what it is now) For Generation #53 SPL : 4943 MUL : 1576 JMZ : 1444 DJN : 1295 SNE : 1258 MOV : 1231 NOP : 1230 ADD : 990 JMN : 856 SUB : 745 JMP : 571 SEQ : 534 DAT : 522 STP : 436 DIV : 356 MOD : 90 LDP : 70 Total = 18412 on the other hand look at how my red directory looks like JMP : 2934 MOV : 2877 SEQ : 2528 SPL : 2216 DAT : 1933 ADD : 1024 DJN : 685 SNE : 300 SUB : 195 JMZ : 140 NOP : 71 JMN : 70 MOD : 63 STP : 50 DIV : 42 LDP : 34 MUL : 8 i must jump quite a lot :) and seqs are from all the qscans, although most of the warriors on this statistics are mine but it's not an accurate aproximation since tehre are many versions of some warriors but if tehre are many versions then logically i probablyw orked on it more so it deserves more attention in the stats :) and sorry no total i'm lazy to type that command in (a bit of awk would do) well .. whatever Franz