From news-rocq.inria.fr!jussieu.fr!mathserv.mps.ohio-state.edu!uwm.edu!spool.mu.edu!howland.erols.net!feed1.news.erols.com!insync!www.nntp.primenet.com!nntp.primenet.com!swrinde!newsfeed.internetmci.com!cdc2.cdc.net!news.texas.net!!tarquin Mon Oct 28 11:22:01 1996 Article: 6369 of rec.games.corewar Path: news-rocq.inria.fr!jussieu.fr!mathserv.mps.ohio-state.edu!uwm.edu!spool.mu.edu!howland.erols.net!feed1.news.erols.com!insync!www.nntp.primenet.com!nntp.primenet.com!swrinde!newsfeed.internetmci.com!cdc2.cdc.net!news.texas.net!!tarquin From: tarquin@texas.net Newsgroups: rec.games.corewar Subject: Grr, My guys never do well :( Date: 25 Oct 1996 21:56:15 GMT Organization: I need to put my ORGANIZATION here. Lines: 80 Message-ID: <54rd1v$53m@news2.texas.net> NNTP-Posting-Host: dnet06-32.austin.texas.net Originator: tarquin@ Every "improvement" I made to Jolt made it score worse :( I went from scan->clear, added a boot, and then tried running the clear concurrently. Here's the last ver I sent to the hill as a test. The scores went from 126/112 (sent v 0.1 recently) to 103 to 62(!?) It's a .80c cmp scanner dropping spl/spl/jmp bombs and goes into a spl/dat coreclear. I tried putting on a boot to help against qscanners, didn't seem to do much. I tried running the coreclear simultaneously so a lucky hit wouldn't kill it (Is there a way to self-split a scanner? I couldn't figure one out) v 0.1 did the best of the three (actually would've gotten 24th if I had sent it to the hill instead of a test) but the second time around it didn't even make it. Here is v 0.3. The concept doesn't seem to bad, but I'm a beginner and coded it badly (uses a djn to tell if it's done, self-bombing was a bit confusing, plus the scanner+clear is ~24 instructions) If you excellent people at CoreWarrior want to do anything with this, go ahead. ;redcode-b test ;name Jolt v0.3 ;author J.A-Worth ;strat boot->scan->clear ;strat v 0.1-original, scan->clear ;strat v 0.2-added boot ;strat v 0.3-parallelized clear ;strat 126->107->62?! special equ 48 org boot bdest equ 4123 // not real one, though this probably won't help anybody. clearp dat #bomb, #100 target dat #100, #112 next add step, target check sne *target, @target add step, target seq *target, @target slt.ab #special, target djn next, #1000 attack mov bomb,}target // yeah, its huge, and it bombs both spots mov bomb,>target mov bomb+1,}target mov bomb+1,>target mov bomb+2,}target mov bomb+2,>target add.f bomb+1, target jmn.b next, attack-1 clear spl #4, #0 mov *clearp, >clearp step dat #42, #42 dat #18,#40 bomb spl #18,#93 spl #-3,>-3 jmp -2, <-3 a for 44 // big decoy, but my scanner will bomb it too dat a,45-a rof bptr dat bomb+3,bdest boot mov {bptr,