;redcode-94 ;name Confused Moth ;author Philip Thorne ;strategy John Metcalf's Spring/Summer 2002 Tournament ;strategy Round 4 [Extinction] Entry #2 Helicoverpa confusa ;strategy ;strategy Switch-on-loss P-Spacer: Scanner & Stone, both booting ;strategy Scanner is Jinx by Christian Schimdt ;strategy Stone is from Recycled Bits by David M Moore ;strategy Brain is from noideas by Ilmari Karonen ;strategy My minor contribution is the mostly dual, but inefficient ;strategy [by 3 instructions each component] boot. I had intended ;strategy a vamp component too for occassional/varied brain-washing. ;strategy ;strategy My benchmark has: CM=165, Jinx alone=172, RecBits=141 ;strategy ;strategy The name summarises my approach to this round. ; ;assert (CORESIZE==8000) ;assert MAXLENGTH==50 ;noideas pbrain [Ilmari Karonen]. ; ; *** switch-on-loss p-brain *** ; 5 cycles on win/tie, 6 on loss pflag equ 248 ; If bw'd to self will give scanner - hope ; is that brain-wash most likely by paper. num equ 2 ; fast ; pflag must not be the a-field of num-1 instructions after switch! first: ldp.ab #pflag, switch ldp.a #0, switch mod.ab #num, @first switch sne.a >0, 0 stp.ba switch, first djn.b wB, @first ; asymmetric if num>2 ;=== boot BOOT equ (-2532) ;Stone boot wA ddd for 10 mov.i 1 ;== dat 0, 0 dat 0, 0 ;=== Jinx scanner [Christian Schmidt] ;10lines, start at top+1 ;This is the starting warrior scanner: head: spl #-4, 3000 loop: mov head, >head ;start here add.f step, scan scan: seq }4, 0 mov.b scan, @loop djn loop, #950 step: spl #-152, >-152 mov clr, >head-5 djn.f -1, >head-5 clr: dat 1, #18 ;== dat 0, 0 dat 0, 0 ;== ;Scanner boot. SBOOT equ (-1532) sBoot: spl SBOOT+(sBoot-boot), scanner+10+(sBoot-boot) wB: mov.i head, (sBoot-wB) + SBOOT-10 mov.i sBoot, boot jmp boot-9