LP hill warrior: Mad LP

Archived from groups: rec.games.corewar (More info?)

Hi all,

here is another p-warrior for the LP hill. Nothing really
sensational in it. The stone/paper is a copy of Roskilde
and the scanner is an LP optimized version of Recon 2.

Christian


;redcode-lp
;name Mad LP
;author Christian Schmidt
;strategy scanner, stone/paper
;assert 1

step equ 6557
ptr equ (scan-8)

dat 19, 19
diff jmp #-step, #-step
spl #0, 0
wipe mov diff, >ptr
w2 mov *wipe, >ptr
djn.a wipe, length

loop sub diff, @s2
scan sne (step*2) - 1, (step*2) - 7
sub diff, scan
s2 seq *scan, @scan
slt.a #20, scan
timer djn loop, #1400


length sub.ba #0, #-7
tweak mov.ab @s2, @wipe
t2 jmn *w2, timer
djn.a <tweak, {t2

for 75
dat 0,0
rof

switch ldp.ba what, what
mod.a #3, what
ldp.ab #0, switch
add.a @switch, what
stp.ab *what, what
what jmp @0, scan
dat 1, scan
dat 0, scan
dat 1, scan
dat 1, scan
dat 2, xGo
dat 0, scan
dat 1, scan

for 82-5
dat 0,0
rof

sAwa1 equ 6897
sAwa2 equ 2842
sAwa3 equ 4582
sStep equ 1564
sStep2 equ 383

pStep1 equ 2990
pStep2 equ 1719

xGo spl pGo
qtab2 spl 2
spl 1

mov <sa1, {sa1
mov <sa2, {sa2
mov <sa3, {sa3
jmp *sa1, }0

sa1 dat pa1+sAwa1, sEnd+1
sa2 dat pa1+sAwa2, sEnd+1
sa3 dat pa1+sAwa3, sEnd+1

mov <sStep2 , 1+sStep
sub 1, -1
sEnd jmp -2, <-sStep

pa1 mov #6, #6
pGo mov <pa1, <j1
jmn -1, @-2
spl @pa1, <pStep2
j1 jmz.b @0, pStep1
b1 dat {1, <1

end switch