info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
[Tiny] Yukikaze
39 views
Skip to first unread message
inversed
Jul 27, 2015, 10:52:50 AM7/27/15
to
;redcode-tiny
;name Yukikaze
;author inversed
;strategy Stone + Imp
;assert CORESIZE == 800
;date submitted: 2015.01.06
; The stone is 1/3c, with the mov.i #1, {1 bomb being a part of the loop. The
; idea was taken from Dwarfia Hyemalis, which was evolved with RedShift from
; a silkdwarf-seeded population. The stone may be hard to work with because
; there is no clear bombing pattern. Sometimes the mov executes twice in a row,
; ruining the global pattern alignment. But thanks to this feature, the stone
; can cover the core almost completely when properly tuned.
; Silk-based imp launcher a-la Silent Snowfall is employed. But there is a
; weird modification discovered by accident. Djn to the silk decrements itself
; on its second execution. As a result, the imp launcher is fed with only 2
; processes. The resulting "spiral" has 11 processes.
step equ 615
sofs equ 611
istep equ 267
bds equ 586
bdi equ 66
stream equ 149
scratch equ 19
qba equ x0+qao+(i-1)*qas
qbb equ x0+qbo+(i-1)*qbs
qao equ 354
qbo equ 762
qco equ 212
qas equ (-20)
qbs equ (-23)
qcs equ (-117)
x0 equ (-CURLINE)
org imp+1
silk spl @ 4 , < istep+1
mov } silk , > silk
spl 2*istep , } 0
imp mov.i # istep/2 , istep
i for 6
mov { qba , qbb
rof
spl 1 , { x0+qco+0*qcs
spl 1 , { x0+qco+1*qcs
mov { silk , { bpi
mov { stone , { bps
bps spl x0+bds+4 , { x0+qco+2*qcs
bpi djn.f x0+bdi+4 , { xx
stone spl # 4 , { stream
loop mov.i # 1 , { 1
add.ba # sofs , # step+1
xx jmp loop , > stone
;name Yukikaze
;author inversed
;strategy Stone + Imp
;assert CORESIZE == 800
;date submitted: 2015.01.06
; The stone is 1/3c, with the mov.i #1, {1 bomb being a part of the loop. The
; idea was taken from Dwarfia Hyemalis, which was evolved with RedShift from
; a silkdwarf-seeded population. The stone may be hard to work with because
; there is no clear bombing pattern. Sometimes the mov executes twice in a row,
; ruining the global pattern alignment. But thanks to this feature, the stone
; can cover the core almost completely when properly tuned.
; Silk-based imp launcher a-la Silent Snowfall is employed. But there is a
; weird modification discovered by accident. Djn to the silk decrements itself
; on its second execution. As a result, the imp launcher is fed with only 2
; processes. The resulting "spiral" has 11 processes.
step equ 615
sofs equ 611
istep equ 267
bds equ 586
bdi equ 66
stream equ 149
scratch equ 19
qba equ x0+qao+(i-1)*qas
qbb equ x0+qbo+(i-1)*qbs
qao equ 354
qbo equ 762
qco equ 212
qas equ (-20)
qbs equ (-23)
qcs equ (-117)
x0 equ (-CURLINE)
org imp+1
silk spl @ 4 , < istep+1
mov } silk , > silk
spl 2*istep , } 0
imp mov.i # istep/2 , istep
i for 6
mov { qba , qbb
rof
spl 1 , { x0+qco+0*qcs
spl 1 , { x0+qco+1*qcs
mov { silk , { bpi
mov { stone , { bps
bps spl x0+bds+4 , { x0+qco+2*qcs
bpi djn.f x0+bdi+4 , { xx
stone spl # 4 , { stream
loop mov.i # 1 , { 1
add.ba # sofs , # step+1
xx jmp loop , > stone
inversed
Jul 27, 2015, 11:00:24 AM7/27/15
to
CorePaste link: http://www.corewar.eu/corepaste/id.php?id=32
ccb...@gmail.com
Jul 27, 2015, 2:18:13 PM7/27/15
to
On Monday, July 27, 2015 at 7:52:50 AM UTC-7, inversed wrote:
> ;redcode-tiny
> ;name Yukikaze
> ;author inversed
> ...
> ;redcode-tiny
> ;name Yukikaze
> ;author inversed
Congratulations!
Yukikaze is the new KotH of the 'Tiny-Koenigstuhl'.
http://users.obs.carnegiescience.edu/birk/COREWAR/koenigstuhl.html#TINY
Christoph
0 new messages