Issue 31 May 27, 1996
______________________________________________________________________________
Core Warrior is a weekly newsletter promoting the game of corewar. Emphasis
is placed on the most active hills--currently the '94 draft hill and the
beginner hill. Coverage will follow where ever the action is. If you have
no clue what I'm talking about then check out these five-star internet locals
for more information:
FAQs are available by anonymous FTP from rtfm.mit.edu as
pub/usenet/news.answers/games/corewar-faq.Z
FTP site is: ftp.csua.berkeley.edu /pub/corewar
Web pages are at:
http://www.stormking.com/~koth ;Stormking
http://www.ecst.csuchico.edu/~pizza/koth ;Pizza
http://pauillac.inria.fr/~doligez/corewar/ ;Planar
Newbies should check the stormking page for the FAQ, language specification,
guides, and tutorials. Post questions to rec.games.corewar. All new players
are infinitely welcome!
If ftp.csua.berkeley.edu is unreachable, you can download pMARS at:
Terry's web page--http://www.infi.net/~wtnewton/corewar/
Planar fpt site--ftp://ftp.inria.fr/INRIA/Projects/para/doligez/cw/pmars
______________________________________________________________________________
Greetings.
Impfinity failed in breaking the 2000 wall, falling very near the goal line,
sorry Planar ;-) Now Therrmite is in sight of the red line. Will it reach
the magic number or fall prey of new stones and qscan detectors? we'll know
that next week.
Ian Oversby had to kill his warrior to avoid flooding his mailbox during
vacations.
To have the chance to change ;redcode attribute in quiet could be the
solution to avoid such things, and we'll appreciate it; unless, of course,
it requires a full rewriting of scripts.
Two warriors in this issue: recently fallen Hall of famer Flurry, by Anton
Marsden and Rosebud, by your favourite author.
--Beppe Bezzi
______________________________________________________________________________
Current Status of the Internet Pizza Server ICWS '94 Draft Hill:
Hill Specs:
coresize: 8000
max. processes: 8000
duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
minimum distance: 100
rounds fought: 200
instruction set: ICWS '94 Draft
# %W/ %L/ %T Name Author Score Age
1 48/ 41/ 11 d-clear bjoern guenzel 155 5
2 42/ 36/ 21 Blur Anton Marsden 148 281
3 35/ 23/ 42 the historian bjoern guenzel 148 8
4 44/ 40/ 16 Scan Man David van Dam 148 294
5 39/ 33/ 28 stoninc Maurizio 146 128
6 43/ 42/ 15 Violent Micro v0.4 basehead 144 139
7 37/ 30/ 33 Osseil John K W 144 34
8 31/ 18/ 51 Rosebud Beppe 143 208
9 28/ 13/ 59 EIF John K W 143 82
10 43/ 43/ 15 HopScotch 1.6 Robert Macrae 143 2
11 37/ 32/ 30 Bomber Boy David van Dam 142 163
12 32/ 22/ 46 blue candle bjoern guenzel 142 122
13 29/ 19/ 52 Test Paper Robert Macrae 139 1
14 43/ 48/ 9 007 David van Dam 138 3
15 37/ 37/ 26 Thermite II Robert Macrae 138 1874
16 38/ 39/ 22 Chameleon M R Bremer 137 1345
17 37/ 38/ 26 Yogi Bear P.Kline 136 25
18 35/ 35/ 30 Grilled Octopus v0.5 David Boeren 136 1090
19 37/ 39/ 24 T.N.T. pro Maurizio Vittuari 135 253
20 38/ 40/ 22 Unknown Anonymous 135 7
21 25/ 16/ 58 Hazy Shade II John K W 134 833
22 35/ 36/ 29 Lithium And Water John K W 133 6
23 39/ 45/ 16 Stepping Stone Kurt Franke 133 687
24 23/ 14/ 63 ompega Steven Morrell 132 135
25 28/ 25/ 47 test rtiqe1 Beppe 132 50
Weekly age: 127 ( 88 last week, 107 the week before )
New warriors: 11 Turnover/age rate 9%
Average age: 311 ( 435 last week, 456 the week before )
Average score: 140 ( 141 last week, 141 the week before )
The top 25 warriors are represented by 13 authors: Wilkinson 5 (including
one as anonymous) van Dam, Guenzel and Mcrae with 3, Vittuari and Bezzi, with 2.
Impfinity is dead. Shortly before reaching 2000 it's fallen off the hill,
losing more than 12 positions in a week; sometimes in the past we spoke of
havoc or carnage; this week I have no word to describe what happened. The
hill lost 30% of his age, including warriors rated 1, 3 and 6 in the what's
old standings, and the standings show all most aged warriors in the lower
half. I cannot remember the imp-ring revolution, but I can guess that what's
happening now is comparable with that, even if I cannot see the reason of
this mass extiction.
After a few week of strong leadership, Scan Man is no more the King;
Guenzel's new warrior d-clear is now in the higher spot, with a good lead
over Blur, the historian, and previous king Scan Man.
Vittuari's stoninc is the only other warrior to hold its position in the top
five.
Apart d- clear a few good warriors have entered the hill: the historian and
Oseesil held king position for a while and also EIF was very near te the
top. Thermite, TNT pro and Chameleon, lose near ten positions from last
week, and the new millenary Grilled Octopus is in the low places of
standings too, 'the new' is advancing and veterans have some troubles to
keep pace.
______________________________________________________________________________
94 - What's New
# %W/ %L/ %T Name Author Score Age
1 48/ 41/ 11 d-clear bjoern guenzel 155 5
3 35/ 23/ 42 the historian bjoern guenzel 148 8
7 37/ 30/ 33 Osseil John K W 144 34
9 28/ 13/ 59 EIF John K W 143 82
10 43/ 43/ 15 HopScotch 1.6 Robert Macrae 143 2
12 32/ 22/ 46 blue candle bjoern guenzel 142 122
13 29/ 19/ 52 Test Paper Robert Macrae 139 1
14 43/ 48/ 9 007 David van Dam 138 3
17 37/ 38/ 26 Yogi Bear P.Kline 136 25
20 38/ 40/ 22 Unknown Anonymous 135 7
22 35/ 36/ 29 Lithium And Water John K W 133 6
25 28/ 25/ 47 test rtiqe1 Beppe 132 50
______________________________________________________________________________
94 - What's No More
# %W/ %L/ %T Name Author Score Age
26 23/ 22/ 55 Impfinity v4g1 Planar 123 1993
26 31/ 38/ 30 Tornado 3.0 Beppe 124 1567
26 28/ 31/ 41 Flurry Anton Marsden 124 835
26 27/ 27/ 47 Walk Like An Egyptian John K W 127 307
26 16/ 13/ 5 Wind-up Toy v1.2c Ian Oversby 53 288
26 34/ 42/ 24 The Core Clear V1.3 David van Dam 127 280
26 1/ 1/ 2 test John K W 4 134
26 3/ 5/ 14 Whisper v0.1 Ian Oversby 25 95
26 0/ 1/ 3 Solomon v0.6 Ian Oversby 4 72
26 10/ 10/ 10 Flimsy v0.4a Ian Oversby 41 66
26 34/ 42/ 24 dummy brain bjoern guenzel 127 44
26 0/ 1/ 3 blue candle bjoern guenzel 3 28
26 37/ 47/ 16 Eggbeater Anton Marsden 127 15
A total age of 5724, yes five thousands seven hundreds twenty four, 30% of
hill age, has been pushed off this week I don't think this record will be
broken too soon.
______________________________________________________________________________
94 - What's Old
# %W/ %L/ %T Name Author Score Age
15 37/ 37/ 26 Thermite II Robert Macrae 138 1874
16 38/ 39/ 22 Chameleon M R Bremer 137 1345
18 35/ 35/ 30 Grilled Octopus v0.5 David Boeren 136 1090
21 25/ 16/ 58 Hazy Shade II John K W 134 833
23 39/ 45/ 16 Stepping Stone Kurt Franke 133 687
Lost Impfinity, Tornado, Flurry and Walk like an Egyptian. No new entries.
Grilled Octopus enters the 'over 1000' club. All veterans are in the lower
half of standings.
______________________________________________________________________________
HALL OF FAME
* means the warrior is still active.
Pos Name Author Age Strategy
1 Impfinity v4g1 Planar 1993 Stone/ imp
2 Thermite II Robert Macrae 1874 * Qscan -> bomber
3 Jack in the box Beppe Bezzi 1620 P-warrior
4 Tornado 3.0 Beppe Bezzi 1567 Bomber
5 Torch t18 P.Kline 1539 Bomber
6 Frontwards v2 Steven Morrell 1420 One shot scanner
7 Chameleon Myer R Bremer 1345 * P-warrior
8 Evol Cap 6.6 John Wilkinson 1299 Imp / stone
9 quiz Schitzo 1262 Scanner/ bomber
10 T.N.T. Maurizio Vittuari 1204 Bomber
11 Grilled Octopus v0.5 David Boeren 1090 * P-warrior
12 Iron Gate 1.5 Wayne Sheppard 926 CMP scanner
13 Agony II Stefan Strack 912 CMP scanner
14 Barrage Anton Marsden 876 Qscan -> replicator
15 Blue Funk Steven Morrell 869 Stone/ imp
16 Flurry Anton Marsden 835 Qscan -> pwarrior
17 Hazy Shade II John Wilkinson 833 * P-warrior
18 Thermite 1.0 Robert Macrae 802 Qscan -> bomber
19 Blue Funk 3 Steven Morrell 766 Stone/ imp
20 Night Train Karl Lewin 755 Replicator
21 Mirage 1.5 Anton Marsden 736 Scanner/ bomber
22 Blizzard Anton Marsden 713 Qscan -> replicator
23 Stepping Stone Kurt Franke 687 * Qscan -> Vampire
24 HeremScimitar A.Ivner,P.Kline 666 Bomber
25 La Bomba Beppe Bezzi 650 Qscan -> replicator
Stepping Stone by Kurt Franke pushes off Paulsson's myVamp.
Thermite II is viewing the 2000 borderline, Tornado stops its run in fourth
place, there are but five living warriors in the HoF, no new warrior is
scheduled to enter soon in the hall.
______________________________________________________________________________
Current Status of the Internet Pizza Server Beginner's Hill:
Hill Specs:
coresize: 8000
max. processes: 8000
duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
minimum distance: 100
maximum age: At age 100, warriors are retired.
rounds fought: 200
instruction set: ICWS '94 Draft
# %W/ %L/ %T Name Author Score Age
1 57/ 31/ 12 Violent Micro v0.4b basehead 184 28
2 54/ 32/ 14 Sapper Andrew Fabbro 176 2
3 50/ 30/ 20 scan test bjoern guenzel 170 56
4 49/ 33/ 17 coal 3.22b bjoern guenzel 166 99
5 45/ 29/ 26 Toxin IVa Edgar 161 71
6 49/ 36/ 15 radar.b John K. Lewis 161 25
7 46/ 31/ 23 Trebuchet Andrew Fabbro 161 9
8 44/ 30/ 26 Toxin IVb Edgar 157 58
9 41/ 25/ 34 Fork v0.1-10p (f) Christoph C. Birk 156 12
10 44/ 33/ 23 Extremely Prejudiced Scott Manley 154 3
11 45/ 40/ 15 8-Ball 2 Iain Hogg 149 89
12 47/ 45/ 8 Drill 1.1 Edgar 149 73
13 42/ 36/ 22 Toxin IV Edgar 148 95
14 43/ 39/ 18 Bloodhound Mk III Andrew Fabbro 147 66
15 43/ 42/ 15 8-Ball Iain Hogg 144 93
16 41/ 39/ 20 pipy-2 Edgar 143 18
17 41/ 41/ 18 pipy-1 Edgar 141 19
18 33/ 32/ 35 Szyzyg v1.01 Scott Manley 135 83
19 41/ 50/ 9 Micro Centurion (mod) Edgar 131 30
20 31/ 35/ 34 Szyzyg v1.0 Scott Manley 127 84
21 22/ 29/ 48 invicta John K. Lewis 116 78
22 26/ 39/ 36 Hunter V 0.7 beta O.Fechner 113 1
23 34/ 55/ 11 Hide'n'Seek debug Edgar 113 42
24 25/ 39/ 36 Hunter V 0.5 beta O.Fechner 111 5
25 24/ 48/ 28 bomberQ2 harleyQ2 99 6
Low traffic also this week, but 35 challenges. Basehead new Violent micro,
a version of which is 6th in the pro hill leads, followed by Fabbro's Sapper.
______________________________________________________________________________
The Hint
Rosebud by Beppe Bezzi
Few weeks ago I had a sort of feeling that there were something wrong in the
hill; no replicators, very many scanners, complex bombers and pspacers.
Having never made a stone before, I call 'stones' but the four five lines
bombers with almost everything inside the main loop, Torch and Tornado are
'bombers' not stones, for me, I decided to try coding one, and results have
been very good.
Five liners are, IMO perhaps the most difficult warriors to make; being
rather slow they have to be very small and make the maximal use of
decrements and the stream to increase attack power.
Stones gains most of their points against comples scanners and bombers,
having a much better size /speed ratio, the main enemies for them are
replicators, obviously, and one shot scanners, because of he small footprint
and the fact that dat 0,0 bombs and decrements aren't very effective against
them. Another problem is inserting a core clear to get rid of self splitting
bombers once disabled.
There is little to invent here, so having to start from something I started
from my favourite, Blue Funk, a warrior I like a lot for his elegance, and
using the same structure I built my warrior.
The stone part is the same of BF, with one more leading spl 0 and a
different mod 5 step. The extra spl is needed because I use an heavier
spiral and because I don't want my stone slowed if imps are stunned.
clr spl #0, 0
split spl #-step, <step
stone mov >-step, step+1
add split, stone
cnt djn.f stone, <djnoff
stptr dat 0,0
dat 0,0
cc dat 0, #stone-4
To better see how Rosebud works use this simple macro, after tracing 'ston'e
line:
st01= @go~if (a!=0 && a!=5)~!~l pc-5,pc+5
it stops when the warrior is executing a mov >0, xx or a mov >cc, xx
The basic is that when we execute a mov >0, xx line we copy xx over itself
and we begin executing another bombing with the pattern shifted one forward,
the cc line is needed because we have to avoid self bombing when executing
the mov >5, yy that after some time cause self destruction. All the merit of
that goes to Steven Morrell, I only fitted his idea in my warrior.
Worth notice is the use of the stream; djnoff and the decrement of split
line both point the same cell, so the stream is interrupted; this has some
advantages: A- can more easily trigger a cmp scan. B- covers a wider area in
the same time -C last but not least, can decrement a warrior without
triggering his decrement detector, usually a
jmz.b loop, #0
When everything goes well, Rosebud will bomb its djn pointer a bit before
the stream reaches itself, restarting it from the beginning and can complete
two full bombings of core before self decrementing and self destructing;
self destruction comes very late in the round, with but 3000 cycles to do.
The imp spiral is a 16 process, jump add launch. The launcher is immediatly
booted away not to give Thermite a free lunch, and uses mov #2667, *0 imps.
Further improvements I planned are: moving spiral placement to avoid losing
that many processes during my bombing (I lose 4) and adding something, don't
know yet what, to kill enemy spirals once my stone killed the partner
warrior. Anyway I'm rather satified of my nice little thing, even if a bit
surprised of how well it does.
;redcode-94
;name Rosebud
;author Beppe
;strategy a nice little thing
;strategy and a bunch of imps
;assert CORESIZE == 8000
;kill Rosebud
step equ -845
djnoff equ split+step
impsize equ 2667
impdspl equ away+impsize+impsize/3+impaway
impaway equ 1000 ;more or less :-)
away equ 4000 ;more or less :-)
org boot
boot
spl impboot
mov cc, @stptr1
for 5
mov {stptr, {stptr1
rof
spl *stptr1,<4000
stptr1 mul.f #split+away,#split+away+2
x dat 0,0
dat 0,0
start
clr spl #0, 0
split spl #-step, <step
stone mov >-step, step+1
add split, stone
cnt djn.f stone, <djnoff
stptr dat 0,0
dat 0,0
cc dat 0, #x
for MAXLENGTH-CURLINE-23
spl #stptr,impptr
rof
impboot
a for 8
mov <impptr,{impptr
rof
impptr jmp vector+impaway,vector
imp mov.i #impsize,*0
mov imp-impaway,imp+impdspl
spl 1,<1000
spl 1,<1100
spl 1,<1200
spl 1,<1300
spl 2,1400
jmp imp+impdspl
add.a #impsize,-1
;-break
vector
dat 0,0
for MAXLENGTH-CURLINE
dat 0,0
rof
end
______________________________________________________________________________
Extra Extra
Flurry by Anton Marsden.
Flurry is a warrior I put together in about 15 minutes. I don't like
it. But it seems to have done well on the Hill. Surprisingly, it remained
in the lower half of the Hill for most of it's life. As the strategy line
says, it's a QScan followed by either CIA (intelligent bomber) or Barrage
(brainwashing paper).
The QScan bombs a wide area and takes far too much time. Using very
simple "logic", CIA or Barrage is executed. The "logic" uses only PSpace
location 0, ie. you can't brainwash it.
Here's Flurry:
;redcode-94
;name Flurry
;author Anton Marsden
;strategy QS -> Barrage | CIA
;assert CORESIZE==8000
;kill Flurry
ORG start
STEP equ 3119
A equ (c-1)
SPYOFF equ STEP*1400
eye equ (A-SPYOFF)
spy equ (A+SPYOFF)
gate:
spy0: dat.f >-SPYOFF,>-SPYOFF
dat.f $-1,$15
cb spl.ab #-2,$15
dat.f $0,$0
dat.f $0,$0
s spl.a #spy0,>4000
loop add.f $c,@2
ip mov.i @1,@spy
mov.i @0,*spy
jmz.b $loop,$eye
c spl.b #STEP,>-STEP
mov.i @2,>gate
mov.i @1,>gate
b0 djn.b $c+1,{cb
FOR 19
dat.f $0,$0
ROF
start:
q FOR 4
sne.i $s+400*q,$s+400*q+100
seq.i $s+400*q+200,$s+400*q+300
mov.ab #s+400*q-f,$f
ROF
jmn.b @0,$f
q FOR 4
sne.i $s+400*(q+8),$s+400*(q+8)+100
seq.i $s+400*(q+8)+200,$s+400*(q+8)+300
mov.ab #s+400*(q+8)-f,$f
ROF
jmn.b @0,$f
q FOR 4
sne.i $s+400*(q+4),$s+400*(q+4)+100
seq.i $s+400*(q+4)+200,$s+400*(q+4)+300
mov.ab #s+400*(q+4)-f-100,$f
ROF
jmz.b $sw,$f
add.ab #100,$f
f: sne.i $d,$0
djn.b $-2,#4
mov.i $b,@f
mov.ba @1,$f
add.f $inc,$f
mov.i *1,@f
mov.i $b,*f
djn.b $-3,#12
ptr: spl.a *1,>2*2667
mov.i *s,$spy
mov.i $1,$-1
b:
inc: dat.f >8,>-8
sw: ldp.ab #0,#0
jmz.f $ptr,<sw
ASTEP EQU 3510; mod-10
BSTEP EQU 2365; mod-5
CSTEP EQU 3315; mod-5
p: spl.b $1,>2000
spl.b $1,>4000
spl.b $1,>6000
s1: spl.b @s1,}ASTEP
mov.i }s1,>s1
s2: spl.b @s2,}BSTEP
mov.i }s2,>s2
mov.i {s2,<s3
s3: jmp.b @s3,>CSTEP
dat.f >1,}1
dat.f }1,>1
stp.b >0,#1
jump: jmp.b $-1,$1
d:
END
______________________________________________________________________________
Questions? Concerns? Comments? Complaints? Mail them to people who care.
authors: Beppe Bezzi <bezzi@nemo.it> or Myer Bremer <bremermr@ecn.purdue.edu>
|
