Issue 54 20 January, 1997
______________________________________________________________________________
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: (Please note new Stormking's address)
http://www.koth.org/ ;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 ftp site--ftp://ftp.inria.fr/INRIA/Projects/para/doligez/cw/pmars
Fechter ftp site--ftp://members.aol.com/ofechner/corewar
A collection of Bezzi's hints in the first issues is available at:
ftp://ftp.volftp.vol.it/pub/pc/msdos/games/solutions/bbhints.zip
Beppe Bezzi web page - http://www.aspide.it/freeweb/Bezzi
______________________________________________________________________________
Greetings.
Happy New Year everyone. Hopefully 1997 will be an active, progressive year
for corewar. It's been about a month since last issue so most of the stats
are for four weeks instead of one.
Also be sure to check Kline's article in the hints section. Paul is always
quick to exploit the composition of the hill and find a niche. RetroQ,
currently sitting pretty on top of the hill, does battle with all those nasty
Q^2 scans dominating the hill.
--M R Bremer
______________________________________________________________________________
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 37.4/ 20.8/ 41.8 RetroQ P.Kline 154.1 4
2 33.4/ 17.7/ 48.9 Return Of The Jedimp John K W 149.1 204
3 38.2/ 27.4/ 34.4 Gigolo Core Warrior staff 149.1 258
4 37.9/ 29.3/ 32.8 Nine Seven Six M R Bremer 146.5 68
5 34.4/ 23.5/ 42.1 unrequited love kafka 145.3 302
6 41.9/ 38.6/ 19.4 NCC-1701-A Philip Kendall 145.2 7
7 45.6/ 47.8/ 6.6 He Scans Alone P.Kline 143.4 19
8 32.7/ 23.4/ 43.8 Trident^2 John K W 142.0 134
9 39.8/ 37.8/ 22.4 Solomon v0.8 Ian Oversby 141.8 50
10 36.8/ 36.6/ 26.6 TEST FFF mod10 full Franz 136.9 1
11 29.3/ 22.1/ 48.6 Impish v0.2 Ian Oversby 136.5 243
12 33.1/ 29.7/ 37.2 Terkonit 0.4 Christian Schmidt 136.4 10
13 35.0/ 34.0/ 31.0 Fast Fast Fast v2 Franz 136.1 18
14 38.6/ 41.4/ 20.0 The Machine Anton Marsden 135.7 110
15 37.9/ 41.6/ 20.5 Instant Wolf Edgar 134.1 25
16 36.9/ 41.2/ 22.0 Scanitator 4 Christian Schmidt 132.6 6
17 36.5/ 40.8/ 22.7 Damage Incorporated Anton Marsden 132.1 245
18 38.1/ 44.5/ 17.4 Oblivion Ian Sutton 131.7 69
19 40.2/ 49.2/ 10.6 Memories Beppe 131.2 118
20 40.7/ 51.1/ 8.3 ScanKiller Christian Schmidt 130.3 2
21 39.2/ 48.3/ 12.5 Wallung Christian Schmidt 130.1 8
22 37.6/ 45.6/ 16.8 Probe Anton Marsden 129.6 382
23 34.6/ 39.8/ 25.6 Tiberius 3.1 Franz 129.3 121
24 36.9/ 45.0/ 18.1 Blur 2 Anton Marsden 128.8 381
25 35.8/ 42.9/ 21.3 Silver Talon 1.2 Edgar 128.7 91
Weekly age: 15 this week ( 43 last issue, 22 the issue before )
New warriors: 7 Turnover/age rate 47%
Average age: 115 ( 119 last issue, 91 the issue before )
Average score: 138 ( 142 last issue, 137 the issue before )
The top 25 warriors are represented by 12 independent authors: Schmidt and
Marsden with 4; Franz with 3; Kline, JKW, Edgar, and Oversby with 2. All
others with one lonely warrior each.
Kline rules the roost, first with He Scans Alone and than with RetroQ. As
usual, his top rated warrior has a comfortable lead over second place Jedimp.
Venerable Blur 2 is in danger of falling of the hill completely.
______________________________________________________________________________
94 - What's New
# %W / %L / %T Name Author Score Age
16 30.0/ 29.2/ 40.8 Terkonit 0.4 Christian Schmidt 130.9 1
2 42.9/ 37.6/ 19.5 NCC-1701-A Philip Kendall 148.2 1
11 37.5/ 38.7/ 23.8 Scanitator 4 Christian Schmidt 136.3 1
16 41.8/ 49.6/ 8.5 ScanKiller Christian Schmidt 134.0 1
1 36.8/ 19.9/ 43.4 RetroQ P.Kline 153.6 1
17 40.2/ 48.1/ 11.8 Wallung Christian Schmidt 132.3 1
10 36.8/ 36.6/ 26.6 TEST FFF mod10 full Franz 136.9 1
New warriors entered the hill in the first and second position, although the
Enterprise has fallen to sixth in recent challenges.
______________________________________________________________________________
94 - What's No More
# %W / %L / %T Name Author Score Age
26 37.1/ 49.8/ 13.1 Scankiller 0.1 Christian Schmidt 124.5 31
26 32.1/ 36.7/ 31.2 YAP Franz 127.5 22
26 1.3/ 1.5/ 1.1 Terkonit 0.3 Christian Schmidt 5.1 7
26 30.7/ 32.1/ 37.2 Falcon v0.3 Ian Oversby 129.4 282
26 33.8/ 42.4/ 23.7 CC Paper 2 Franz 125.2 91
26 1.0/ 1.9/ 1.1 Tsunami v0.5 Ian Oversby 4.1 12
26 31.3/ 35.6/ 33.1 scTest P.Kline 127.1 25
The average age drops as Oversby's Falcon v0.3 falls from the hill.
______________________________________________________________________________
94 - What's Old
# %W / %L / %T Name Author Score Age
22 37.6/ 45.6/ 16.8 Probe Anton Marsden 129.6 382
24 36.9/ 45.0/ 18.1 Blur 2 Anton Marsden 128.8 381
5 34.4/ 23.5/ 42.1 unrequited love kafka 145.3 302
3 38.2/ 27.4/ 34.4 Gigolo Core Warrior staff 149.1 258
17 36.5/ 40.8/ 22.7 Damage Incorporated Anton Marsden 132.1 245
11 29.3/ 22.1/ 48.6 Impish v0.2 Ian Oversby 136.5 243
2 33.4/ 17.7/ 48.9 Return Of The Jedimp John K W 149.1 204
JKW enters the bottom of the list.
______________________________________________________________________________
HALL OF FAME
* means the warrior is still active.
Pos Name Author Age Strategy
1 Thermite II Robert Macrae 2262 Qscan -> bomber
2 Impfinity v4g1 Planar 1993 Stone/ imp
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 Chameleon Myer R Bremer 1437 P-warrior
7 Frontwards v2 Steven Morrell 1420 One shot scanner
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 1154 P-warrior
12 Hazy Shade II John Wilkinson 1102 P-warrior
13 Stepping Stone Kurt Franke 1049 Qscan -> Vampire
14 Rosebud Beppe Bezzi 993 Stone/ imp
15 Iron Gate 1.5 Wayne Sheppard 926 CMP scanner
16 T.N.T. pro Maurizio Vittuari 925 Bomber
17 Agony II Stefan Strack 912 CMP scanner
18 Barrage Anton Marsden 876 Qscan -> replicator
19 Blue Funk Steven Morrell 869 Stone/ imp
20 Flurry Anton Marsden 835 Qscan -> pwarrior
21 Thermite 1.0 Robert Macrae 802 Qscan -> bomber
22 Blue Funk 3 Steven Morrell 766 Stone/ imp
23 Night Train Karl Lewin 755 Replicator
24 Mirage 1.5 Anton Marsden 736 Scanner/ bomber
25 Blizzard Anton Marsden 713 Qscan -> replicator
No changes this week, and I think we won't see any for long time, but we at
the Core Warrior are working on a new hall of fame that addresses testing,
etc. Stay tuned for details or mail me with your opinions.
______________________________________________________________________________
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
Last challenge: Thu Jan 16 17:26:45 PST 1997
# %W / %L / %T Name Author Score Age
1 49.9/ 33.7/ 16.4 Silver Talon 1.1 Edgar 166.1 89
2 48.1/ 33.1/ 18.8 Hexagram J.A.Denny 163.0 8
3 50.4/ 38.5/ 11.1 Pentagram 2.a J.A.Denny 162.3 10
4 40.4/ 18.9/ 40.7 Terkonit 0.4 Christian Schmidt 161.9 5
5 46.2/ 33.4/ 20.4 Flimsy v0.6 Ian Oversby 159.0 9
6 49.3/ 42.2/ 8.5 Scankiller 0.1 Christian Schmidt 156.4 20
7 43.2/ 31.3/ 25.5 Scanitator 3.0 Christian Schmidt 155.1 15
8 40.9/ 29.0/ 30.1 Hammermill 4.0 Matt Lewinski 152.8 95
9 47.5/ 43.6/ 8.9 The Stainless Steel Rat Christian Schmidt 151.5 31
10 45.9/ 40.9/ 13.2 Goothmonger v1.2 Ian Sutton 150.9 33
11 35.8/ 20.8/ 43.4 DemonSpawn J.A.Denny 150.8 18
12 42.3/ 34.3/ 23.5 Versatility 1.7 Ross Morgan-Linial 150.3 36
13 40.8/ 31.5/ 27.6 Apocalypse Matt Lewinski 150.1 66
14 43.4/ 41.6/ 15.0 Mostly Harmless v1.1 Justin Kao 145.3 39
15 41.6/ 39.4/ 19.0 Vivid Radiation 2.0 Matt Lewinski 143.8 57
16 42.6/ 43.0/ 14.5 Flamberge 13.4 Matt Lewinski 142.2 58
17 37.1/ 32.7/ 30.2 Dust 0.7.5 Justin Kao 141.6 35
18 33.8/ 26.2/ 40.0 Quantum Christian Schmidt 141.4 12
19 40.0/ 40.5/ 19.5 Microsoft Office '96 Justin Kao 139.6 85
20 33.7/ 30.6/ 35.7 Heartworm Edgar 136.8 6
21 32.7/ 30.9/ 36.5 Escargot 0.5 Justin Kao 134.4 55
22 37.5/ 46.2/ 16.3 Bash the Rat V1.4 Ian Sutton 128.8 50
23 32.8/ 57.8/ 9.4 Trap94_1 Waldemar Bartolik 107.8 4
24 29.8/ 58.6/ 11.5 Trap94_2 Waldemar Bartolik 101.0 2
25 28.4/ 61.3/ 10.2 Gisela 7110 Andrzej Maciejczak 95.5 1
26 8.9/ 83.0/ 8.1 Forker v1.0 Andrew Bortz 34.8 0
Top 25 Averages:
40.6/ 37.6/ 21.8 143.5 33
______________________________________________________________________________
The Hint
On QScans and CoreWar Strategy.
QScans, particularly Q^2 scans modeled on Probe, are very successful on
the '94 Hill. A couple of tests show why. My little silk-imp program
'dTest' scores 120 points and gets zero wins against other replicators
and imps. Adding a Q^2 scan increased the score by 20 points and obtained
20-30% wins against those other programs. In fact, a dummy program
consisting of the Q^2 scan and a 20-line do-nothing loop scored 52 points,
with 20-30% wins against several programs (40% against Scanitator)!
Probe's Q^2 scan checks 50 core locations in 25 instructions, jumping
immediately to the calc/bomb routine when something is found. His scan
code is about 70 instructions long, making him a good target for other
Q^2 scans. Running the dummy Q^2 program against itself gives about
45% wins to each for a 90% 'find' rate between the two of them, which
is amazing. This means that in a battle between two Q^2 scanning
opponents, nearly 90% of the battles could be decided by the Q^2 scan,
with the two of them exchanging wins at 3 points apiece. No wonder
non-scanners are struggling - by not cooperating with the Q^2 program,
they have to slug it out with "real" strategies :-)
Of course, any time you have several programs doing the same thing on
the Hill, a niche is created which is also an opportunity. The
program 'He Scans Alone' puts a short decoy at minus-100 locations from
the cmp-scanner, which sometimes diverts the Q^2 scanner's attack.
Another approach is to put the Q^2 scan into reverse. Scanning backwards
along the mirrored trail of a normal Q^2 scan will result in the two
scanners finding each other at the same time. Then it's a race to
see who can bomb the other faster, putting a lucky bomb in the right
place before executing the opponent's bomb. Still an even chance
to win or lose.
However, we don't want an even chance, we want more! Either more wins
or fewer losses. A non-cooperating Q^2 scan that gets his share
of wins but withholds them from others :-) Which is where Retro-Q
comes in.
Retro-Q does a mirrored scan of the usual Q^2 so that he and a regular
Q^2 scanner find each other at the same time. Then he splits his attack
into two routines which are located such that the opponent's attack will
not kill both. Now both programs bomb each other, but at least one of
Retro-Q's routines will survive long enough to see his opponent die.
How well does it work? Using Probe's Q^2 scan, Retro-Q scored 42/42
against Probe. Using the RQ^2 scan he scored 58/25. Against CCPaper2
he went from 29/25 to 38/7. Against Gigolo from 21/30 to 30/20.
Of course this only works because Anton's scan routine has been copied
as-is into several programs. Jumbling up the scan order (as opposed
to tweaking the scan 'step') would change the environment for Retro-Q,
robbing him of his advantage. Still, he will continue to have an
advantage against non-redundant post-scan attacks in those matches
where they find each other at the same time, mabye 1-2 times per
200 fights.
If the RQ^2 scan fails to win, the silk-imps are launched. Imps are like:
mov.i #2*spcsz,}-spcsz*1
with multiple processes spaced 'spcsz' locations apart. This means that
the imp is copying itself backward one step as well as forward one step
at the same time, which gives it enormous strength against spl carpets.
By themselves the silk-imps go 50/20 against The Machine and Scanitator.
Since they don't dat-bomb they can only win by surviving and overwriting
the opponent, like a good spiral should :-)
(The term "Retin-A" comes from a wrinkle cream that came out a while back)
;redcode-94
;name RetroQ
;kill RetroQ
;author P.Kline
;assert CORESIZE == 8000
;strategy Retro Q^2 scan
;strategy silk-imps with retin-A - clear up those ugly spl's
spcsz equ (2000+381)
impsz equ (127*22)
; RQ^2, like Anton's Probe only backward
QB EQU (start-350)
QS EQU (0-200)
QD EQU (0+100)
GAP EQU 8
REP EQU 15
datz equ (start-200) ; save one instruction
dat 10*QS, 2*QS ; can get 21 values from this table
table: dat 4*QS, 1*QS ; and can also use the initial value
dat 23*QS, 3*QS ; of fnd
qbomb: dat -GAP,GAP
qincr dat -GAP,2*GAP
dat start+1,0 ; very important!
tab: add.a table ,table ; translate found location
slow: add.ab @tab ,fnd
fast: add.b *tab ,@slow
which: sne.i datz ,@fnd
add.ab #QD ,fnd
;break
add.ba fnd ,fnd
spl attack2 ; start primary bomber
attack1 add.b fnd,fnd2 ; redundant bomber bombs left-to-right
bmb1 mov.i qbomb ,@fnd2
fnd2 mov.i qbomb ,@fnd-(GAP*REP)/2-GAP/2
add #2*GAP ,fnd2
bcnt djn -3 ,#REP
spl RetinB ,<RetinB ; start silk-imps
jmz RetinA ,RetinA ; if not already started
dat 0,0
RetinA spl 1,0
mov.i -1,#0
spl 1
b0 spl @0,spcsz ; two-level silk spawner
mov }b0,>b0
c0 spl @0,impsz
mov }c0,>c0
spl c0,{c0 ; loop and split
mov.i #2*spcsz,}-spcsz*1
mov.i #2*spcsz,}-spcsz*1
mov.i #2*impsz,}-impsz*1
start: seq.i QB+QS*0,QB+QS*0+QD
jmp which,}QS*13 ; qinc+GAP
seq.i QB+QS*1,QB+QS*1+QD
jmp fast,}QB+QS*1+QD/2
seq.i QB+QS*2,QB+QS*2+QD
jmp fast,{tab
seq.i QB+QS*3,QB+QS*3+QD
jmp fast,}tab
seq.i QB+QS*13,QB+QS*13+QD
jmp fast,{fast
seq.i QB+QS*4,QB+QS*4+QD
jmp >fast,}QB+QS*4+QD/2
seq.i QB+QS*5,QB+QS*5+QD
jmp slow,}QB+QS*5+QD/2
seq.i QB+QS*6,QB+QS*6+QD
jmp slow,{tab
seq.i QB+QS*7,QB+QS*7+QD
jmp slow,}tab
seq.i QB+QS*10,QB+QS*10+QD
jmp >fast,<tab
seq.i QB+QS*11,QB+QS*11+QD
jmp slow,<tab
seq.i QB+QS*12,QB+QS*12+QD
djn.f slow,tab
seq.i QB+QS*23,QB+QS*23+QD
jmp >fast,>tab
seq.i QB+QS*24,QB+QS*24+QD
jmp slow,>tab
seq.i QB+QS*17,QB+QS*17+QD
jmp slow,{fast
seq.i QB+QS*8,QB+QS*8+QD
jmp <fast,}QB+QS*8+QD/2
seq.i QB+QS*9,QB+QS*9+QD
jmp tab,}QB+QS*9+QD/2
seq.i QB+QS*15,QB+QS*15+QD
jmp tab,<tab
seq.i QB+QS*16,QB+QS*16+QD
jmp tab,{tab
seq.i QB+QS*20,QB+QS*20+QD
djn.f <fast,tab
seq.i QB+QS*21,QB+QS*21+QD
jmp tab,{fast
seq.i QB+QS*22,QB+QS*22+QD
djn.f tab,tab
seq.i QB+QS*27,QB+QS*27+QD
jmp <fast,>tab
seq.i QB+QS*28,QB+QS*28+QD
jmp tab,>tab
RetinB spl 1,0
mov.i -1,#0
spl 1
mov <1,{1 ; make another copy and start him
spl 4000+7,7 ; safer than going back to RetinA
; which might have been spl-bombed by now
d0 spl @0,spcsz ; two-level silk spawner
mov }d0,>d0
e0 spl @0,impsz
mov }e0,>e0
spl e0,{e0 ; loop and split
mov.i #2*spcsz,}-spcsz*1
mov.i #2*spcsz,}-spcsz*1
mov.i #2*impsz,}-impsz*1
attack2 mov qbomb ,*fnd ; primary bomber
mov.i qbomb ,@fnd ; bombs left and right from found location
fnd: mov.i -GAP/2 ,@QB
add qincr ,fnd
djn -4 ,bcnt
spl RetinA ,<RetinA ; start silk-imps
jmz RetinB ,RetinB ; if not already started
end start
______________________________________________________________________________
Questions? Concerns? Comments? Complaints? Mail them to people who care.
authors: Beppe Bezzi <bezzi@nemo.it> or Myer Bremer <bremermr@ecn.purdue.edu>
or Anton Marsden <amarsden@etherspace.gen.nz>
|
