Issue 39 July 22, 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: (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
______________________________________________________________________________
Greetings.
The '94 draft hill is quiet this week, as Pizza was (is?) down. My mail is
currently bouncing off the server. Hopefully the problems will be fixed soon
and service restored. All the Stormking hills are open and with noticeably
faster response times. You can satisfy your corewar craving there. Enjoy.
--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 39.6/ 25.6/ 34.8 Simple v0.4b Ian Oversby 153.7 12
2 29.3/ 13.1/ 57.6 Return Of The Jedimp John K W 145.4 49
3 41.2/ 37.8/ 21.0 Probe Anton Marsden 144.7 61
4 33.2/ 23.7/ 43.1 Armory II John K W 142.8 200
5 43.4/ 45.8/ 10.8 Q^2 Miro Anders Ivner 141.1 111
6 39.4/ 38.7/ 21.9 Yogi Bear P.Kline 140.1 265
7 41.4/ 42.8/ 15.8 Goldfinch P.Kline 140.1 2
8 41.5/ 43.2/ 15.4 Mist P.Kline 139.8 1
9 41.6/ 43.6/ 14.8 Blur 2 Anton Marsden 139.7 60
10 33.1/ 28.1/ 38.8 Pulp v0.2 Ian Oversby 138.2 156
11 40.7/ 43.6/ 15.7 myVamp5.4 Paulsson 137.7 109
12 40.1/ 43.4/ 16.5 Earthquake v0.2 Bjoern & Ian 136.8 18
13 30.6/ 24.4/ 45.0 Rosebud Beppe 136.7 734
14 37.8/ 39.4/ 22.8 T.N.T. pro Maurizio Vittuari 136.2 779
15 26.4/ 17.0/ 56.6 ompega Steven Morrell 135.8 176
16 37.1/ 39.3/ 23.6 Dura v0.1 Ian Oversby 135.0 90
17 29.9/ 26.9/ 43.2 blue flame c1/10 bjoern guenzel 133.0 45
18 31.5/ 30.8/ 37.7 Jack in the box II Beppe Bezzi 132.1 418
19 36.2/ 40.7/ 23.2 Paper, Scissors and Stone David van Dam 131.6 189
20 36.8/ 43.3/ 19.9 Twister Beppe Bezzi 130.3 515
21 35.4/ 40.9/ 23.7 airBag Paulsson 129.9 102
22 31.7/ 33.4/ 34.9 the historian bjoern guenzel 129.9 534
23 35.2/ 40.7/ 24.1 Scimitar 2 P.Kline 129.7 127
24 36.9/ 46.2/ 16.9 Harmony P.Kline 127.5 26
25 37.3/ 49.4/ 13.3 test mef2 Beppe Bezzi 125.2 9
Weekly age: 31 ( 84 last week, 90 the week before )
New warriors: 6 Turnover/age rate 19%
Average age: 191 ( 184 last week, 158 the week before )
Average score: 137 ( 141 last week, 141 the week before )
The top 25 warriors are represented by 11 authors: Kline with 5; Bezzi,
and Oversby with 4; Guenzel with 3; JKW, Marsden, and Paulsson with 2.
King Report: The top 3 warriors have all been vying for the top slot, but
Ian's new Simple version 0.4b is slightly more effective than the previous
version giving Oversby a commanding 8 point lead over second place. Other
warriors seen on the top of the hill include Mist (quite a few times) and
Q^2 Miro and Blur 2 (once each).
Most of the veterans are in the middle of the pack. Twister and the
historian have not recovered and are in danger of being pushed off. Armory
gets a nice boost since last week moving up to the 4th position.
Question of the day: Is Macrae going to put a Q^2 scan on Thermite and take
up space on the hill for months to come? Or has someone already beat him to
it?
______________________________________________________________________________
94 - What's New
# %W / %L / %T Name Author Score Age
6 42.1/ 42.3/ 15.6 Goldfinch P.Kline 141.9 1
8 41.5/ 43.2/ 15.4 Mist P.Kline 139.8 1
1 36.4/ 26.0/ 37.5 Simple v0.4b Ian Oversby 146.8 1
18 38.9/ 44.1/ 16.9 Harmony P.Kline 133.7 1
25 38.0/ 49.9/ 12.1 test mef2 Beppe Bezzi 126.1 1
13 41.4/ 42.2/ 16.4 Earthquake v0.2 Bjoern & Ian 140.6 1
Goldfinch was hanging around in the middle of the pack until Paul decided to
replace it. The new warrior is holding its place well in the top ten.
______________________________________________________________________________
94 - What's No More
# %W / %L / %T Name Author Score Age
26 1.8/ 2.1/ 0.0 Goldfinch P.Kline 5.5 206
26 1.3/ 1.5/ 1.2 Harmony P.Kline 5.1 8
26 1.6/ 1.7/ 0.7 Mist P.Kline 5.4 30
26 1.2/ 1.3/ 1.6 Simple v0.4 Ian Oversby 5.1 146
26 24.0/ 23.7/ 52.3 Ties, Ties, Ties! (+3) Ross Morgan-Linial 124.4 2
26 33.5/ 42.4/ 24.1 Goliath David van Dam 124.7 181
Kline revamps three of his warriors while Oversby updates sometime-king
Simple. The only real loss is van Dam's Goliath.
______________________________________________________________________________
94 - What's Old
# %W / %L / %T Name Author Score Age
14 37.8/ 39.4/ 22.8 T.N.T. pro Maurizio Vittuari 136.2 779
13 30.6/ 24.4/ 45.0 Rosebud Beppe 136.7 734
22 31.7/ 33.4/ 34.9 the historian bjoern guenzel 129.9 534
20 36.8/ 43.3/ 19.9 Twister Beppe Bezzi 130.3 515
18 31.5/ 30.8/ 37.7 Jack in the box II Beppe Bezzi 132.1 418
No new entries. No losses.
______________________________________________________________________________
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 Iron Gate 1.5 Wayne Sheppard 926 CMP scanner
15 Agony II Stefan Strack 912 CMP scanner
16 Barrage Anton Marsden 876 Qscan -> replicator
17 Blue Funk Steven Morrell 869 Stone/ imp
18 Flurry Anton Marsden 835 Qscan -> pwarrior
19 Thermite 1.0 Robert Macrae 802 Qscan -> bomber
20 T.N.T. pro Maurizio Vittuari 779 * Bomber
21 Blue Funk 3 Steven Morrell 766 Stone/ imp
22 Night Train Karl Lewin 755 Replicator
23 Mirage 1.5 Anton Marsden 736 Scanner/ bomber
24 Rosebud Beppe 734 * Stone/ imp
25 Blizzard Anton Marsden 713 Qscan -> replicator
T.N.T. pro moves up two spots while Rosebud climbs to 24th position.
______________________________________________________________________________
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 58.3/ 20.9/ 20.8 Yet 4b Justin Kao 195.8 48
2 45.5/ 16.5/ 38.0 (-: :-) Ross 174.6 71
3 49.6/ 29.0/ 21.5 Velveeta Shift-F shar 170.1 46
4 45.4/ 23.2/ 31.4 Versatility 1.5 Ross Morgan-Linial 167.6 38
5 46.1/ 26.2/ 27.7 Saboteur v0.4k shar 165.9 79
6 40.3/ 27.8/ 31.9 Inferno 1.0 Philip Kendall 152.8 61
7 29.5/ 11.0/ 59.5 Microsoft v1.0 Justin Kao 148.0 4
8 22.7/ 10.7/ 66.7 Ties, Ties, Ties! (+2) Ross 134.6 73
9 26.2/ 23.2/ 50.6 Utility Knife Robert J. Street 129.2 81
10 19.8/ 13.8/ 66.4 HAL 204 Justin Kao 125.7 1
11 14.0/ 8.5/ 77.6 Wasps 1.3 Ross 119.5 72
12 15.0/ 10.8/ 74.2 more testing Anonymous 119.2 32
13 11.9/ 6.9/ 81.2 Nematode v1.4e Jonathan Stott 116.8 12
14 13.2/ 9.6/ 77.3 Ties and Wasps Ross 116.7 25
15 11.5/ 7.0/ 81.4 Nematode v1.4b Jonathan Stott 116.0 45
16 16.0/ 16.9/ 67.1 Fork 4/13 Christoph C. Birk 115.1 65
17 11.6/ 27.7/ 60.8 Handy Man Robert J. Street 95.5 39
18 2.5/ 16.9/ 80.5 Mama's Boy Robert J. Street 88.2 26
19 11.3/ 35.2/ 53.6 silken stomp harleyQ2 87.4 21
20 3.5/ 19.8/ 76.6 silkbombQ2 harleyQ2 87.3 44
21 3.3/ 21.4/ 75.2 Snail Edgar 85.2 10
22 10.1/ 36.8/ 53.1 silken train harleyQ2 83.4 14
23 6.4/ 38.6/ 55.0 b3 harleyQ2 74.1 13
24 3.8/ 33.8/ 62.4 b harleyQ2 73.7 20
25 3.2/ 33.8/ 63.0 abc.i harleyQ2 72.6 2
I see a lot of familiar names on the beginner's hill. Is Kao implying the
Microsoft programs are really just big viruses with his warrior Microsoft
v1.0 or is he predicting Gates and Co. will dominate KotH as the do the
desktop? (-: :-), very cute.
______________________________________________________________________________
The Hint
In a weak attempt to gain a foothold on the hill, I thought I would develop
a new type of p-brain (pea-brain as it turns out). I usually use the basic
switch on loss or tie p-logic like Chameleon's.
res ldp.ab _RES, #0
str ldp.a _STR, str1 ;load strategy in use
seq.ab #1, res ;check result
lost add.a #1, str1 ;lost change
mod.a #2, str1 ;secure result
win stp.ab str1, _STR
str1 jmp @0, strat1
dat 0, strat2
The p-logic is quick and doesn't self destruct when brainwashed. Quickness
is especially important now with the new Q^2 quickscans flooding the hill.
The downside is no matter what _effective_ strategy you are using, a spurious
loss will switch you to a less effective one. It may take even another loss
to return to the optimal strategy if you are switching more than one
component. To combat this phenomena, I developed a switch on two consecutive
tie/loss combinations. Strategies will only change on a loss loss, tie loss,
loss tie, or tie tie.
A loss or tie will advance the table pointer contained in _STRAT. If the
_STRAT pointer is 0 or 2, that strategy will get its 'second chance'. If the
index is 1 or 3, strategies will switch. A win should reset the pointer to
either 0 or 3 depending on what strategy is currently in use. I do this by
taking the index modulo 2 and subtracting the result from the current index.
_RES EQU #0
_STRAT EQU #1
result ldp.ab _RES, wlt
strat ldp.a _STRAT, choice
wlt sne.ab #1, #0
jmp win ; win has occured
add.a #1, choice ; loss or tie has occured
mod.a #4, choice
stp.a choice, _STRAT
choice jmp @0, strat1 ; _STRAT = 0, choose strat1
nop 0, strat1 ; _STRAT = 1, choose strat1
nop 0, strat2 ; _STRAT = 2, choose strat2
nop 0, strat2 ; _STRAT = 3, choose strat2
win mov.ab choice, #0 ; reset index becuase of win
mod.ab #2, win
sub.ba win, choice
jmp choice-1
strat1 jmp 0
strat2 jmp 0
All in all, the brain is much bigger (8 vs. 15) and a bit slower. But I
thought it had the smarts necessary to defeat non-pspace programs quite
handily. However this wasn't the case. When switching the components in
Chameleon (mod 5 bomber and .66c scanner), the new logic performed similarly
or worse! The optimal strategy would win for awhile, but the occasional loss
tie combination would cause it to switch. The secondary strategy (being a
somewhat general purpose strategy) would not switch out as fast due to the
new switching rules. This seemed to balance the extra wins gained. I
believe the p-logic would be much more effective if the secondary strategy
is very specialized against one type of program and losses quickly against
everything else. But don't take my word for it. Use your nifty cut and
paste and try a few components.
______________________________________________________________________________
Extra Extra
by Beppe Bezzi
Here is Jack in the box II, there is nothing really innovative in it.
The bomber is in practice the same published in Twister, there is a little
variation in the bomb pattern, but nothing worth mention.
The replicator is a part of that in the old Jack, just booted away to have a
faster reaction time against qscanners, even if not fast enough against Q^2
(It would need a lighter paper)
The switcher is exacltly the same of old Jack, no need to change it being
safe and brainwash resistant.
Overall I don't expect it to live much longer than old version, even if
nobody knows :-)
;redcode-94
;name Jack in the box II
;author Beppe Bezzi
;kill Jack
;strategy Same old strategy, improved components
;strategy p-switching Tornado and silk
;assert CORESIZE == 8000
org think ;v 2.03
PIN 3371
_RES equ #0
_STR equ #1 ;more or less 500 :-)
step equ -45
DIST
away equ 4000 ;more or less 3900 :-)
gate1 equ (gate-4)
A0 equ 3488
A1 equ 1860
A2 equ 3740
marcia
spl 1, <300 ;\
spl 1, <400 ;-\ generate 15
spl 1, <500 ;-/ parallel processes
mov -1, 0 ;/
mov <bootslk,{bootslk
bootslk spl 4000, silk+15 ;more or less 4000
silk spl @0, }A0 ;split
mov.i }-1, >-1 ;copy
mov.i bomb1, >123 ;bombing
silk2 spl @0, }A1 ;split
mov.i }-1, >-1 ;copy
mov.i bomb1, >1001 ;bombing
mov.i bomb , }2042 ;A-indirect bombing
mov.i {silk2, <silk3 ;copy
silk3 jmp @0, >A2 ;jmp new copy
bomb dat.f >2667, >5334 ;anti-imp bomb
bomb1 dat >1, }1 ;anti clear bomb
for 12
dat 0,0
rof
think
res ldp.ab _RES, #0
str ldp.a _STR, str1 ;load strategy in use
sne.ab #0, res ;check result
lost add.a #1, str1 ;lost change
mod.a #2, str1 ;secure result
win stp.ab str1, _STR
str1 jmp @0, tornado
dat 0, marcia
for 12
dat 0,0
rof
tornado
mov gate, }pt2
mov gate, *pt2
mov last, <pt1
spl 1, 1
mov {pt1, <pt1
mov {pt1, <pt1
mov {pt1, <pt1
mov {pt1, <pt1
mov {pt1, <pt1
go djn.b @pt1, #2 ;start Tornado
mov gate, <pt1
pt1 div.f #last, #last+1+away
pt2 div.f #gate+away-5,#1
;--- Tornado 3.8
for MAXLENGTH-CURLINE-14
dat 0,0
rof
start
dat -25, last-gate1+5
warr
gate
dat -25, last-gate1+15
bombs spl #(step+1), -step ;hit spl
start1 sub incr, @b1
stone mov (0*step)+jump+1,*(1*step)+jump
b2 mov bombs, @stone
b1 mov bombm, *stone
jump jmz.b start1, #0 ;hit by spl
clr mov @djmp, >gate1
mov @djmp, >gate1
djmp djn.b clr, {bombs
incr dat >-3*step,>-3*step
last
bombm dat <1, {1
end
______________________________________________________________________________
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@comp.vuw.ac.nz>
|
