Sections
Home
Hills
Infinite Hills
Tournaments
Software
Evolving
Optimizer
Community
Newsletter
Discussion
History
Sections
 
For Beginners
First Steps
FAQ
Guides
Lexicon
Benchmarks
For Beginners
> Home > The Corewar Newsletters > Core Warrior > Issue #1

Issue 47                                                   14 October, 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

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

NEW
Beppe Bezzi web page - http://www.aspide.it/freeweb/Bezzi
______________________________________________________________________________
Greetings.

The first round of the tournament has been a real debacle for Core Warrior
staff. Myer and Anton, pitted against two opponents like the young lion
Franz and the expert Steven Morrell, lost their matches and now are in the
bottom of the standings. One cannot leave them alone for a moment :-)

The 94 hill lost two of the three older warriors, TNT pro and Jack in the
box II. All happened when Robert Mcrae, back to the arena after some time
for the tournament and under an obvious attack of battle lust, submitted a
dozen test warriors to the hill leaving it with perhaps the lowest average
age from its creation, only 117

Thos modified the format of his hills on the web page; now you can look at
warriors headers, or mail the author, with a click of the mouse. Good work.


--Beppe Bezzi
______________________________________________________________________________
Tournament time

First round of my tournament is over, not without some surprising result.
Some of the 'battle scarred veterans' have been beated by jounger player
used at the beginners hill more than at the hall of fame, in particular the
victories of Matt Lewinsky and Justin Kao against two veterans of the
caliber of Robert Mcrae and J.K. Wilkinson are worth mention, congratulations.

Here are first round results and standings:

CC Paper 3T by Franz scores 369
Chameleon by M R Bremer scores 180
Results: 106 43 51

invicta by John K. Lewis scores 95
mayhem.q by harleyQ2 scores 455
Results: 15 135 50

Falcon v0.5 by Ian Oversby scores 529
T-1 vX by Ross Morgan-Linial scores 64
Results: 174 19 7

Yet 4b Clear by Justin Kao scores 277
evoltmp by John K W scores 271
Results: 75 73 52

ompega by Steven Morrell scores 279
Damage Incorporated by Anton Marsden scores 225
Results: 61 43 96

Fleetfoot in a Rose-Garden by Robert Macrae scores 120
Flamberge 101 by Matt Lewinski scores 465
Results: 35 150 15

Tangle Trap by David Moore scores 230
Sapphire by Philip Kendall scores 341
Results: 67 104 29

myConfuser by Paulsson scores 383
Negative Energy by Zul Nadzri scores 200
Results: 122 61 17

digital breakdown v01b by bjoern guenzel scores 181
Linalg by George Eadon scores 253
Results: 5 29 166

Standing after round 1 are the following:

Pos.    #       Player                Score  
-----------------------------------------------
 1       1      Franz                   3
 1       4      Joshua                  3
 1       5      Ian Oversby             3
 1       7      Justin Kao              3
 1       9      Steven Morrell          3
 1      12      Matt Lewinski           3 
 1      14      Philip Kendall          3
 1      15      Magnus Paulsson         3
 1      18      G. Eadon                3
10       2      M. R. Bremer            0
10       3      J. K. Lewis             0
10       6      Ross Morgan-Linial      0
10       8      J. K. Wilkinson         0
10      10      Anton Marsden           0
10      11      Robert McRae            0
10      13      David M. Moore          0
10      16      Zul Nadzri              0
10      17      Bjoern Guenzel          0

Next round will be played withlimited processes rules; redcode and coresize
are those of first round but warriors have to work under following assert:

;assert MAXPROCESSES == 8 && MAXLENGTH == 200

You can see tournament results at my new web page:
http://www.aspide.it/freeweb/Bezzi

the tournament internal link is still floating, because the page is under
construction, but I hope all of you will be able to find them even if most
of the text is writen in italian (Corewar part is in english, don't worry)

I'm working at something interesting for next rounds, stay tuned.
______________________________________________________________________________
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

The current ICWS '94 Draft hill:
 #   %W /  %L /  %T                       Name               Author Score  Age
 1  36.7/ 20.8/ 42.5                    Gigolo   Core Warrior staff 152.6  116
 2  45.6/ 45.0/  9.4                  Memories               Beppe  146.2    7
 3  32.9/ 20.4/ 46.6           unrequited love                kafka 145.4  160
 4  39.2/ 33.5/ 27.3       Damage Incorporated        Anton Marsden 144.9  103
 5  29.3/ 15.0/ 55.7      Return Of The Jedimp             John K W 143.6   62
 6  30.2/ 16.8/ 53.0               Impish v0.2          Ian Oversby 143.5  101
 7  41.1/ 39.1/ 19.8                    Blur 2        Anton Marsden 143.1  239
 8  30.4/ 18.4/ 51.1                   Rosebud                Beppe 142.4  913
 9  34.4/ 29.0/ 36.7                  Tiberius                Franz 139.7   23
10  24.3/ 10.9/ 64.8                testing t2             John K W 137.8    2
11  24.2/ 10.7/ 65.1                 Trident^2             John K W 137.7    8
12  41.8/ 46.1/ 12.1                 Eggbeater        Anton Marsden 137.6   21
13  23.4/  9.9/ 66.7                      test             John K W 136.8    1
14  32.0/ 27.9/ 40.2              Simple v0.4b          Ian Oversby 136.0  191
15  37.3/ 39.1/ 23.6                     Probe        Anton Marsden 135.5  240
16  24.7/ 14.4/ 60.9                    ompega       Steven Morrell 135.0  355
17  31.3/ 27.5/ 41.2               Falcon v0.3          Ian Oversby 135.0  145
18  36.4/ 39.1/ 24.5                 Goldfinch              P.Kline 133.8  181
19  40.3/ 47.5/ 12.2                  Q^2 Miro         Anders Ivner 133.1  290
20  38.2/ 43.4/ 18.4                   Bodge 1        Robert Macrae 133.0   32
21  38.3/ 43.9/ 17.8                   Harmony              P.Kline 132.8   26
22  31.9/ 31.4/ 36.8              CC Paper 3.3                Franz 132.3   58
23  35.5/ 40.2/ 24.3          Instant Wolf 3.4                Edgar 130.9  123
24  33.3/ 36.6/ 30.1                     Frogz                Franz 129.9  120
25  34.6/ 40.7/ 24.7              Instant Wolf                Edgar 128.5   40

I like this hilltop :-)

Weekly age: 56 ( 46 last issue, 13 the week before )
New warriors: 9  Turnover/age rate 16%
Average age: 142 ( 169 last issue, 140 the week before )
Average score: 138 ( 133 last issue, 140 the week before )
The top 25 warriors are represented by 15 authors: Marsden with 4; Franz and
Oversby with 3; Beppe, Kline and Edgar with 2; everyone else with 1.

King Report: Lot of movement in the top, no warrior held the King spot for
long. Blur 2, Damage Incorporated, Return of Jedimp and Unrequited Love kept
it for some challenge.

Q^2 scans are still at the top in various combinations. Paper warriors are
losing a bit while anti imp scanners are back. Pspacers are having very hard
times; the first of them is in 18th position

Lots of testing this week, more than 70 test challenges. 
To submit a test warrior you have to put the word 'test' where usually goes
quiet or verbose. Example to test on the 94 hill:

;redcode-94 test
;name Test
...

______________________________________________________________________________
94 - What's New

 #   %W /  %L /  %T                       Name               Author Score  Age
 2  45.1/ 46.2/  8.7                  Memories               Beppe  144.0    1
 5  37.7/ 28.6/ 33.7                  Tiberius                Franz 146.7    1
 8  20.1/  9.2/ 70.7                 Trident^2             John K W 131.0    1
11  22.3/ 11.2/ 66.5                testing t2             John K W 133.3    1
13  37.2/ 40.0/ 22.7              Instant Wolf                Edgar 134.5    1
15  40.7/ 43.2/ 16.1                   Harmony              P.Kline 138.3    1
15  19.5/ 10.0/ 70.5                      test             John K W 129.0    1
16  42.6/ 47.7/  9.7                 Eggbeater        Anton Marsden 137.5    1
20  38.7/ 48.3/ 13.0                   Bodge 1        Robert Macrae 129.1    1

Three new entries in the top ten: new version of Memories, Tiberius and
Trident^2
______________________________________________________________________________
94 - What's No More

 #   %W /  %L /  %T                       Name               Author Score  Age
26  32.3/ 40.4/ 27.3                T.N.T. pro    Maurizio Vittuari 124.2  925
26  26.8/ 30.7/ 42.5        Jack in the box II          Beppe Bezzi 123.0  559
26  33.6/ 42.4/ 24.0                  mrb-test           m r bremer 124.8  114
26   2.1/  1.3/  0.6                      Test        Anton Marsden   7.0   95
26  32.5/ 42.1/ 25.4               Inferno 2.4       Philip Kendall 123.0   84
26  37.1/ 48.8/ 14.1              Taking Names              P.Kline 125.4   68
26  30.8/ 44.3/ 24.9                    mayhem             harleyQ2 117.3   19
26  21.6/ 35.9/ 42.5             Barbarian 3.1                Franz 107.3    9
26   3.7/ 21.9/ 74.4                       sin             harleyQ2  85.5    4

T.N.T. pro ends its run at 925 age, Jack II at 559.
______________________________________________________________________________
94 - What's Old

 #   %W /  %L /  %T                       Name               Author Score  Age
 8  30.4/ 18.4/ 51.1                   Rosebud                Beppe 142.4  913
16  24.7/ 14.4/ 60.9                    ompega       Steven Morrell 135.0  355
19  40.3/ 47.5/ 12.2                  Q^2 Miro         Anders Ivner 133.1  290
15  37.3/ 39.1/ 23.6                     Probe        Anton Marsden 135.5  240
 7  41.1/ 39.1/ 19.8                    Blur 2        Anton Marsden 143.1  239

Rosebud is now the oldest warrior.
______________________________________________________________________________
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  T.N.T. pro             Maurizio Vittuari  925    Bomber
16  Rosebud                Beppe Bezzi        913 *  Stone/ imp
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

T.N.T. pro ends its run in 15th position. Rosebud moves up some spots.
______________________________________________________________________________
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

The current Beginner hill:
 #   %W /  %L /  %T                       Name               Author Score  Age
 1  48.2/ 16.0/ 35.8                  Dust 0.7           Justin Kao 180.5    3
 2  48.2/ 38.5/ 13.3          Velveeta Shift-I                 shar 157.9   93
 3  47.5/ 37.9/ 14.7       Paper Shredder 3500           Justin Kao 157.1   30
 4  47.0/ 38.6/ 14.4                  mayhem.q             harleyQ2 155.3    6
 5  47.8/ 41.4/ 10.8             Flamberge 101        Matt Lewinski 154.3    7
 6  45.8/ 38.7/ 15.5          nSplit Lite v1.2                 shar 152.9   81
 7  37.8/ 23.9/ 38.3                  Q Yogurt                Franz 151.7   26
 8  44.6/ 40.7/ 14.7                 Recover 3                Franz 148.5    1
 9  44.7/ 41.2/ 14.1                  Eagle.b1       Philip Kendall 148.2   58
10  42.3/ 38.6/ 19.1                    mayhem             harleyQ2 146.0   43
11  34.6/ 24.3/ 41.0                 Gravel 11           Justin Kao 144.9   54
12  34.1/ 25.0/ 40.9              FastScan 1.0       Philip Kendall 143.3   73
13  29.5/ 20.9/ 49.6                Papyrus 13           Justin Kao 138.1   55
14  40.8/ 44.9/ 14.3  Neverending Brutality 10                Franz 136.7   22
15  29.1/ 23.8/ 47.1                   testing            Anonymous 134.4    4
16  29.7/ 25.1/ 45.2                  Sapphire       Philip Kendall 134.3   52
17  37.5/ 40.9/ 21.7                  hack.exp             harleyQ2 134.1    8
18  35.9/ 38.2/ 25.9                  murder.p             harleyQ2 133.6   75
19  39.7/ 47.1/ 13.2               Conquest II           Zul Nadzri 132.4   74
20  37.1/ 42.9/ 20.0         Microsoft NT v2.5           Justin Kao 131.3   44
21  26.2/ 23.5/ 50.4           Sand Storm v0.7           Justin Kao 128.8   38
22  35.0/ 42.4/ 22.6             Autumn Leaves        Matt Lewinski 127.7   27
23  28.3/ 30.4/ 41.3                      Frog                Franz 126.3   95
24  27.3/ 30.2/ 42.4                    Linalg         George Eadon 124.4    9
25  27.2/ 31.1/ 41.7               Barbarian 3                Franz 123.3   23

Dust 0.7 holds the head with an incredible lead over the second; in fact the
warrior did well also in the 94 hill; Justin it's time to play in the pros
and leave the -b hill to beginners, same thing for Franz of course.
______________________________________________________________________________
The Hint
Better Warriors (Papers and Stones)
by Franz

I'm writing this because I don't think these topics have been covered, or
I haven't read about them anywhere.

And I am still a beginner so my coding is not so good. These are the
things that I have found useful. (what I'm trying to say is this code may
suck)

STONES
======

Making stones anti-imp ... well here are a few ideas that DO work. At
least I think they do!

If writing regular stone (.33c) have the counter where you are bombing,
well in front of the warrior, that way when you add to it, you might kill
an imp. your bomb line can also be something like 

...
mov >gate,@where
...

But what about those .6c bombers which have the counter as one of the mov
lines ... well put that line as far up as you can and the add line as low
as you can. here's an exemple

(opt is the step number, stbm is where to start bombing)

gate    dat opt,xxx ;this is the gate line decremented and used as a bomb
...                 ;also the a field is used for the where line bombing

...
stone   mov gate,@where             ;no bdecrement we have to use the gate
where   mov >stbm+opt,*stbm+(opt*2) ;use anything but here we increment
                                    ;the a field which will be bombed in
                                    ;the next step
        mov >gate,*where
        add.f jump,where            ;jump is my coreclear spl# line
        jmp stone,>gate
...

...
jump    spl #opt*3,opt*3
...

This is quite effective against imps large spirals especially ... try your
old stone configuration against this.

Self splitting might be used in this warrior too ... be sure to reverse
the instructions on selfsplitting stones as they are executed backwards.
This has been covered many times, by many people ... but I still make that
mistake sometimes and then wonder why it doesn't work right. this is
usually just small speed decrease but in some cases the speed can go down
by .2c and you wander why your bomber is slow.

I have found the >gate is probably the most effective gate there is. Watch
how mov #step,*0 can get through

If you increment the *0 to *1 it becomes 

mov #xxx,*1

which can still live as a normal imp (1 point) and could potentially hurt
your code or at least start executing your code and tie ... This is what
you can do to prevent it

gate   dat 0,0
...
       dat -20,0  
...
;your code here

this will kill such an imp since the imp uses the a-field of the next
instruction for movement and this will kill it. Since it copies itself
behind and then just falls off and dies on the dat -20,0

PAPERS
======

What to do with papers.

Silk is probably the only paper worth doing at this time. But how to
do a good silk.

From the tutorial you saw how to do a small yet inefficent silk .. since
Status: R

it has to take care of the pointers and reset them. Remember each
instruction in a silk is executed many times, so just resetting a pointer
takes many turns and slows down the paper. Here's how to avoid it

spl @0,>step1  ;silk1
mov }-1,>-1
spl @0,>step2  ;silk2
mov }-1,>-1
spl @0,>step3  ;silk3
mov }-1,>-1
.. your attack goes here

this will make copies 3 kinds one will have 3 copiers and be an exact
copy, one will have two silks and one will have one silk.

This is probably the standard way to silk .. maybe with a timescape
routine at the end which makes one more copy.

Your attack:

1) make sure it never just falls off. It has to continue or you have to
make sure it dies. Since if an enemy executes your code he will usually
fall htrough to the attack and you want him to die. I did this by making
the last instruction copied with the paper a bomb which i used for the
attack and then it would die. the problem with falling off is that the
enemy migth go through your attack and then fall back to his own code.

2) NEVER, NEVER use DJN.F 0, <-xx for bombing. I tried this and it's VERY
hostile against silks but a lot of other warriors are immune or little
affected by this.

3) Remember the longer the paper the longer it takes to copy, do not make
the bomb routine long, instead bomb a few times and let it die but in the
meantime you have several other copies already bombing. Bombing by copying
several lines after you is also not too effective .. it's better to just
use a bomb routine which can be more effective.

good bomb routines is a set of

...
mov >-x,<x
mov >-x*2,<x*2
mov >-x*3,<x*3
...

or similiar ... it increments one locations decrements another and creates
a bomb carpet on the second location. so depending on the numebr of
proccesses n, you bomb n locations decrement 1 n times and increment 1 n
times. if you just made n higher and copy few bombs with you you would
lose 2 cycles to every bomb since the copy routine works at .5c and the
bomb routine works at 1c for bombs.

OK that's enough for today.

______________________________________________________________________________
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>
2002-2005 corewar.info. Logo C. Schmidt