From: msmoulton@aol.com (MSMoulton)
Subject: Re: FAQ frequency
Date: 1995/09/01
Message-ID: <428j54$ltp@newsbf02.news.aol.com>#1/1
references: <4265qh$3pb@news.asu.edu>
newsgroups: rec.games.corewar

In article <4265qh$3pb@news.asu.edu>, sieben@imap1.asu.edu writes:

>
>How about a mini FAQ that mainly tells about the location of the faq and 
>soda. This could be posted more frequently without the overload.
>
>

I think that posting the FAQ weekly would be best.  Once or twice a month
is not enough, but daily is annoying.  But a mini FAQ would just be
annoying.  It's easier just to post the whole thing instead of worrying
about 2 different things to post



From: pk6811s@acad.drake.edu
Subject: Re: p-space and switching routines
Date: 1995/09/01
Message-ID: <1995Sep1.155605@acad.drake.edu>#1/1
references: <427a0f$820@mozo.cc.purdue.edu>
newsgroups: rec.games.corewar

(Myer R. Bremer) writes:
>...
> strategy.  When p-space warriors battle each other, they may have
> a tendency to hash.  One warrior wins with its stone.  It stays
>...

Yes.  But it might be possible to pick up your opponents cycle and
beat him.  Or you could confuse him by picking at random.

But normal deviation in single-battle outcomes confuse the switcher anyway.

> 
> Right now, specialized components are creaming the "dumb" programs on
> the hill.  We'll see how those same components do when the non-pspace
> warriors are gone. 
> 

Generalized components will become more important for sure.  Let's say
your switcher is based on the win/loss ratio with no points for a tie.
Then if your 'bad choice' component only gets ties it will be switched out
but it will not have given much away.

Component switching is not the only possibility.  You could try optimizing
a step size, or adjusting the number of times you bomb before launching Silk,
or using a positive increment instead of negative.

Paul Kline
pk6811s@acad.drake.edu



From: jwilkinson@utexas.edu
Subject: Re: Is John Perry out there?
Date: 1995/09/01
Message-ID: <425tk9$o3k@geraldo.cc.utexas.edu>#1/1
references: <420j6a$d0g@newsbf02.news.aol.com> <tom.9.304654E0@halls1.cc.monash.edu.au>
newsgroups: rec.games.corewar


>any idea where I can get a copy of this paper?  sounds really interesting...
>
>please email or post, I don't care ;)
>
>-Tom
>	tom@halls1.cc.monash.edu.au
>	http://yoyo.cc.monash.edu.au/~madmick

Well, you can read the Corewars FAQ, that'll give you sites to check.
Or, check http://www.stormking.com/~koth , I -know- it's there...


<to Stefan> Post the FAQ more often, please... 




From: tom@halls1.cc.monash.edu.au (Tom Paton)
Subject: Re: Is John Perry out there?
Date: 1995/09/01
Message-ID: <tom.9.304654E0@halls1.cc.monash.edu.au>#1/1
references: <420j6a$d0g@newsbf02.news.aol.com>
newsgroups: rec.games.corewar

In article <420j6a$d0g@newsbf02.news.aol.com> msmoulton@aol.com (MSMoulton) writes:
>Does anyone know John Perry's e-mail address?  He's the author of the
>paper "Core Wars Genetics: The Evolution of Predation".  I tried the

any idea where I can get a copy of this paper?  sounds really interesting...

please email or post, I don't care ;)

-Tom
	tom@halls1.cc.monash.edu.au
	http://yoyo.cc.monash.edu.au/~madmick



From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: P-Space Switcher
Date: 1995/09/01
Message-ID: <427r7g$ibp@lastactionhero.rs.itd.umich.edu>#1/1
references: <lewinDE87uo.42w@netcom.com>
newsgroups: rec.games.corewar

Karl Lewin (lewin@netcom.com) wrote:
: Here is another bit of code to determine how to rotate warriors using 
: p-space.  I'm not sure how good it is because of my limited understanding 
: of how strong my component warriors are and how they interact with what 
: is the current opponents.  Anyways, I though I'd throw it out in case 
: anyone has any comments or insights.

: The basic premise is to rotate to the next warrior only when there are 
: more losses than wins.  I think it works ok in practice but still can be 
: fooled be some early losses.  It can be adapted to use more or less 
: component warriors by changing the values given to losses/wins and the 
: number of component values.

: Anyone have any thoughts on what the perfect components should be?  My 
: feelings are that they should do extremely well against a certain class 
: of opponents and fail miserably against all others.  This way a pspace 
: warrior will "quickly" rotate to the opponent most suited to face each 
: opponent.  Is this logic flawed.  

: (FYI- The current "working" version uses a simple stone, a modified form 
: of Ryooki, and a slightly modified old version of Rave as components)

: Well enough rambling.......

: _RES	EQU	#0
: _W	EQU	#1
: _L	EQU	#2
: _CUR	EQU	#3

: 	ldp.ab	_RES,	check
: 	ldp.ab	_W,	wins
: 	ldp.ab	_L,	losses
: 	ldp.a	_CUR,	launch
: check	sne.ab	#0,	#0	
: losses	add.ab	#2,	#0      ;loss value
: 	sne.ab  #1,	check
: wins	add.ab  #5,	#0      ;win value
: 	stp.b	losses, _L
: 	slt.b	losses,	wins
: switch	add.a	#1,	launch
: 	mod.a	#3,	launch  ;# of components
: 	stp.ab	launch,	_CUR
: 	stp.b	wins,	_W
: launch	jmp	@1,	scissor	;3rd
: 	dat	0,	paper	;1st
: 	dat	0,	rock	;2nd


: paper	jmp 0
: rock    jmp 0
: scissor jmp 0

: -- 


Here is my version of the switching program.  I choose to switch if there was a lose last 
turn.  This is also the fastest "switch" I`ve made (or seen).

;redcode-94b
;author John K. Lewis
;strategy "Smart" program.  Uses P-Space to remember last fight.
;name Obvious

S equ 3

result	ldp #0,#0	; load in last result
prog	ldp #S, #0	; load in last program (1-4)
	seq #1, -2	; stick with a winner
	add #1, prog	; if we lost try next program
	seq #0,prog
	mov.ba prog, 1	; grab program pointer
	jmp.a 1, 0	; eventual pointer
jmp.a rusty		; index of choices
jmp.a pidge
jmp.a impring
rusty stp #1,#S		; indexed choice of '4' gets reset to '1'

**************
*   Rusty    *
**************

pidge stp #2,#S

**************
*   Pidge    *
**************


impring stp #3,#S

**************
*  impring   *
**************






From: jwilkinson@utexas.edu
Subject: Re: P-Space Switcher
Date: 1995/09/01
Message-ID: <427oa9$15l@geraldo.cc.utexas.edu>#1/1
references: <lewinDE87uo.42w@netcom.com> <1995Sep1.090321@acad.drake.edu>
newsgroups: rec.games.corewar

>A couple of years ago we had a discussion about how many possible battles
>there are between any two opponents with the answer somwhere around 7900.
>That also implied 7900 possible series-of-100 rounds since the 'random'
>sequencing is pseudo-random.  Now with pspace there are many possible
>series based on each pair of starting locations.  So even with identical
>starting locations two series can have very different outcomes - Chaos Theory!  
>And there are many more than 7900 possible sequences in every series-of-100.

Um, if the pMARS battle locations essentially random?

Well, I can only assume it is.  If so, then you would have 8000*8000*7900
possibilities for two programs which only accessed Pspace -ONCE-.  For
each additional Pspace access, you multiply by 8000 again...  I believe
the Pspace warrior I sent read from pSPACE 3 times.  That means that even
fighting a "dumb" warrior I've got 4.0448e+15 possible starting settings.

-JKW




From: bremermr@cartoon.ecn.purdue.edu (Myer R. Bremer)
Subject: p-space and switching routines
Date: 1995/09/01
Message-ID: <427a0f$820@mozo.cc.purdue.edu>#1/1
newsgroups: rec.games.corewar

Greetings.

I've been following the discussion about what kind of components
to include in the new p-space warriors.  Current favor seems
to fall towards very specialized components.  As the hill fills
with p-space warriors, I don't think this will be an effectvie   
strategy.  When p-space warriors battle each other, they may have
a tendency to hash.  One warrior wins with its stone.  It stays
stone for the second round.  The other warrior switches to paper
and wins.  It stays paper for the third round.  But the other 
warrior has booted its scanner this time around.  Sure, the 
sequencing may be off due to the semantics of a specific switching
algorithm, but hopefully you see my point.  A better strategy MAY
be to use a stone with some kind of paper defense.  Use a replicator
that can have a good chance of killing scanners.  Use scanners with
some type of stone killing mechanism.  In essence, use the same
multi-strategy / high survival warriors we were before p-space with
some kind of switching strategy.  Of course this would also make a
warrior more resistant to attacks targeted directly at their p-space.

Right now, specialized components are creaming the "dumb" programs on
the hill.  We'll see how those same components do when the non-pspace
warriors are gone. 

M R Bremer
 



From: jwilkinson@utexas.edu
Subject: Re: FAQ frequency
Date: 1995/09/01
Message-ID: <4276el$hte@geraldo.cc.utexas.edu>#1/1
references: <9508241558.AA03078@idnsun.gpct.Vanderbilt.Edu.noname> <422hfp$ddu@geraldo.cc.utexas.edu> <4265qh$3pb@news.asu.edu>
newsgroups: rec.games.corewar

>How about a mini FAQ that mainly tells about the location of the faq and 
>soda. This could be posted more frequently without the overload.

Yes, yes.  That would work good.





From: pk6811s@acad.drake.edu
Subject: Re: P-Space Switcher
Date: 1995/09/01
Message-ID: <1995Sep1.090321@acad.drake.edu>#1/1
references: <lewinDE87uo.42w@netcom.com>
newsgroups: rec.games.corewar

(Karl Lewin) writes:

>...
> Anyone have any thoughts on what the perfect components should be?  My 
> feelings are that they should do extremely well against a certain class 
> of opponents and fail miserably against all others.  This way a pspace 
> warrior will "quickly" rotate to the opponent most suited to face each 
> opponent.  Is this logic flawed.  

This is also what I came to after some experimentation.  I tried using
DieHard (averages < 1 loss per opponent :-) in the rotation, but gave up.
The thing wanted to 'play it safe' - and you just can't score big with
a lot of ties.  Use components that are very good against one or more
common styles and make your scoring system an agressive one.

Your switcher looks very credible.  And it is innoculated against Alzheimers,
by ALWAYS performing MOD on the strategy before doing the jump.  Yeesh.
The report is lost, but someone was beating my test 99 out of 100 until I
figured this out.  Randy???  Beppe???  one of you guys is too too clever :-)

As to components - a simple mod-x stone is probably one requirement as it
is guaranteed to rack up wins against a stand-alone scanner.  Boot it
away from the rest of the code however.  And it will quickly be
switched out if the opponent is paper.

--

A couple of years ago we had a discussion about how many possible battles
there are between any two opponents with the answer somwhere around 7900.
That also implied 7900 possible series-of-100 rounds since the 'random'
sequencing is pseudo-random.  Now with pspace there are many possible
series based on each pair of starting locations.  So even with identical
starting locations two series can have very different outcomes - Chaos Theory!  
And there are many more than 7900 possible sequences in every series-of-100.

Paul Kline
pk6811s@acad.drake.edu



From: lewin@netcom.com (Karl Lewin)
Subject: P-Space Switcher
Date: 1995/09/01
Message-ID: <lewinDE87uo.42w@netcom.com>#1/1
newsgroups: rec.games.corewar

Here is another bit of code to determine how to rotate warriors using 
p-space.  I'm not sure how good it is because of my limited understanding 
of how strong my component warriors are and how they interact with what 
is the current opponents.  Anyways, I though I'd throw it out in case 
anyone has any comments or insights.

The basic premise is to rotate to the next warrior only when there are 
more losses than wins.  I think it works ok in practice but still can be 
fooled be some early losses.  It can be adapted to use more or less 
component warriors by changing the values given to losses/wins and the 
number of component values.

Anyone have any thoughts on what the perfect components should be?  My 
feelings are that they should do extremely well against a certain class 
of opponents and fail miserably against all others.  This way a pspace 
warrior will "quickly" rotate to the opponent most suited to face each 
opponent.  Is this logic flawed.  

(FYI- The current "working" version uses a simple stone, a modified form 
of Ryooki, and a slightly modified old version of Rave as components)

Well enough rambling.......

_RES	EQU	#0
_W	EQU	#1
_L	EQU	#2
_CUR	EQU	#3

	ldp.ab	_RES,	check
	ldp.ab	_W,	wins
	ldp.ab	_L,	losses
	ldp.a	_CUR,	launch
check	sne.ab	#0,	#0	
losses	add.ab	#2,	#0      ;loss value
	sne.ab  #1,	check
wins	add.ab  #5,	#0      ;win value
	stp.b	losses, _L
	slt.b	losses,	wins
switch	add.a	#1,	launch
	mod.a	#3,	launch  ;# of components
	stp.ab	launch,	_CUR
	stp.b	wins,	_W
launch	jmp	@1,	scissor	;3rd
	dat	0,	paper	;1st
	dat	0,	rock	;2nd


paper	jmp 0
rock    jmp 0
scissor jmp 0

-- 







From: sieben@imap1.asu.edu
Subject: Re: FAQ frequency
Date: 1995/09/01
Message-ID: <4265qh$3pb@news.asu.edu>#1/1
references: <9508241558.AA03078@idnsun.gpct.Vanderbilt.Edu.noname> <422hfp$ddu@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

How about a mini FAQ that mainly tells about the location of the faq and 
soda. This could be posted more frequently without the overload.

Nandor.




From: pk6811s@acad.drake.edu
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/02
Message-ID: <1995Sep2.184553@acad.drake.edu>#1/1
references: <429s4o$87u@mikasa.iol.it>
newsgroups: rec.games.corewar

(Beppe Bezzi) writes:
> Having got the O.K. from Randy, I hope Paul has nothing against it,

Beppe, this is a very fine analysis and I learned something from it.....
which I'm going to incorporate into Pink :-)

Which is to put its 'gate' on a mod-4 distance from the spl. 
A neat idea.  Thanks.  Sometimes it's the little things.

(Of course you could go with a mod-1 step which approximates a mod-4 by having
a very low find-4 number)

---

And in answer to Stefan's question of several weeks ago, "Of what use is
communication between warriors in a 2-warrior battle?"  Simple, make
them fight a win-or-lose strategy instead of a tying one so they both
get 50 wins instead of 100 ties.  

In my pspace switcher, recording a win takes a little longer than a loss,
so the loser records his score first and the winner overwrites it after.
So both think they are winning and stick with that strategy.  No extra
work or lines, just adding a PIN # caused Pink to go from scoring
20/20/60 against himself to 55/0.

Actually I'm not sure if self-plays are counted except in the first 
submission.  When the Hill is challenged by another player, does
each of the 'presiding' programs count its own self-play in its score?

Paul Kline
pk6811s@acad.drake.edu



From: bezzi@iol.it (Beppe Bezzi)
Subject: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/02
Message-ID: <429s4o$87u@mikasa.iol.it>
newsgroups: rec.games.corewar

Having got the O.K. from Randy, I hope Paul has nothing against it,
I'm publishing how I managed to score 56/42/2 wins against Porch Swing
and 61/36/3 against Whitershins Thrice; all this with a warrior,
Tornado 1.2, scoring well 97.2 points on the 94 hill.

Whitershins, and then Porch, shocked the hill when they entered. They
got really high scores and, most important, outscored nearly every
other warrior with the only exception of Agony. Having to make a
multi-component p-warrior, and hoping to score very high <g>, I began
thinking at how to beat these two deadly guys with, at least, one of
my components. Paper, perhaps the kind of warrior I can make better,
was not effective against their deadly spl-spl-dat clear, so were
stones/imp (Blue Funk and Juliet scored low with them); the only
warrior scoring well was Agony but I was (am) not able to make one so
good and older cmp-scan were not effective against PS and WT; fast
bombers like HeremScimitar and his offspring Copyright notice, scored
well but not well enough; maybe we are on the right way;  O.K. I need
a specialized warrior able to hit them where they are softer.

Let's give a short look at the code and how it works. 

;redcode-94
;name Porch Swing
;kill Porch Swing
;Author Randy Graham
;assert 1
;strategy Swing with a little wider range.
;strategy Now 80% bomb/scan with djn-stream once-thru

STEP    equ     14	;guess Randy didn't published it
EXTRA   equ     4
DJNOFF  equ     5000	;still gues but I don't think it changes a lot

        dat.f  #gate-10,  step-gate+5
gate    dat.f  #gate-10,  sneer-STEP+1
dat2    dat.f  #gate-20,  step-gate+5
dat1    dat.f  #gate-25,  step-gate+5
site2   spl.a  #gate-30,  step-gate+5
site    spl.a  #gate-40,  step-gate+5
for EXTRA
        dat.f  0,         0
rof

adder   sub.f  sweeper,   sneer
hithigh mov.i  step,      *sneer
hitlow  mov.i  step,      @sneer
sneer   sne.i  @gate+STEP*6-1-EXTRA, *gate+STEP*3-EXTRA  ;so we bomb
step
looper  djn.b  adder,     <DJNOFF
setup   add.f  sneer,     gate
sweeper spl.a  #-STEP*4+1,<-STEP*4+1
mover   mov.i  @swinger,  >gate
swinger djn.b  mover,     {site
step    dat.f  <STEP,     <-STEP

end     hithigh

W.T. is very similar, both drop two bombs and use them as pointers for
a sne instruction (2 bombs and 2 scans in 5 instructions, very fast),
when the sne finds something they begin claring core in this position.
They begin with spl 0 (spl #something is same as spl 0) to stun
opponent then, when the djn stream reaches -swinger- with spl again
and then dat forever. Imps are attacked when they reach the gate
instruction and the clear continues where the imp is copying itself
(the imp became the pointer of the clear) very deadly and still very
simple.

We have seen the strengths, now the weaknesses. 
-It is a single process warrior, so a single hit is enough to kill it.
-It is a rather large target, at least while it's still
scanning/bombing.
-Once it begin the clear its speed reduces from 80%c to less than 50%c
even if its size reduces too from 10 (including step but not gate) to
4 instructions plus the gate or better 3 plus the gate, hitting the
djn don't stop the clear)
It scans bombs once thru, and it's forced to do so forward not to hit
himself, so we can hide behind a decoy.
How to attack this monster? We have some different options.
1- Be as small as I can, hoping to pass through the scan and hit him
while it's still big. All my trials with a 3/4 lines stone aborted,
maybe someone else will be more successful.

2- Try to force him clearing core and attack the core clear.

3- Bomb faster that it does. (80% c is nearly a top limit)

4- A mix of the above. More or less what I did.

I resumed Tornado, a very fast bomber I wasn't able to make function
well, I made it boot near 1500 lines forward a big decoy to be sure I
finish my bombing run, and I gave him bombs able to attack the clear
once started.
Now Tornado (again; I published it together with K.Y.E.)
After boot it looks so:

gate    dat     #step, #-step		;step equ 52
   
start   mov     bombd+20,   *stone
	mov     bombd+20,   @stone
stone   mov     *(2*step)+1, *(3*step)+1         
	add     incr+50,stone           
jump    djn.b   start,  #count		;count equ 665
	spl     #step,  #0
clr     mov     gate,   }gate-5
;....				
;....some space away
;....	
incr    dat     3*step, 3*step        

bombd   dat     #step, #1         ;hit dat

Tornado drops 3 bombs in 5 instructions, using the first as pointer
for the third, dropping near 2000 bombs in 3325 cycles, the time P.S.
clear starting at the decoy reaches me. The step 52 is a tight modulo
4 with a good find 8 and find 12 number. If I manage to hit PS while
it's still bombing well, it's dead if not look at what could happen;
now P.S. executing part is so:

 1-gate    dat.f  #gate-10,  sneer-STEP+1

some instruction not vulnerable

16-sweeper spl.a  #-STEP*4+1,<-STEP*4+1
17-mover   mov.i  @swinger,  >gate
18-swinger djn.b  mover,     {site
19-step    dat.f  <STEP,     <-STEP

Instruction 18 in not vulnerable. Even if I hit it the clear
continues, only I die immediatly (without struggling for the spl)
because it uses a dat (line 19) to overwrite me.

Instructions 1 and 17 are in the same modulo pattern in P.S. (not in
W.T.) Hitting them I stop the clear, no more mov, and my clear will
soon finish the paralized opponent.

Instruction 19 is my other unlucky chance, I missed it, can only hope
my clear will reach him before its reaches me.

Instruction 16 is the most subtle chance. Hitting it the clear
continues, but it's no more feed of processes by the split. So every
time its djn train reaches a b-field of 1, and there are as many as I
laid bombs, very many, it loses a process and soon or after, dies of
starvation, sometimes after having overwritten me with spl 

Whitershins case is a bit different, it's two lines longer so I get an
extra chance, hitting the -gate- instruction. Doing so causes the
clear self destruct.

gate	dat  52,   1
	.... 
clr	mov  @1,   >gate
	djn  -1,   {wipe1

W.T. will start clearing gate+1 then gate+2, .. the end.

It's finished. I only hope, after looking this dissection of a
warrior, others will continue publishing their warriors ;-) I'll do.

Soon coming "How to beat Marcia Trionfale" by Randy.


-Beppe Bezzi

bezzi@iol.it


--------
Beppe Bezzi
bezzi@iol.it

Gaudeamus igitur..




From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: PVamps
Date: 1995/09/02
Message-ID: <1995Sep2.093404.2850@rhodes>#1/1
newsgroups: rec.games.corewar

Last night I had visions of a PVamp.  I don't know that this is
something I will use, but I will certainly protect my warriors against
this segment:

      spl.a   #0,   <-12
      stp.ab  #500, #-1
      jmp.a   -2,   <-1

Any warrior caught by this that uses a jmp.a @0, str1 will never win
again once hit.  The easiest defense is to ALWAYS mod your strategy
before using it.  I can asuure you my pspace warriors have been
protected against this for future reference.  They may get their
counters messed up, but they won't be jumping off into random spots in
core at least.  Just thought I'd share.

Randy



From: jwilkinson@utexas.edu
Subject: (no subject)
Date: 1995/09/02
Message-ID: <428nu0$mju@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

What is the best scanner out there right now?  Are there any scanners in
existence which can destroy even a really strong silk?

I've been looking at Rave's interraction with my COTTON series of
silk-ish warriors.  For some subtle reason, Rave isn't able to go
quite fast enough to kill off all the processes.  It usually leaves
about 2 to 10 behind.  Those then proceed to "go forth and be fruitful." :)

This generally results in the death of Rave.  Is there someway to do
a nice SPL/DAT carpet on the core, without adding to the size?  That
would do the trick nicely...

-JKW




From: jwilkinson@utexas.edu
Subject: Re: FAQ frequency
Date: 1995/09/02
Message-ID: <428nmd$mju@geraldo.cc.utexas.edu>#1/1
references: <4265qh$3pb@news.asu.edu> <428j54$ltp@newsbf02.news.aol.com>
newsgroups: rec.games.corewar

>I think that posting the FAQ weekly would be best.  Once or twice a month
>is not enough, but daily is annoying.  But a mini FAQ would just be
>annoying.  It's easier just to post the whole thing instead of worrying
>about 2 different things to post

If you posted a mini-FAQ, then you would no longer need to post the
big FAQ.  And, why would it be annoying?  It'd be a small message,
so it wouldn't take long to receive, and after once or twice, you'd
recognize the subject, and skip over it, eh?

-JKW




From: lewin@netcom.com (Karl Lewin)
Subject: Re: P-Space Switcher
Date: 1995/09/02
Message-ID: <lewinDE942n.CA7@netcom.com>#1/1
references: <lewinDE87uo.42w@netcom.com> <1995Sep1.090321@acad.drake.edu>
newsgroups: rec.games.corewar

pk6811s@acad.drake.edu wrote:
: (Karl Lewin) writes:

: Your switcher looks very credible.  And it is innoculated against......
: by ALWAYS performing MOD on the strategy before doing the jump.  Yeesh.
: The report is lost, but someone was beating my test 99 out of 100 until
: figured this out.  Randy???  Beppe???  one of you guys is too too .....

One addition that I made was to always stay with a winner.  There were 
certain situations where the "correct" strategy lost the first time 
through and then the switcher would never "catch-up" (loss points would 
always be higher than win points)  So I made a change and any strategy 
that wins always get another shot.  This seemed to help stengthen things 
overall.
-- 







From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/03
Message-ID: <42d8v0$tbq@mikasa.iol.it>#1/1
references: <429s4o$87u@mikasa.iol.it> <1995Sep2.184553@acad.drake.edu>
newsgroups: rec.games.corewar

pk6811s@acad.drake.edu wrote:

>(Beppe Bezzi) writes:
>> Having got the O.K. from Randy, I hope Paul has nothing against it,

>Beppe, this is a very fine analysis and I learned something from it.....
>which I'm going to incorporate into Pink :-)

>Which is to put its 'gate' on a mod-4 distance from the spl. 
>A neat idea.  Thanks.  Sometimes it's the little things.

>(Of course you could go with a mod-1 step which approximates a mod-4 by having
>a very low find-4 number)

And about a mod-5 step ? Yes you can put gate 20 lines far from the
split and be good for both but, so doing, you lenghten your warrior
and:
1- I found difficult to put three warriors, their boot code, and a
switching routine in but 100 lines and I think you too.
2- You expose more yourself to replicator launched streams of anti imp
bombs  - dat <2*2667, <2667 - that create, hitting your clear a large
safe area in which they can live in peace.

Seems that, in Corewar, every attack form has its counter measures.
Not long ago it seemed that fast scanners with forward multipass
clears were the ultimate weapon, now we have an 'old style' stone like
juliet storm outscoring both Porch and Whitershins, I think because
it's very good against their enemies, that are growing in number. 
The game is, IMO, very well balanced and even the now dominating
p-space warriors will have hard times at throwing out the best
standard warriors.


>---
>...
>work or lines, just adding a PIN # caused Pink to go from scoring
>20/20/60 against himself to 55/0.

>Actually I'm not sure if self-plays are counted except in the first 
>submission.  When the Hill is challenged by another player, does
>each of the 'presiding' programs count its own self-play in its score?

Thanx. May be this is the solution I needed for 1/6/93 problem of KYE
against itself.

I made some calculations and seems that the myself result is kept in
the score, so better have a *decent* one seen that this is the only
result one is sure to keep forever ;-)

>Paul Kline
>pk6811s@acad.drake.edu

--------
Beppe Bezzi
bezzi@iol.it

Gaudeamus igitur..




From: tpoindex@nyx.cs.du.edu (Tom Poindexter)
Subject: Re: "Robots" games
Date: 1995/09/03
Message-ID: <42duaj$6sc@nyx.cs.du.edu>#1/1
references: <41v6j9$2jf@newsbf02.news.aol.com>
newsgroups: rec.games.corewar

In article <41v6j9$2jf@newsbf02.news.aol.com>,
MSMoulton <msmoulton@aol.com> wrote:
>I have CRobots, Pascal Robots 3.1, and PC Robots 1.41.  Am I using the
>newest versions?  Also, where can I get C++Robots and what is RoboWar?


RobotWars was a game for the Apple ][, circa 1983 or so.  It was the program
that gave me the inspiration to write CROBOTS.  Language used was a proprietary
one, similar to Basic, a limited number of variables, (A-Z, if I recall 
correctly),  simple IF cases (one statement, usually an assignment (STORE) or
GOTO or GOSUB).

Another 'robot' type game is my TclRobots, avaialable at
	ftp://ftp.aud.alcatel.com/tcl/code/tclrobots-1.0.tar.gz

TclRobots is written in and uses the Tcl language, available via ftp for most
Unix/X11 systems.  Tcl is also available on the same ftp machine.
-- 
Tom Poindexter   
tpoindex@nyx.cs.du.edu
http://nyx10.cs.du.edu:8001/~tpoindex/



From: sriverhi@snowcrest.net (Ananda)
Subject: Re: PCROBOTS
Date: 1995/09/03
Message-ID: <42d1oe$ar5@news.snowcrest.net>#1/1
references: <41ceoe$pji@ra.ins.de> <41ln94$7ro@news.snowcrest.net> <41t69o$icm@agate.berkeley.edu>
newsgroups: rec.games.corewar

In article <41t69o$icm@agate.berkeley.edu>, mconst@soda.CSUA.Berkeley.EDU 
says...

>Actually, discussion of PCRobots, C++Robots, RoboWar, etc. is welcome
>here.  The rec.games.corewar charter says that this group is for the
>discussion of "corewar and related games".  Just about any programming
>game qualifies as "related" to corewar, and it's not as if we're over-
>whelmed with traffic here anyway :-)

Ahh... Ok.  It is really nice to read a postitive posting like this :)
Does anyone have a filename for C++Robots?  I have never seen it.  Thanks.

-Ananda




From: everard@infi.net (M. Scott Everard)
Subject: Re: FAQ frequency
Date: 1995/09/03
Message-ID: <42bark$qs5@allnews.infi.net>#1/1
references: <4265qh$3pb@news.asu.edu> <428j54$ltp@newsbf02.news.aol.com> <428nmd$mju@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar


What if the mini-FAQ simply gave the ftp address for the FAQ?


Just a thought.

Scott



From: kb5kjn@PROBLEM_WITH_INEWS_DOMAIN_FILE (Bill Jones)
Subject: Programmers Game
Date: 1995/09/03
Message-ID: <42cp26$jc5@news.blkbox.com>#1/1
newsgroups: rec.games.corewar

Im working on a new Programmers Game.

It is a cross between Crobots and Omega. Thus 

   I. Program is written in asmembly
      A. Like DSP
         1. CPU is a Floating Point CPU
  II. Tanks are not all alike
      A. Diffrent classes of tanks
         1. power plants
         2. movement rates
         3. kinds of armor
      B. Diffrent kind of weapons
         1. Damage done per hit
         2. Rate of fire
         3. Range
      C. Diffrent kinds of scanners
         1. Adjustable range
         2. Adjustable width
 III. Graphical User interface 
      A. Battle field is large
      B. There are obsticals
  IV. Player
      A. Single PC
      B. Internet Play
   V. Program Binaries will be FREE


Currently, I am writting for the Linux platform.

If anyone is would like to help code this game with me, I would love the help
please reply via email if you would. My address in kb5kjn@kb5kjn.com

The current working instructions that have been coded are as follows.

#DATA = hard coded data
<direct address> = the contents @ <address>
<indirect address> = the contents of the address stored in <address>
<absolute> = the precise address
<relitive> = the program counter + <address>

OPCODE	GROUP	SUB OPCODE	ACTION

  1	ADD
		    1		add ACC, #DATA
		    2		add ACC, <direct address>
		    3		add ACC, <indirect address>

  2	SUB
		    1		sub ACC, #DATA
		    2		sub ACC, <direct address>
		    3		sub ACC, <indirect address>

  3	MUL
		    1		mul ACC, #DATA
		    2		mul ACC, <direct address>
		    3		mul ACC, <indirect address>

  4	DIVIDE
		    1		div ACC, #DATA
		    2		div ACC, <direct address>
		    3		div ACC, <indirect address>

  5	MOV
		    1		mov <direct address>, #DATA
		    2		mov <direct address>, <direct address>
		    3		mov <direct address>, <indirect address>

  6	JMP
		    1		jmp <absolute address>
		    2		jmp <relitive address>

  7	CMP
		    1		cmp ACC, #DATA, <relitive>
  		    2		cmp ACC, <direct address>, <relitive>


I will tank <yes thats a pun> all suggestion and try to put them into the 
game.

Thanks Bill
kb5kjn@kb5kjn.com




From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - Multiwarrior Experimental 94
Date: 1995/09/04
Message-ID: <199509040400.AAA01652@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information


Current Status of the StormKing Industries MultiWarrior Experimental 94 CoreWar Hill:

 #                      Name               Author   Score     Age
 1          TimeScapeX (0.1)       J. Pohjalainen    3550      32
 2                  Lucky 13        Stefan Strack    3105      34
 3                  Paperone          Beppe Bezzi    3097      17
 4                     jaded           M R Bremer    2888       3
 5                      life        Nandor Sieben    2663      33
 6                 lifedwarf        Nandor Sieben    2583       9
 7            Miss Treatment           Derek Ross    2344      12
 8                    Hidden M.C.Diskett Bullfrog    2263       2
 9             Miss Careless           Derek Ross    2222       1
10                 Whirlwind              Bob Uhl    2161      19
11                   Shwing!         T. H. Davies    2148      28
12                   Piggy 3              Bob Uhl    1857       5
13            AB Scanner 2.9         Chris Hodson    1840      21
14                   Piggy 2              Bob Uhl    1785       7
15            Illusion-94/55         Randy Graham    1656      11
16        Miss Understanding           Derek Ross    1616       8
17                Miss Carry           Derek Ross    1556      13
18               Whirlwind 2              Bob Uhl    1367      20
19                   Unknown            Anonymous    1130      10
20                   MuDwarf             G. Eadon     809      22

21                Veeble Jr.         T. H. Davies     785      29



From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Re: PVamps
Date: 1995/09/04
Message-ID: <1995Sep4.205750.2851@rhodes>#1/1
references: <1995Sep2.093404.2850@rhodes> <42fjic$4jh@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu wrote:
: > [I wrote...]
: >      spl.a   #0,   <-12
: >      stp.ab  #500, #-1
: >      jmp.a   -2,   <-1

: Well, although that's an interesting idea, I have my doubts as to the
: effectiveness.  I mean, after a SPL 0's, you'll be skipping over large
Yes, I realized that after I posted.  How about:

     stp.ab   #500,  #-1
     spl.a    -1,    <-12
     jmp.a    -1,    <-2

Tried it and it works correctly.  By the way, I am working on a vamp
using this, but using a mov.i  x, <x for the stp, and a jmp.a -1, }x
for the jmp.  Make 'x' be another fang, and this will run a sweep of
fangs through core.  Almost made the hill first try with it.  I think
I can get it on there, but I'm not sure.  Also had a new idea for a
core-clear that is smaller than that in Withershins and Porch, but
works even in '88.  Still working on that one.

: -JKW

Randy



From: jwilkinson@utexas.edu
Subject: (no subject)
Date: 1995/09/04
Message-ID: <42fv4g$cgi@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Hey all you KOTH server guys,
  Is there any way to tell the KotH server to either

1) Stop sending me mail from a warrior that I sent as "verbose"?
 
 or

2) Send the mail to a new location?

If not, can you add this feature?


Also, is there any protection whatsoever against me faking mail from
"stst@vuse.vanderbilt.edu" and putting ";kill Agony II" as the first
line?  I know, it's a cheap little thing to do, but could the server
stop me?




From: jwilkinson@utexas.edu
Subject: Pvamp
Date: 1995/09/04
Message-ID: <42fked$5cu@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

I rethought the subject.  How about this one?
I think it'd be reasonably effective.

Assuming you get a vamp strike early, this pretty much should clear the entire
Pspace, should it not?

stp #666, #0
spl -1, <-100
jmp -1, <-2




From: jwilkinson@utexas.edu
Subject: Re: PVamps
Date: 1995/09/04
Message-ID: <42fjic$4jh@geraldo.cc.utexas.edu>#1/1
references: <1995Sep2.093404.2850@rhodes>
newsgroups: rec.games.corewar

>      spl.a   #0,   <-12
>      stp.ab  #500, #-1
>      jmp.a   -2,   <-1
>
>Any warrior caught by this that uses a jmp.a @0, str1 will never win
>again once hit.  The easiest defense is to ALWAYS mod your strategy
>before using it.  I can asuure you my pspace warriors have been

Well, although that's an interesting idea, I have my doubts as to the
effectiveness.  I mean, after a SPL 0's, you'll be skipping over large
segments of pspace in between STP's.  Perhaps a SPL 1 arrangement might
work better.

-JKW




From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: "Robots" games
Date: 1995/09/04
Message-ID: <42ekpu$nma@agate.berkeley.edu>#1/1
references: <41v6j9$2jf@newsbf02.news.aol.com> <42duaj$6sc@nyx.cs.du.edu>
newsgroups: rec.games.corewar

Tom Poindexter <tpoindex@nyx.cs.du.edu> wrote:
>MSMoulton <msmoulton@aol.com> wrote:
>>I have CRobots, Pascal Robots 3.1, and PC Robots 1.41.  Am I using the
>>newest versions?  Also, where can I get C++Robots and what is RoboWar?
>
>RobotWars was a game for the Apple ][, circa 1983 or so.

To clarify: RoboWar (the original poster's spelling was correct) is a
modern robot-programming game for the Mac.  It uses an assembler-level
language which is not quite like any real assembler (to execute commands,
you push operands onto a stack; commands take operands off the stack and
put return values back on the stack).  It's a very fun game, and it makes
a pleasant diversion from corewar.

If you are interested in RoboWar, an archie for "robowar*" should find
it.  I believe the current version is 4.1.
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - ICWS Tournament
Date: 1995/09/04
Message-ID: <199509040400.AAA23910@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information


Current Status of the StormKing Industries Annual ICWS Tournament CoreWar Hill:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  61/ 11/ 28                 Cannonade           Paul Kline     212      56
 2  53/ 36/ 11             Miss Careless           Derek Ross     171      10
 3  50/ 39/ 12            Old Tire Swing         Randy Graham     161      13
 4  48/ 39/ 14          DoubleStone v0.6    Christoph C. Birk     157       2
 5  49/ 42/ 10                Miss Carry           Derek Ross     155      20
 6  49/ 44/  7                   Agony T        Stefan Strack     154      57
 7  43/ 33/ 24              Giskard v0.5           Ken Mitton     154      29
 8  42/ 45/ 13              Slaver v1.1i    Christoph C. Birk     139      17
 9  40/ 41/ 19                warrior 42      stefan roettger     139      69
10  39/ 39/ 23                     stone  matthew householder     139      68
11  42/ 47/ 11            Miss Treatment           Derek Ross     138      21
12  41/ 45/ 13                 Maya v1.6    Christoph C. Birk     137      30
13  44/ 52/  4                       xtc      stefan roettger     136      61
14  35/ 46/ 19                  scissors  matthew householder     123      71
15  37/ 54/ 10                  Illusion         Randy Graham     120      15
16  34/ 52/ 14                  Cat v2.0           Tim Scheer     117      53
17  34/ 54/ 12             Smartbomb 4.0     Devin Kilminster     113      55
18   3/ 48/ 49          Mr Administrator           Derek Ross      58       1
19   4/  3/  3                      test    Christoph C. Birk      14       5
20   4/  3/  3                      test    Christoph C. Birk      14       8

21   3/  0/  1        Miss Understanding           Derek Ross      12      12



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - Standard
Date: 1995/09/04
Message-ID: <199509040400.AAA25176@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information

Current Status of the StormKing Industries Standard KotH CoreWar Hill :

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  30/ 15/ 55              Blue Funk 88       Steven Morrell     144      57
 2  27/ 12/ 61                 Cannonade              P.Kline     142      93
 3  29/ 18/ 53   CAPS KEY IS STUCK AGAIN       Steven Morrell     141      59
 4  30/ 21/ 49    Der Zweiter Blitzkrieg      Mike Nonemacher     140      88
 5  35/ 31/ 34              Keystone t21              P.Kline     139      80
 6  27/ 14/ 59                     Peace            Mr. Jones     139      67
 7  28/ 20/ 51                      Test       Wayne Sheppard     137      82
 8  41/ 45/ 14                  bigproba        nandor sieben     136      81
 9  36/ 37/ 27               Christopher       Steven Morrell     135      58
10  37/ 41/ 22              Request v2.0     Brant D. Thomsen     134      29
11  24/ 14/ 62                      ttti        nandor sieben     134      43
12  34/ 35/ 31              Giskard v0.5           Ken Mitton     134      31
13  39/ 44/ 18                 Iron Gate       Wayne Sheppard     134     187
14  21/  9/ 70         Imps! Imps! Imps!       Steven Morrell     133     104
15  36/ 40/ 24       Beholder's Eye V1.7         W. Mintardjo     133     137
16  26/ 19/ 55                     Hydra      Stephen Linhart     132     167
17  24/ 18/ 58           jmp/add crasher         Randy Graham     131      17
18  34/ 38/ 29             Miss Careless           Derek Ross     129       3
19  18/ 12/ 70              Evol Imp v5a            Wilkinson     125       4
20  10/ 61/ 29             stone thrower        Darren Bergen      60       1

21   0/ 46/ 54          Mr Administrator           Derek Ross      54       2



From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Re: pMARS Feature
Date: 1995/09/05
Message-ID: <1995Sep5.211020.2856@rhodes>#1/1
references: <42hu40$kq6@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu wrote:
: Clear enough?  I want to test my warrior's ability to recover from
: a hypothetical attack which blanked/scrambled my warrior's pspace
: in the previous round.
Well, there's always the pvamp I published a couple of days ago...

: -JKW

Randy



From: tom@halls1.cc.monash.edu.au (Tom Paton)
Subject: Re: FAQ frequency
Date: 1995/09/05
Message-ID: <tom.14.304C024E@halls1.cc.monash.edu.au>#1/1
references: <4265qh$3pb@news.asu.edu> <428j54$ltp@newsbf02.news.aol.com> <428nmd$mju@geraldo.cc.utexas.edu> <42bark$qs5@allnews.infi.net>
newsgroups: rec.games.corewar

In article <42bark$qs5@allnews.infi.net> everard@infi.net (M. Scott Everard) writes:
>What if the mini-FAQ simply gave the ftp address for the FAQ?
>Just a thought.
>Scott

how about just a subject line with _no message_ such as:

	SUBJECT: FAQ @ ftp.blah.blah.blah

	?
Wouldn't take anytime to look at it and virtually no bandwidth...

personally I would say that posting a FAQ list every day or so would be more 
annoying that the FAQ's themselves...

ciao,


-Tom
   tom@halls1.cc.monash.edu.au
   http://yoyo.cc.monash.edu.au/~madmick
-={ My opinions are mine and should be yours }=-




From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/05
Message-ID: <199509051855.TAA09763@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

Paul Kine wrote:


>Had some more thoughts about those mod-4 alignments.  In the example
>given the two locations, 'gate' and 'spl' were the ones most vulnerable
>to attack.  If either is hit the program will die, and the question is:
>what is the best spacing between them if the opponent is bombing with
>a mod-4 step?

>If they are spaced a mod-4 distance (4,8,12,16...) then in three out
>of four battles both will be safe - the opponent will be bombing 1, 2, or
>3 locations off from them.  In the other one-of-four battles the
>danger is doubled, since both are now in the path of the bombing stream.

>If they are not spaced a mod-4 distance then only in two of four battles
>are both safe.  In the other two-of-four, one of them is in the path
>of the bombing stream, but only one.

>So maybe it's a toss-up.

Maybe the question is: what's the chance I have to catch my opponent
before it finishes the bombing run, i.e. bombs every mod-4 location.
In case of a fast bomber, Tornado bombs the 2000 mod-4 locations in
3350 cycles, the time a clear fills 1500, is better to be aligned; in
case of a slower stone the breakpoint is at speed near 25%
I don't know for sure, but most stone are faster so it's better be
aligned too.

>Beppe Bezzi wrote:
>> 1- I found difficult to put three warriors, their boot code, and a
>> switching routine in but 100 lines and I think you too.

>..
>with other styles of pspacers.  Take Torch and vary the amount of
>bombing before sweeping.  Could be done with very little extra code.

Isn't it always better to bomb the longer time possible, at least with
incendiary bombs?

>> clears were the ultimate weapon, now we have an 'old style' stone like
>> juliet storm outscoring both Porch and Whitershins, I think because

>What's 'old style' about Juliet Storm?   A nice spl-spl stone
>with an imp-spiral.  Very sheek and modern :-)

I have nothing against juliet, this deadly little thing is a really
fine piece of code. I was saying that in the, short, time I'm playing
corewar I have seen many fast scanners challenging the hill and but
one, Night Crawler if I remember well, stone with imps; I have seen
many stone-imps in last year issues of the newsgroup when imps where
the coolest argument of discussion. 'Old style' is a very relative
concept in a fast paced game like Corewar.

BTW some months are missing at csua, no chance to get them? and old
hill results? I'm curious to see how programs like Blue Funk and Agony
did when first appeared on hill.

>Paul Kline





From: jwilkinson@utexas.edu
Subject: pMARS Feature
Date: 1995/09/05
Message-ID: <42hu40$kq6@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Nescessity being the mother of invention, I've stumbled blindly upon
a pMARS feature that I now need.  An inter-round Pspace scrambler.

Clear enough?  I want to test my warrior's ability to recover from
a hypothetical attack which blanked/scrambled my warrior's pspace
in the previous round.

I could always do it by hand, or add some code to the front of my
warrior to do it myself.  However, the former it slow, and the
latter gives my opponent quite a bit of extra time to be doing something
while mine is just blanking it's own pspace...

-JKW




From: jwilkinson@utexas.edu
Subject: Re: FAQ frequency
Date: 1995/09/05
Message-ID: <42htp1$kes@geraldo.cc.utexas.edu>#1/1
references: <4265qh$3pb@news.asu.edu> <428j54$ltp@newsbf02.news.aol.com> <428nmd$mju@geraldo.cc.utexas.edu> <42bark$qs5@allnews.infi.net> <tom.14.304C024E@halls1.cc.monash.edu.au>
newsgroups: rec.games.corewar

>how about just a subject line with _no message_ such as:
>
>	SUBJECT: FAQ @ ftp.blah.blah.blah

>	?
>Wouldn't take anytime to look at it and virtually no bandwidth...

>personally I would say that posting a FAQ list every day or so would be more 
>annoying that the FAQ's themselves...

Yeah right, as though people will look at that and
immediately understand what that means?  To the
contrary, it would likely increase the confusion
of corewar newbies... which isn't anybody's goal...

-JKW




From: pk6811s@acad.drake.edu
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/05
Message-ID: <1995Sep5.092308@acad.drake.edu>#1/1
references: <429s4o$87u@mikasa.iol.it> <1995Sep2.184553@acad.drake.edu> <42d8v0$tbq@mikasa.iol.it>
newsgroups: rec.games.corewar


Had some more thoughts about those mod-4 alignments.  In the example
given the two locations, 'gate' and 'spl' were the ones most vulnerable
to attack.  If either is hit the program will die, and the question is:
what is the best spacing between them if the opponent is bombing with
a mod-4 step?

If they are spaced a mod-4 distance (4,8,12,16...) then in three out
of four battles both will be safe - the opponent will be bombing 1, 2, or
3 locations off from them.  In the other one-of-four battles the
danger is doubled, since both are now in the path of the bombing stream.

If they are not spaced a mod-4 distance then only in two of four battles
are both safe.  In the other two-of-four, one of them is in the path
of the bombing stream, but only one.

So maybe it's a toss-up.

Beppe Bezzi wrote:
> 1- I found difficult to put three warriors, their boot code, and a
> switching routine in but 100 lines and I think you too.

It's a problem with multi-component pspacers, but maybe not so much
with other styles of pspacers.  Take Torch and vary the amount of
bombing before sweeping.  Could be done with very little extra code.

> 2- You expose more yourself to replicator launched streams of anti imp
> bombs  - dat <2*2667, <2667 - that create, hitting your clear a large
> safe area in which they can live in peace.

Well, Timescape won't 'live' in the space between the gate and the
clear-code.  It executes its code once and moves on.  But others
might do that.

> Seems that, in Corewar, every attack form has its counter measures.
> Not long ago it seemed that fast scanners with forward multipass
> clears were the ultimate weapon, now we have an 'old style' stone like
> juliet storm outscoring both Porch and Whitershins, I think because

What's 'old style' about Juliet Storm?   A nice spl-spl stone
with an imp-spiral.  Very sheek and modern :-)

Paul Kline
pk6811s@acad.drake.edu



From: jwilkinson@utexas.edu
Subject: Re: pMARS Feature
Date: 1995/09/06
Message-ID: <42kbai$cdf@geraldo.cc.utexas.edu>#1/1
references: <42hu40$kq6@geraldo.cc.utexas.edu> <1995Sep5.211020.2856@rhodes>
newsgroups: rec.games.corewar

>: Clear enough?  I want to test my warrior's ability to recover from
>: a hypothetical attack which blanked/scrambled my warrior's pspace
>: in the previous round.
>Well, there's always the pvamp I published a couple of days ago...
>
>Randy

Well, how am I supposed to pvamp my program every round, and still
get an effective count of how it performs?

Yes, I supposed I could go to the debugger, make it jump to the vamp,
and then, the next round make it not do that.

Of course, that would involve keeping track of statistics on a sheet
of paper...

Please, if anyone can think of a simple way around this, I'd be glad
to hear it...



From: rognlie@lute.gcr.com (Richard W. Rognlie)
Subject: Richard's C++Robots Server Monthly Post
Date: 1995/09/06
Message-ID: <42krp9$9j1@ukelele.qnet.com>
newsgroups: rec.games.corewar,comp.lang.c,comp.lang.c++,rec.games.programmer

Richard's C++Robots Server Monthly Posting

A generic Play-By-eMail Server has been set up at pbmserv@vtsu.prc.com.  It
currently supports a variety of games.  Of particular interest to this forum
is C++Robots.

To get more information send mail to pbmserv@vtsu.prc.com with 'help' as the
subject line.

Games Currently Supported

C++Robots (Copyright (c) 1994 Richard Rognlie)

    An ongoing "King of the Hill" (KotH) tournament in which players
    use ANSI C or C++ to create a control program for a robot.  Your
    robot then fights each of the other robots "on the hill".  If you
    do well enough, your robot will "make the hill", bumping the lowest
    robot from the hill.

    Robots have the ability to scan for opponents, fire a cannon, move,
    and determine current position and status.

    Conceptually based on C-Robots written for the IBM PC by Tom Pointdexter.

Abalone

    On a hexagonal board (radius 5) two to six players have armies of
    marbles.  Players take turns "pushing" 1, 2 or 3 linearly connected
    marbles, attempting to push their opponents' marbles off the board.

Ataxx 

    On a 7x7 board, the two players of ataxx fight to controll a majority
    of the board via growth and jumps that flip opponents pieces to their
    color.

Backgammon
  
    A classic.

Connect4
 
    On a 7x6 board, two players alternate dropping their pieces from the top
    of the board, down a column, attempting to form four in a row.

Connect4x4

    On an 8x8 board, two players alternate inserting their pieces from the
    edges of the board, across a row or up/down a column, attempting to form
    four in a row.

Gomoku

    On a 15x15 board, the two players of gomoku try to be the first
    to create a line of 5 or more stones in a row of their color.

Hex

    On a 11x11 diamond board, players take turns placing stones of
    their color on the board. The object is to connect your sides of
    the board while preventing your opponent from doing the same.

Jungle

    Jungle is sort of a cross between Chinese chess and Stratego. It's
    popular in China as a children's "stepping-stone" to Chinese chess.
    It's also an interesting game in its own right.

Lines-of-Action

    The object of the game is to move all your pieces into a configuration
    where they are in a single group connected horizontal, vertically, or
    diagonally.  Pieces may move horizontally, vertically, or diagonally,
    but they must move exactly the number of spaces as there are pieces
    on the row they are moving in.  You may not jump over opponent's
    pieces, nor may you land on your own piece.  If you land on an
	opponent's piece, it is captured and removed from the game.

Neutron (Copyright (c) 1978 Charles Wetherell)

    On a 5x5 board, the two players of neutron fight to either move the
    neutron to their back row or trap it so the opponent cannot move it.

    The winner is the player who is able to trap the neutron or gets the
    neutron to his or her own back row. It does not matter if it is your
    opponent who moves the neutron to your back row -- you still win.

Othello (Copyright (c) 1973,1990  Pressman Toy Co.)

    On a 8x8 board, the two players of othello fight to control the majority
    of the board by outflanking and flipping their opponents pieces.

Pente

    On a 19x19 board, the two players of pente try to be the first
    to create a line of 5 or more stones in a row of their color *or*
    try to capture 5 pairs of their opponents stones.  You capture
    a pair of stones any time you sandwich the stones between a
    pair of your stones.
 
Philosopher's Football

    On a 19x19 board, players take turns either adding men to the field, or
    moving the football.  The football moves by jumping lines of adjacent
    men (and removing them from the board).  The object is to move the
    football to (or past) your goal line.
    
Plotto (Copyright (c) 1995 David Smith)

    The players take turns placing one hex shaped piece in turn onto an open
    space (no board).  Pieces are numbered either 1, 2, 3 or 4 and you may
    play a piece of any number at each turn.  The object is to place a pair
    of pieces with your number in a straight line with two pieces in
    between.
 
Qubic

    On a 4x4x4 grid, two players alternate placing their pieces, attempting
    to form four in a row in any direction.
 
Score4

    On a 4x4 grid of pegs, two players alternate dropping their pieces from
    the top of a peg, down a column, attempting to form four in a row in any
    direction.

Survival (Copyright (c) 1995 David Smith))

    Survival is played on a hexagonal board made up of 19 numbered hexagons.
    Two players take turns placing pieces on the board with the "arrow" of
	the piece dictating the direction in which the next piece played by that
	player must be played.  The first player who can not move loses the game.
 
Tanbo & Tanbo3d (Copyright (c) 1995 Mark Steere)

    Played on a Go board, Tanbo crudely models a system of plant roots.  Roots
    which are growing, competing for space, and dying.  In beginner play, the
    roots grow much as the roots in a garden. Over time, the roots become
    shrewd and calculating.

    To win, a player must eliminate all eight of his opponent's roots.  One
    player will always win.  It's impossible to repeat a board configuration
    in Tanbo.  Therefore a game cannot result in a draw.  

    Tanbo3d extends the game of Tanbo into three dimensions.

Terrace (Copyright (c) 1995 by Siler/Siler Ventures.  All Rights Reserved)

    Terrace is played on an 8x8 board consisting of 16 'L' shaped terraces. 
    Two corners of the board are "High" and the other corners are "Low". 
    Each player has pieces of 4 sizes ('A', 'B', 'C' and 'D'). 'A' pieces are
    the smallest, 'D' pieces are the largest. 'T' pieces are the same size as
    'A' pieces and are each player's "key" piece.

    The object of the game is to capture your opponent's "T" or move your "T"
    to the lowest square on your opponent's side of the board.

Trax & StdTrax (Copyright (c) 1983 David Smith)

    Trax is a game played with square tiles.  Each tile is identical
    to all other tiles, one side has a white line connecting opposite
    edges and a black line connecting the other edges, and the other
    side has a white line connecting 2 adjacent edges and a black line
    connecting the other edges.

    The object of the game is to create a loop of your color while
    preventing your opponent from doing the same.  An alternate winning
    condition is to create a line extending from one edge of the board
    to the opposite edge of the board when the board is at least 8
    tiles wide (or tall).

    StdTrax limits the board to an 8x8 area.  Normal Trax allows to board
    to grow to whatever size is necessary.  Normal Trax is also known as
    SuperTrax.

TwixT (Copyright (c) Avalon Hill)

    On a 24x24 board, players take turns placing pegs of their color
    on the board.  Any time a peg is placed a "knight's move" from 
    another peg of the same color, a strut is placed, connecting them.
    A strut can not cross over (through) another strut.  The object is 
    to connect your sides of the board while preventing your opponent
    from doing the same.


--
  /\/\/\  | Richard Rognlie / Pr. Computer Analyst / PRC Inc. / McLean, VA
 /  \ \ \ | E-Mail: rrognlie@qnet.com       rrognlie@vtsu.prc.com
 \  / / / | Phone:  (Home) (703) 361-4764   (Office) (703) 556-2458
  \/\/\/  |                                 (Fax)    (703) 556-1174



From: jimc@mainelink.net
Subject: ICWS '94 for DOS or Windows
Date: 1995/09/06
Message-ID: <42j8is$bgk@news.mainelink.net>#1/1
newsgroups: rec.games.corewar

Is there an emulator (preferably with debug) for DOS or 
Windows that runs '94.  I can only find mars88 which isn't bad 
but doesn't support the new instructions.




From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: (no subject)
Date: 1995/09/06
Message-ID: <42ipd1$ntu@agate.berkeley.edu>#1/1
references: <42fv4g$cgi@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

In article <42fv4g$cgi@geraldo.cc.utexas.edu>,  <jwilkinson@utexas.edu> wrote:
>Also, is there any protection whatsoever against me faking mail from
>"stst@vuse.vanderbilt.edu" and putting ";kill Agony II" as the first
>line?

No, there isn't.  The reason there's no security on the server is that
we're all more interested in spending time writing corewar than writing
security for the Hill.  We trust each other.
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: FAQ frequency
Date: 1995/09/06
Message-ID: <199509061644.RAA16541@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

>>how about just a subject line with _no message_ such as:
>>
>>	SUBJECT: FAQ @ ftp.blah.blah.blah

>>	?
>>Wouldn't take anytime to look at it and virtually no bandwidth...

>>personally I would say that posting a FAQ list every day or so would be more 
>>annoying that the FAQ's themselves...

>Yeah right, as though people will look at that and
>immediately understand what that means?  To the
>contrary, it would likely increase the confusion
>of corewar newbies... which isn't anybody's goal...

We can post a very short message something like:

SUBJECT: HOW TO FIND FAQ

You can find the Corewar FAQ at:

FTP:	rtfm.mit.edu/pub/usenet/news.answers/games/corewar-faq

WWW:	http://www.stormikn.com/~koth/

And post it weekly, alternated with the FAQ document; should be
enought for most servers.


>-JKW

-Beppe




From: jwilkinson@utexas.edu
Subject: Re:  pMARS Feature
Date: 1995/09/06
Message-ID: <42lccs$6mn@geraldo.cc.utexas.edu>#1/1
references: <9509062016.AA06753@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

>>You can easily blank or set all P-space cells to the same value with this 
>>sequence of cdb commands:

>or define a macro:
>
>del=@psp~fill0,$~0~@psp off

Ok, I guess that sounds like the best solution for now.
Although scrambling would be ideal.

I'll try out the macro.  Thanks.

-JKW




From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Re: FAQ frequency
Date: 1995/09/06
Message-ID: <9509062025.AA06757@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

Concerning a weekly mini-FAQ, the stormking standings that are posted weekly
already contain a pointer to www.stormking.com/~koth. I'll ask Tuc to mention
the FAQ as well. BTW, Thomas, whatever happened to the weekly reports from
pizza?

-Stefan



From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Re:  pMARS Feature
Date: 1995/09/06
Message-ID: <9509062016.AA06753@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

jwilkinson@utexas.edu writes:
>Nescessity being the mother of invention, I've stumbled blindly upon
>a pMARS feature that I now need.  An inter-round Pspace scrambler.

You can easily blank or set all P-space cells to the same value with this 
sequence of cdb commands:

(cdb) pspace
(cdb) fill 0,$
fill with: 0
(cdb)

or define a macro:

del=@psp~fill0,$~0~@psp off

As for "scrambling", i.e. filling with arbitrary values, the only way to do 
this right now is by editing each P-space cell individually. If you can 
compile pMARS, there's a one-line patch to cdb.c that allows you to input
formulas instead of just values when prompted for P-cell contents. The
macro below requires the patch and sets each P-cell to a pseudo-random number:

scramble=@psp~!!~@ed .~x=(x+1101)%9019~@+1~if .<=$~!~@psp off

If you can't recompile pMARS, you'll have to wait for the next minor upgrade.

-Stefan



From: bezzi@iol.it (Beppe Bezzi)
Subject: Re (no subject)
Date: 1995/09/06
Message-ID: <199509061644.RAA16543@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar


>>>Also, is there any protection whatsoever against me faking mail from
>>>"stst@vuse.vanderbilt.edu" and putting ";kill Agony II" as the first
>>>line?
>>
>>No, there isn't.  The reason there's no security on the server is that
>>we're all more interested in spending time writing corewar than writing
>>security for the Hill.  We trust each other.
>>--
>>            Michael Constant (mconst@soda.csua.berkeley.edu)

>Why do you trust everyone?  You couldn't possibly know them all...
>You certainly don't know me.  I'd hate to see the hill get corrupted
>by a newbie who gets pissed and decides to remove everyone's warrior
>with a quick series of ";kill" messages at 2 in the morning...


Some times ago, for a mistake of the server, one of my warriors
submitted to the beginners hill was published with another's name. I
tried to kill it but, to do that, I had to ask the other to do it for
me. So a minimal security is there on the hill server.

A person able to fake a mail address, I don't know if it's easy or
difficult, could better send e-mail to my bank to get money (don't
worry I don't use e-mail with bank ;-) than kill my programs at Pizza
seen one don't get a penny for having a warrior on hill.
Apart that one able to crack a system is sure able to make a good
warrior without cheating.

>Ah well, I suppose most people don't buy security systems until -after-
>they're robbed too...

The problem of security is that it's expensive and time consuming; I
see no reason to spend time and money to better protect the hill.

>-JKW


-Beppe Bezzi
bezzi@iol.it

P.S. If you have to make any experiment with one of my warriors,
please kill a 'test' *not* Marcia Trionfale. TIA :-)





From: lewin@netcom.com (Karl Lewin)
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/06
Message-ID: <lewinDEI682.GAM@netcom.com>#1/1
references: <429s4o$87u@mikasa.iol.it> <42kc6b$cdf@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu wrote:

: STEP    equ     14	;guess Randy didn't published it
: EXTRA   equ     4
: DJNOFF  equ     5000	;still gues but I don't think it changes a lot

....<deleted>....

: Is that the same code as the Porch Swing currently residing on the hill?
: I've tested a warrior that I'm making against that code up there, and I
: consistently beat the Swing around 50/30/20 or so.  However, when I sent
: my warrior to Pizza, Swing beat me 24/57/19.

: Help?  Suggestions?

: -JKW

I think that the constants (at least the DJNOFF) are not quite what they 
are in the "production" version.  I doubt that it is optimum but try -500 
for DJNOFF and the code in question seems to perform much better.

Karl
-- 







From: jwilkinson@utexas.edu
Subject: Re: (no subject)
Date: 1995/09/06
Message-ID: <42iu9k$f96@geraldo.cc.utexas.edu>#1/1
references: <42fv4g$cgi@geraldo.cc.utexas.edu> <42ipd1$ntu@agate.berkeley.edu>
newsgroups: rec.games.corewar

>>Also, is there any protection whatsoever against me faking mail from
>>"stst@vuse.vanderbilt.edu" and putting ";kill Agony II" as the first
>>line?
>
>No, there isn't.  The reason there's no security on the server is that
>we're all more interested in spending time writing corewar than writing
>security for the Hill.  We trust each other.
>--
>            Michael Constant (mconst@soda.csua.berkeley.edu)

Why do you trust everyone?  You couldn't possibly know them all...
You certainly don't know me.  I'd hate to see the hill get corrupted
by a newbie who gets pissed and decides to remove everyone's warrior
with a quick series of ";kill" messages at 2 in the morning...

Ah well, I suppose most people don't buy security systems until -after-
they're robbed too...

-JKW




From: pk6811s@acad.drake.edu
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/06
Message-ID: <1995Sep6.111515@acad.drake.edu>#1/1
references: <199509051855.TAA09763@iol-mail.iol.it>
newsgroups: rec.games.corewar

(Beppe Bezzi) writes:
> 
> Isn't it always better to bomb the longer time possible, at least with
> incendiary bombs?

Actually the mov/spl bomb is a bit weak against Silk, especially the
ones like Timescape that execute their code once and move on.  I think
even Withershins would score higher against Timescape if it just skipped
the bomb/scan part and started the sweep right off.  Least some recent
testing seems to support that :-)

Paul



From: jwilkinson@utexas.edu
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/06
Message-ID: <42kc6b$cdf@geraldo.cc.utexas.edu>#1/1
references: <429s4o$87u@mikasa.iol.it>
newsgroups: rec.games.corewar

;redcode-94
;name Porch Swing
;kill Porch Swing
;Author Randy Graham
;assert 1
;strategy Swing with a little wider range.
;strategy Now 80% bomb/scan with djn-stream once-thru

STEP    equ     14	;guess Randy didn't published it
EXTRA   equ     4
DJNOFF  equ     5000	;still gues but I don't think it changes a lot

        dat.f  #gate-10,  step-gate+5
gate    dat.f  #gate-10,  sneer-STEP+1
dat2    dat.f  #gate-20,  step-gate+5
dat1    dat.f  #gate-25,  step-gate+5
site2   spl.a  #gate-30,  step-gate+5
site    spl.a  #gate-40,  step-gate+5
for EXTRA
        dat.f  0,         0
rof

adder   sub.f  sweeper,   sneer
hithigh mov.i  step,      *sneer
hitlow  mov.i  step,      @sneer
sneer   sne.i  @gate+STEP*6-1-EXTRA, *gate+STEP*3-EXTRA  ;so we bomb
step
looper  djn.b  adder,     <DJNOFF
setup   add.f  sneer,     gate
sweeper spl.a  #-STEP*4+1,<-STEP*4+1
mover   mov.i  @swinger,  >gate
swinger djn.b  mover,     {site
step    dat.f  <STEP,     <-STEP

end     hithigh

---

Is that the same code as the Porch Swing currently residing on the hill?
I've tested a warrior that I'm making against that code up there, and I
consistently beat the Swing around 50/30/20 or so.  However, when I sent
my warrior to Pizza, Swing beat me 24/57/19.

Hmmmm... I suppose it could be an anomoly(sp?).  However I would guess
that this is a less effective reproduction of Swing than the original...
Either that or I've got my pMARS set up wrong somehow...


Help?  Suggestions?

-JKW




From: jwilkinson@utexas.edu
Subject: SHORT FAQ:  READ ME FIRST.
Date: 1995/09/06
Message-ID: <42kbf4$cdf@geraldo.cc.utexas.edu>#1/1
references: <42j8is$bgk@news.mainelink.net>
newsgroups: rec.games.corewar

Check out this address for any and all Corewars stuff you need.

http://www.stormking.com/~koth/




From: ruhl@phoebe.cair.du.edu (Robert A. Uhl)
Subject: Re: Pizza Weirdness!
Date: 1995/09/07
Message-ID: <42m4qr$ai@hermes.cair.du.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu>
newsgroups: rec.games.corewar

In article <42lm66$81e@agate.berkeley.edu>,
Michael Constant <mconst@soda.CSUA.Berkeley.EDU> wrote:
>
>This was a decision made quite a while ago.  In fact, Pizza used to have
>random battle placements.  The change was made for a couple of reasons:
>so that one could meaningfully compare scores between slightly different
>versions of the same program without worrying about the effect of program
>placement, and so that people wouldn't be able to submit the same program
>multiple times in order to get better scores.

  Wouldn't be simpler to have the program be placed in a different
location each battle?  Well, not simpler, but it would be more valid
as far as results are concerned.  I might have a great program that is
placed to close to a scanner or bomber and therefore lose, even though
I should win.  Not that any of my programs are that good...

  And don't worry about the effects of battle placements; over several
games, they should even out.  Not that I know how many games are
necessary to do this.  Probably quite a few more than 100.

-- 
+------------------------------------------------------------------------+
| Bob Uhl | Spectre                  | `En touto nika' +                 |
| U of D  | Baron Robert von Raetzin | http://mercury.cair.du.edu/~ruhl/ |
+------------------------------------------------------------------------+



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's '94 Pizza Hill
Date: 1995/09/07
Message-ID: <42lhv3$sv3@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server ICWS '94 Draft Hill:
(accessed with ";redcode-94")

Hill Specs:
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: ICWS '94 Draft

Last challenge: Wed Sep  6 17:06:04 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  44/ 27/ 29                      Pink              P.Kline     160      15
 2  45/ 41/ 13                  Agony II        Stefan Strack     149     687
 3  35/ 27/ 38              juliet storm           M R Bremer     142     303
 4  35/ 27/ 38          Marcia Trionfale          Beppe Bezzi     142     202
 5  42/ 42/ 17             HeremScimitar      A.Ivner,P.Kline     141     571
 6  37/ 34/ 29                 bwtest 03          Beppe Bezzi     141       6
 7  41/ 42/ 17            Mr Speculative           Derek Ross     140      14
 8  34/ 27/ 39               Cobblestone       Steven Morrell     140       3
 9  43/ 46/ 11                       PGP       Steven Morrell     140      19
10  34/ 28/ 38               Blue Funk 3       Steven Morrell     139     752
11  38/ 37/ 25                 Torch t17              P.Kline     139     240
12  39/ 39/ 22                endpoint .           M R Bremer     138      60
13  43/ 47/ 10                  PeepHole         Randy Graham     138       4
14  40/ 43/ 18             Miss Careless           Derek Ross     137      48
15  40/ 43/ 18                 sdrawkcaB       Steven Morrell     137       2
16  38/ 40/ 22                 Firestorm          Beppe Bezzi     135       7
17  18/  1/ 81                  Die Hard              P.Kline     135       1
18  37/ 39/ 24         Circle of Friends         Randy Graham     134     109
19  36/ 39/ 25              Thermite 1.0        Robert Macrae     134     410
20  41/ 51/  9               Porch Swing         Randy Graham     131     157

21  22/ 56/ 21               Obvious 2.0           John Lewis      88       0
--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "help koth" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: pk6811s@acad.drake.edu
Subject: Re: Pizza Weirdness!
Date: 1995/09/07
Message-ID: <1995Sep7.090716@acad.drake.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu> <42ln57$ebf@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu writes:

[discussion of fixed placements]

> 
> Allright.  Well, I can only say that the idea of fixed placements may
> have come and gone.  With the advent of pSPACE programs, I would say
> that it is vital to have random placements, because initial placements
> of programs in just the first round can totally knock the results out
> of whack.
> 
> I've been testing this with Swing.  The standard deviation of results
> jumps dramtically higher with pspace warriors.
> 
> Perhaps what we need now are 4 groups of 25 battles, or 2 groups of 50,
> with the pspace being reset to zero, with the 0th spot returned to the
> "-1" value again.

The effect of fixed placements is to freeze the existing programs on the
Hill in place while the submitter is free to change his boot distance,
step size, internal spacing, etc.  But even when we had random spacing
between submissions the changes in win/loss nearly always averaged out
over the 20 opponents.

With pspace the deviation in results is enormous.  A small change causes
the pspacer to take a different path, leading it to abandon or commit to
different strategies affecting the outcome greatly.  But 25 battles is
hopelessly small, and easily defeated by a non-sophisticated program
that starts with one strategy and switches to another after 5-10 rounds.
We might think about increasing the number of battles to 500 or so.
If the pspacer has a good strategy it should discover it by then :-)

Question: even if the starting locations are fixed, is there a possibility
that the starting order might not be?  Or does the submission get to
go first in every battle?

Paul Kline
pk6811s@acad.drake.edu



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's Beginner Hill
Date: 1995/09/07
Message-ID: <42lhva$sv6@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server Beginner Hill:
(accessed with ";redcode-b")

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.
  instruction set: ICWS '94 Draft

Last challenge: Tue Sep  5 15:48:35 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  65/  8/ 27               hit'em 1.03     Ulrich Schlechte     222      67
 2  56/ 30/ 13                   Obvious        John K. Lewis     183      17
 3  49/ 17/ 34 2-Copy Paper-Stone-Imp co            Mneumenth     182      76
 4  54/ 30/ 16               Miss Timing           Derek Ross     177      90
 5  48/ 34/ 18                Maya v1.6n    Christoph C. Birk     161      86
 6  45/ 32/ 24              Giskard v0.5           Ken Mitton     158      96
 7  46/ 38/ 16                  Eft v1.2    Maurizio Vittuari     155       1
 8  43/ 39/ 18               SpYkD! v1.0               Gollum     146      50
 9  45/ 46/  9                     Rusty           John Lewis     145      70
10  44/ 44/ 12             Mythicon v1.1             G. Eadon     144      34
11  39/ 34/ 27               Tiny papers            Mneumenth     144      75
12  42/ 46/ 12                   Unknown            Anonymous     138      72
13  36/ 39/ 25                   Skimmer               Gollum     132      60
14  38/ 44/ 18              Pebbles v1.2    Christoph C. Birk     131      22
15  37/ 45/ 17                Stone v0.1    Christoph C. Birk     129      23
16  40/ 58/  2             i really suck         Kurt Ostfeld     123      12
17  35/ 47/ 18  Everyone Must Get Stoned       Nathan Summers     122      87
18  31/ 48/ 21            Pink Rain v1.1      Pliny the Elder     114      46
19  31/ 50/ 19        The Anhilator v1.1                 Mark     111      56
20  26/ 68/  6                      test    Christoph C. Birk      85       8

21  15/ 69/ 16              BigBomb v0.3    Christoph C. Birk      60       2
--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "help koth" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's Experimental '94 Hill
Date: 1995/09/07
Message-ID: <42lhv8$sv5@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server Experimental '94 Hill:
(accessed with ";redcode-94x")

Hill Specs:
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: ICWS '94 Draft

Last challenge: Sun Sep  3 07:39:19 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  46/ 31/ 23            Illusion-94/55         Randy Graham     161       6
 2  43/ 31/ 26                Ptest 01 X          Beppe Bezzi     155       2
 3  32/ 21/ 47                   Aleph 1              Jay Han     144      51
 4  40/ 37/ 23               Stimpy v2.0     Brant D. Thomsen     144      44
 5  32/ 24/ 43                  Lucky 13        Stefan Strack     141     195
 6  42/ 46/ 11          I Can't Drive 55      Randy W. Graham     139      13
 7  29/ 22/ 49                 Blue Funk       Steven Morrell     135      43
 8  29/ 23/ 48 Der Zweite Blitzkrieg - 9      Mike Nonemacher     135     151
 9  30/ 26/ 44             Variation G-1              Jay Han     134     153
10  38/ 42/ 20                   Squeeky         Randy Graham     133      10
11  37/ 40/ 23                  ivscan6b            J.Layland     133      53
12  36/ 40/ 25        Miss Understanding           Derek Ross     132       4
13  27/ 23/ 50              NotSoBigImps        James Layland     130      49
14  19/  7/ 74          TimeScapeX (0.1)       J. Pohjalainen     130      15
15  36/ 42/ 22             Miss Careless           Derek Ross     129       1
16  35/ 41/ 24                     Fscan              Jay Han     129      37
17  39/ 51/ 10 still testing... big vers      Mike Nonemacher     128      14
18  37/ 51/ 12                 Rave B4.1        Stefan Strack     124     150
19  23/ 23/ 54    Marcia Trionfale 0.2 x          Beppe Bezzi     124       5
20  37/ 51/ 11                      Test        Stefan Strack     124      25

21  32/ 48/ 20              Genocide 94x      Mike Nonemacher     116      19
--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "help koth" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: jwilkinson@utexas.edu
Subject: Re: Pizza Weirdness!
Date: 1995/09/07
Message-ID: <42ln57$ebf@geraldo.cc.utexas.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu>
newsgroups: rec.games.corewar

>>I must then ask: Why doesn't Pizza have random battle placements?  And,
>>what affect does this fact have on my challenges??
>
>This was a decision made quite a while ago.  In fact, Pizza used to have
>random battle placements.  The change was made for a couple of reasons:
>so that one could meaningfully compare scores between slightly different
>versions of the same program without worrying about the effect of program
>placement, and so that people wouldn't be able to submit the same program
>multiple times in order to get better scores.
>
Ok... I understand that idea.

>The fixed placements should not affect your challenges at all, except that
>(of course) you cannot submit the same warrior multiple times to see the
>effect of different placements.  If you want to see how the fixed placement
>system works, look at the pmars documentation for the -F option.  Of course,
>the number used for the -F option on the hill is kept secret.

Allright.  Well, I can only say that the idea of fixed placements may
have come and gone.  With the advent of pSPACE programs, I would say
that it is vital to have random placements, because initial placements
of programs in just the first round can totally knock the results out
of whack.

I've been testing this with Swing.  The standard deviation of results
jumps dramtically higher with pspace warriors.

Perhaps what we need now are 4 groups of 25 battles, or 2 groups of 50,
with the pspace being reset to zero, with the 0th spot returned to the
"-1" value again.







From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: Pizza Weirdness!
Date: 1995/09/07
Message-ID: <42lm66$81e@agate.berkeley.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu wrote:
>I must then ask: Why doesn't Pizza have random battle placements?  And,
>what affect does this fact have on my challenges??

This was a decision made quite a while ago.  In fact, Pizza used to have
random battle placements.  The change was made for a couple of reasons:
so that one could meaningfully compare scores between slightly different
versions of the same program without worrying about the effect of program
placement, and so that people wouldn't be able to submit the same program
multiple times in order to get better scores.

The fixed placements should not affect your challenges at all, except that
(of course) you cannot submit the same warrior multiple times to see the
effect of different placements.  If you want to see how the fixed placement
system works, look at the pmars documentation for the -F option.  Of course,
the number used for the -F option on the hill is kept secret.
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: jwilkinson@utexas.edu
Subject: Pizza Weirdness!
Date: 1995/09/07
Message-ID: <42lcr6$6mn@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Well, I suppose "Weirdness" might be too strong of a word..

Anyway, I mentioned before my curiosity at finding myself
beaten by Porch Swing, when I thought my program should
cream it.

In order to determine if my loss to Porch Swing was an exception,
I mailed my warrior to Pizza three consecutive times.  I was hoping
to recieve a better picture as to how badly Swing would beat me
in the long run.  As I compared the 3 new results to the earlier
result, I was startled to find them to be all exactly the same.
They had run about 6 hours apart, and had the same results.  In fact,
EVERY battle outcome was exactly the same in all four of my challenges.

This has led me to the inescapable conclusion that I am either the
luckiest man who has ever lived, or Pizza doesn't have random battle
placements.  My intuition leads me to believe in the latter of the
two. :)

I must then ask: Why doesn't Pizza have random battle placements?  And,
what affect does this fact have on my challenges??

-JKW






From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: (no subject)
Date: 1995/09/07
Message-ID: <42llbo$7rk@agate.berkeley.edu>#1/1
references: <42fv4g$cgi@geraldo.cc.utexas.edu> <42ipd1$ntu@agate.berkeley.edu> <42iu9k$f96@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

jwilkinson@utexas.edu wrote:
>Michael Constant (mconst@csua.berkeley.edu) wrote:
>>No, there isn't.  The reason there's no security on the server is that
>>we're all more interested in spending time writing corewar than writing
>>security for the Hill.  We trust each other.
>
>Why do you trust everyone?  You couldn't possibly know them all...
>You certainly don't know me.

We trust everyone because we hope that everyone will notice that this is
an environment of trust, and thus not betray that trust.  The fact is that
no one really wants to bother writing a secure hill system, and until we
have any problems with the system we're using now, I don't think that will
change.

Who would *want* to kill other people's warriors?  It's not as if anyone
will think they're really cool for hacking the hill system.  It's not as
if they'd cause irreparable damage, because hill stats are posted to the
newsgroup weekly and the hill admin could restore the hill from a copy of
the posted stats.  While it's lots of fun to conquer the hill, there's no
prize for winning... and if someone just wants to see a hill report with
their program in first place, they can just as easily forge mail from
pizza to themselves :-)
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: bezzi@iol.it (Beppe Bezzi)
Subject: ICWS tournement
Date: 1995/09/07
Message-ID: <199509070943.KAA20992@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

I read from the FAQs:

>..
>the ten best warriors submitted to the EBS tournament are entered
>into the annual ICWS tournament.  .....
>..

That means that the ten best warriors in the ;redcode-icws stormking
hill automatically partecipate to the annual ICWS championship?

TIA

-Beppe Bezzi
bezzi@iol.it





From: ruhl@phoebe.cair.du.edu (Robert A. Uhl)
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42q6vm$6jd@hermes.cair.du.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.vanderbilt.edu.noname> <42oa6q$5l5@charnel.ecst.csuchico.edu>
newsgroups: rec.games.corewar

In article <42oa6q$5l5@charnel.ecst.csuchico.edu>,
Dr. Manhattan <sd@ecst.csuchico.edu> wrote:
>
>Pizza's hills are practically on a dedicated machine.  They are
>currently running on "webhead", the school's dedicated web server.
>It almost never has a load above 0.6, and worst-case battles (100
>ties) only take a little more than a minute.  Upping the battles
>to 200 or even 250 would still get you results back in less than
>half an hour in most cases.

  I didn't realise that it was _that_ fast.  Of course, I normally run
Redcoder with all the bells and whistels going full blast, so it's
pretty slow.

>I'll do it if there's a demand.

  Demand!  Demand!

>On another note, should I get to work on the ";password" switch?
>jwilkinson makes an interesting point, it is easy to forget mail,
>and although killing other peoples warriors will probably never
>happen, an ounce of prevention is worth a pound of cure.

  Why not add verification?  If mail is sent requesting a ;kill, just
mail back with a unique passwd.  The user sends it in with a ;rkill or
something, and it kills the old warrior and deletes the passwd from
the passwd log (so that it may be used again).  Also, an ;nkill
<passwd> to say, 'No, I don't want to remove my warrior.'  Nice,
simple and IMHO elegant.  Also, passwords would not be predicatable,
which is a Good Thing.

-- 
+------------------------------------------------------------------------+
| Bob Uhl | Spectre                  | `En touto nika' +                 |
| U of D  | Baron Robert von Raetzin | http://mercury.cair.du.edu/~ruhl/ |
+------------------------------------------------------------------------+



From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <9509081553.AA07321@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

jwilkinson@mail.utexas.edu wrote:
>Can someone explain an affirmative reason for the -F option?

I think Michael and Paul did a very good job of explaining why -F.
Essentially, it allows you optimize your warrior without the "noise"
of random score fluctuations.

>I don't think I'm gonna be submitting the same warrior over and
>over, in the hope that I'll get a lucky break...  But, I would
>like to be able to re-submit to look for anomolies in the battle
>outcomes.

Explain what you mean by "anomalies in the battle outcomes".

-Stefan



From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/08
Message-ID: <199509082347.AAA02923@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

jwilkinson@mail.utexas.edu wrote
>bezzi@iol.it (Beppe Bezzi) wrote:
>>
>>>: STEP    equ     14	;guess Randy didn't published it
>>>: EXTRA   equ     4
>>>: DJNOFF  equ     5000	;still gues but I don't think it changes a lot
>>
>>>....<deleted>....
>>
>>I based my guess on the two missing equ on those facts:
>>
>I found that with a larger step, a different djn placement, I was
>able to delay the triggering of the spl stream until after the core
>had been quick scanned...  which would seem to be advantagous no matter
>what type of program Swing is fighting.

As I said I did'n spent lots of time in trying to guess the exact
constants Randy used. A larger step is more effective against a medium
size (near 10 lines) warrior but has less chances to catch a small 4
lines stone like Blue Funk. My posting was mainly based on how to
attack Porch and Whitershins after they begin clearing and, whichever
is the step, booting Tornado behind a large decoy make me sure, unless
PS happens to begin the battle beetween me and my decoy, to be
shielded from its scanning part. Against Marcia Trionfale and
Timescape a simpler thing like
.	...
begin	add.f	step,	comp
	seq	10,	20
	jmp	clear
	jmp	begin
.	....
is more effective than the whole Porch but, in my home test, not
against all opponents.

>>The djn stream has the double function: wound opponents catched and
>>...

>Yeah, I found that a properly placed djn was the only thing that gave
>P Swing much of a chance to beat one of my Cotton replicators.

I don't know your Cotton replicators, happy to give a look if you want
to send them to me, so I cannot say.
BTW I have too a replicator able to beat Porch (not Agony, not miss
Careless); I submitted it to the hill, entering but scoring *far* less
than Marcia in the total score so I kept it home; I also have a
replicator better, in home test, than Marcia, I hope I learnt
something in the last two months :-), but I keep it, should Marcia be
put off the hill.

Please Randy publish the equ and stop that. We have now to think
"Pink" ;-)

>-John

-Beppe




From: sd@ecst.csuchico.edu (Dr. Manhattan)
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42oa6q$5l5@charnel.ecst.csuchico.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.vanderbilt.edu.noname>
newsgroups: rec.games.corewar

Stefan Strack <stst@idnsun.gpct.Vanderbilt.Edu> wrote:
>pk6811s@acad.drake.edu wrote:
>
>>We might think about increasing the number of battles to 500 or so.
>>If the pspacer has a good strategy it should discover it by then :-)
>
>That's of course the best option. The downside is that a submission backlog
>might develop because of the increased time it takes to run the battles. 
>But then again we might start a collection among newsgroup readers to buy a
>workstation dedicated to corewar :-)
>
>>
>>Paul Kline
>
>-Stefan

Pizza's hills are practically on a dedicated machine.  They are
currently running on "webhead", the school's dedicated web server.
It almost never has a load above 0.6, and worst-case battles (100
ties) only take a little more than a minute.  Upping the battles
to 200 or even 250 would still get you results back in less than
half an hour in most cases.

I'll do it if there's a demand.

On another note, should I get to work on the ";password" switch?
jwilkinson makes an interesting point, it is easy to forget mail,
and although killing other peoples warriors will probably never
happen, an ounce of prevention is worth a pound of cure.

-Thos



From: y91magpa@ida.liu.se (Magnus Paulsson)
Subject: Wariors wanted
Date: 1995/09/08
Message-ID: <y91magpa.810559105@news>#1/1
newsgroups: rec.games.corewar

Hi
I'm thinking of taking up corewars again but I've throw out all my (and others)
wariors (from my dinky 40 M(m)b HD), could some one mail me (or post) a few
wariors so I have something to compare my wariors with.
Only send 94 wariors and maybee pspace ones.

/Magnus Paulsson



From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42qbvv$3sa@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu>
newsgroups: rec.games.corewar

>>But then again we might start a collection among newsgroup readers to buy a
>>workstation dedicated to corewar :-)

>I have a Pentium/90 at home which sits idle for a good portion of the day
>and night, and there's a chance that I might get a real net connection
>sometime in the near future (I'm using SLIP right now).  If this happens,
>I would be happy to run a KotH server on my machine.

>Now if the newsgroup wants to contribute to a net connection for me... :-)

>            Michael Constant (mconst@soda.csua.berkeley.edu)

Net connection...?  How much do those cost anyway, and why would you
want one?  Also, why would you need a genuine net connection in order
to be able to run battles?

It would seem as though you'd simply need to leave you're computer
dialed into your local server 24h/day, with a script waiting to run
the programs that were mailed over to you...




From: tusk@daimi.aau.dk (Martin M|ller Pedersen)
Subject: Re: PCROBOTS
Date: 1995/09/08
Message-ID: <42q92b$9vg@krone.daimi.aau.dk>#1/1
references: <41ceoe$pji@ra.ins.de> <41ln94$7ro@news.snowcrest.net> <41t69o$icm@agate.berkeley.edu> <42d1oe$ar5@news.snowcrest.net>
newsgroups: rec.games.corewar

Thus spake sriverhi@snowcrest.net (Ananda):

>In article <41t69o$icm@agate.berkeley.edu>, mconst@soda.CSUA.Berkeley.EDU 
>says...

>>Actually, discussion of PCRobots, C++Robots, RoboWar, etc. is welcome
>>here.  The rec.games.corewar charter says that this group is for the
>>discussion of "corewar and related games".  Just about any programming
>>game qualifies as "related" to corewar, and it's not as if we're over-
>>whelmed with traffic here anyway :-)

>Ahh... Ok.  It is really nice to read a postitive posting like this :)
>Does anyone have a filename for C++Robots?  I have never seen it.  Thanks.

>-Ananda

try http://www.daimi.aau.dk/~tusk/pbmserv




From: wsheppar@st6000.sct.edu (Wayne Sheppard)
Subject: Re: FAQ frequency
Date: 1995/09/08
Message-ID: <42pqn0$1561@st6000.sct.edu>#1/1
references: <4265qh$3pb@news.asu.edu> <42bark$qs5@allnews.infi.net> <tom.14.304C024E@halls1.cc.monash.edu.au> <42htp1$kes@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

How about this:

Once a week, post the Table of Contents of the FAQ with a pointer
to where the answers can be found.

Wayne

-- 
Wayne Sheppard
wsheppar@st6000.sct.edu



From: jwilkinson@mail.utexas.edu
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42pleq$fol@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack) wrote:
>>[discussion of fixed placements]
>
>To which I should add that the value of the -F option changes every month,
>so if you feel that you've been a victim of bad placement, submit again
>next month.
>
Ugh, am I missing something here?
Can someone explain an affirmative reason for the -F option?
I don't think I'm gonna be submitting the same warrior over and
over, in the hope that I'll get a lucky break...  But, I would
like to be able to re-submit to look for anomolies in the battle
outcomes.

>>We might think about increasing the number of battles to 500 or so.
>>If the pspacer has a good strategy it should discover it by then :-)
>
>That's of course the best option. The downside is that a submission backlog
>might develop because of the increased time it takes to run the battles. 
>But then again we might start a collection among newsgroup readers to buy a
>workstation dedicated to corewar :-)
>
I don't think there'd be much of a backlog.  So far, I have yet to submit
a warrior, and have it take more that an hour to get results back.  And,
sometimes I even submitted multiple warriors at once, and they came back
quickly...

-John




From: jwilkinson@mail.utexas.edu
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/08
Message-ID: <42pl5i$fol@geraldo.cc.utexas.edu>#1/1
references: <199509071404.PAA22396@iol-mail.iol.it>
newsgroups: rec.games.corewar

bezzi@iol.it (Beppe Bezzi) wrote:
>
>>: STEP    equ     14	;guess Randy didn't published it
>>: EXTRA   equ     4
>>: DJNOFF  equ     5000	;still gues but I don't think it changes a lot
>
>>....<deleted>....
>
>I based my guess on the two missing equ on those facts:
>
I found that with a larger step, a different djn placement, I was
able to delay the triggering of the spl stream until after the core
had been quick scanned...  which would seem to be advantagous no matter
what type of program Swing is fighting.

Just now, I submitted P Swing, with the new constants, to see if I'm
getting close to Randy's...

>The djn stream has the double function: wound opponents catched and
>delay other scanners. Having to test against him a pure bomber and a
>replicator, the position of the stream is, for me, irrilevant, for you
>may be different. 
>
Yeah, I found that a properly placed djn was the only thing that gave
P Swing much of a chance to beat one of my Cotton replicators.

-John




From: jwilkinson@mail.utexas.edu
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42pjo5$eip@geraldo.cc.utexas.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu> <42ln57$ebf@geraldo.cc.utexas.edu> <1995Sep7.090716@acad.drake.edu>
newsgroups: rec.games.corewar

>[discussion of fixed placements]

>The effect of fixed placements is to freeze the existing programs on the
>Hill in place while the submitter is free to change his boot distance,
>step size, internal spacing, etc.  But even when we had random spacing
>between submissions the changes in win/loss nearly always averaged out
>over the 20 opponents.
>
Indeed, however with random placements, at least I would be able to resubmit
my warrior a couple of times, and see which losses and wins were the
rule, and which were the exception.  That seems fair...

>With pspace the deviation in results is enormous.  A small change causes
>the pspacer to take a different path, leading it to abandon or commit to
>different strategies affecting the outcome greatly.  But 25 battles is
>hopelessly small, and easily defeated by a non-sophisticated program
>that starts with one strategy and switches to another after 5-10 rounds.
>We might think about increasing the number of battles to 500 or so.
>If the pspacer has a good strategy it should discover it by then :-)
>
True, it might find a good strategy, but that would be unfair to non-pspace
warriors.  I recommend that the KotH program runs a series of 5, 100 round
meta-battles.  The pspace would be reset 5 times, once before each meta-battle.

Averaging these scores would be a step towards not getting out-of-whack
scores.

As a personal test, I ran my warrior against Porch Swing 10 times in a row,
with 100 round battles.  9 out of the times I won with something like 50/20/30.
However, one of the ten times, I lost 15/16/69.  Obviously, this is a serious
anomoly, which really scews the battle curve...

-John




From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: Pizza Weirdness!
Date: 1995/09/08
Message-ID: <42oqcd$scv@agate.berkeley.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

Stefan Strack <stst@idnsun.gpct.Vanderbilt.Edu> wrote:
>But then again we might start a collection among newsgroup readers to buy a
>workstation dedicated to corewar :-)

I have a Pentium/90 at home which sits idle for a good portion of the day
and night, and there's a chance that I might get a real net connection
sometime in the near future (I'm using SLIP right now).  If this happens,
I would be happy to run a KotH server on my machine.

Now if the newsgroup wants to contribute to a net connection for me... :-)
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Re: ***deleted for security reasons***
Date: 1995/09/09
Message-ID: <1995Sep9.165044.2877@rhodes>#1/1
references: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

John Wilkinson (jwilkinson@mail.utexas.edu) wrote:
: >>> The easiest defense is to ALWAYS mod your strategy before using it.
: >Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
: >for your warrior to execute.
: Incrementing afterwards takes more time, just start off with "JMP @1, 0"
: and add to it.  Doesn't that make more sense?  Well, maybe not, obviously
: I don't know much since I haven't made a succesful pwarrior yet. :)
I think he meant, instead of jmp @0, 5, use jmp >0, 5.  This doesn't
erase the boot pointer, but makes an opponent use a little more
"thinking" to get you.

: -John

Randy



From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: New feature on SKI-ICWS
Date: 1995/09/09
Message-ID: <42t76q$ol7@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

Hello everyone,

	I decided to take my first stab at perl (Oh ok, stop complaining
that it took me this long) and decided to put up a new feature on
http://www.stormking.com/~koth .  It'll be in the section where you
can find out about the scores of the different hills.

		Warriors currently awaiting battle!

	This will be a list of the warriors that are currently in the
hopper to be put into battle.  PLEASE let me know if you have any
ideas, suggestions, etc.  The guest book is always ready!

	For mail users, I just need to find a good name to use 
(";waiting" ";backlog" ";listing" ) and that will be implemented.


		Thanks, Scott

--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: Michael N Nonemacher <schitzo+@CMU.EDU>
Subject: Re: Pizza Weirdness!
Date: 1995/09/09
Message-ID: <ckITTem00YUv0Gi_J9@andrew.cmu.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu>
newsgroups: rec.games.corewar

Excerpts from netnews.rec.games.corewar: 9-Sep-95 Re: Pizza Weirdness!
by Michael Constant@soda.CS 
>That's the problem.  My access provider (my school) won't let me leave my
>modem connected 24hrs/day unless I subscribe to their $70/month service.
>I am considering that service, and I am also considering getting ISDN from
>my school (but at $170/month + $30/month phone charges, I'm not sure if I
>can afford it).  If neither of those options is viable, I will probably
>get some sort of commercial service.

Excerpts from netnews.rec.games.corewar: 9-Sep-95 Re: Pizza Weirdness!
by John Wilkinson@mail.utex 
> I know that I could stay connected 24h/day to UT.  They only charge
> by the day.  It's 12 cents a day... and some charge per megabyte of
> transfer or something...  :/

Kinda makes me feel spoiled here at CMU, with direct 'net connections
all over the place (4 in my 3-person dorm room, in every classroom, on
the telephone poles near the soccer field...), just waiting for an
Ethernet card to plug into...  Of course, if you want to run KotH
scripts on my 386dx/25, just lemme know... :)

Mike
schitzo+@cmu.edu
root@cactus.res.cmu.edu



From: daredevl@enterprise.america.com (pagefault)
Subject: PSPACE question
Date: 1995/09/09
Message-ID: <42sm7q$7b7@enterprise.america.com>#1/1
newsgroups: rec.games.corewar

I have some questions about PSPACE.  What happens with this situation:

1. You lose, change to strat #2
2. You lose again w/Strat #2, how does your program know on the next 
round not to use strat #2?

I guess my baseline question is if PSPACE stays the same between 
rounds...? Other than the win/loss that is.
-- 
     p a g e f a u l t   
    /~-~-~-~-~-~-~-~-~~\
-==|unseen terror - UDIC|==-
    \__c c++ vb xbase__/
       ~~~~~~~~~~~~~~



From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Re: ***deleted for security reasons***
Date: 1995/09/09
Message-ID: <1995Sep9.122302.2875@rhodes>#1/1
references: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

Steven C. Morrell (morrell@math.utah.edu) wrote:
: My newsreader crashed on the last two messages that I posted, so I am
: trying out the Stormking list server.  Its feed is about 36 hours before
: my regular feed, so I'm getting tomorrow's news today!
Whoa, so tell me how I'm going to do with my next submission tomorrow
morning.  *-)

: snide remarks or comparing the success of Porch Swing/Withershins Twice-
: type programs (there must be a shorter name for these things!) with
I use Porch and Withers or PS and WT.  Or even Stone/scanners

: (The poor man's quote mechanism:)
: >> From: graham@hal.mathcs.rhodes.edu (Randy Graham)
: >> Date: 2 Sep 95 09:34:04 -0500
: >> The easiest defense is to ALWAYS mod your strategy before using it.
: Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
: for your warrior to execute.
VERY GOOD point.  I will use that.

: Actually, this makes vector-selecting pretty slow unless you can cram lots
: of warriors into 100 instructions.
Had an idea for doing that.  Should be able to get 3-4 good or 4-6
decent warriors in there and still switch.  Might get tricky on
launching, but I think with code reuse and a couple of setup
instruction for pointers and process buildup, you can really cram
quite a bit in there.  But, if the switcher isn't any good, it won't
really matter.

: If this post works correctly, I shall post a little on the theory and
: implementation of so-called PVamps, including the source to PGP, which I
: wrote mostly to show off.  Sorry to keep folks waiting.
I'm dying (grin) to see this.  Wondered what PGP was al about.

: Steven
: -- 
: Steven Morrell                       morrell@math.utah.edu

Randy
 (any news on the book?   ever recover from that crash?)



From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/09
Message-ID: <42shf8$gk1@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu> <42scr5$dsa@valhalla.stormking.com>
newsgroups: rec.games.corewar

>Folks........

>	All someone needs to do is come up with a new hill concept, and 
>pitch it.  If people show an interest, I'll put it on Stormking.Com, no
>problems.  You want a 500 round game, sure, not a problem.  Which hills?
>;redcode-94ml (long)?

No no, not a 500 round game.  I mean, you -could- do that.  But, I'd
rather see a set of 5 or 10, 100 round battles, with the scores averaged
normally.

That way traditional warriors wouldn't be forced out by smart pspacers.

-John




From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: Re: FAQ frequency
Date: 1995/09/09
Message-ID: <42qoip$o3o@valhalla.stormking.com>#1/1
references: <199509061644.RAA16541@iol-mail.iol.it>
newsgroups: rec.games.corewar

Beppe Bezzi (bezzi@iol.it) wrote:
: >>how about just a subject line with _no message_ such as:
: >>
: >>	SUBJECT: FAQ @ ftp.blah.blah.blah

: >>	?
: >>Wouldn't take anytime to look at it and virtually no bandwidth...

: >>personally I would say that posting a FAQ list every day or so would be more 
: >>annoying that the FAQ's themselves...

: >Yeah right, as though people will look at that and
: >immediately understand what that means?  To the
: >contrary, it would likely increase the confusion
: >of corewar newbies... which isn't anybody's goal...

: We can post a very short message something like:

: SUBJECT: HOW TO FIND FAQ

: You can find the Corewar FAQ at:

: FTP:	rtfm.mit.edu/pub/usenet/news.answers/games/corewar-faq

: WWW:	http://www.stormikn.com/~koth/

Incase anyone want, that address is really http://www.stormking.com/~koth

				Tuc

--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/09
Message-ID: <42sh9j$gk1@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu>
newsgroups: rec.games.corewar

>Well, the answers to those questions are seperate.  I want a real net
>connection so I can run various servers on my home computer, and so that I
>can access other internet computers at a reasonable speed.  A real net
>connection is by no means required for a KotH server; all that's required
>is a 24-hour connection.  Unfortunately, it is hard for me to get a 24-
>hour connection without actually getting a real net connection.  See below.

Interesting.  How would you manage to get faster access with an internet
connection?  Aren't you still limited by you modem speed?

>That's the problem.  My access provider (my school) won't let me leave my
>modem connected 24hrs/day unless I subscribe to their $70/month service.
>I am considering that service, and I am also considering getting ISDN from
>my school (but at $170/month + $30/month phone charges, I'm not sure if I
>can afford it).  If neither of those options is viable, I will probably
>get some sort of commercial service.

Hmmm... so what does ISDN give you?

I know that I could stay connected 24h/day to UT.  They only charge
by the day.  It's 12 cents a day... and some charge per megabyte of
transfer or something...  :/

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/09
Message-ID: <42sh08$gk1@geraldo.cc.utexas.edu>#1/1
references: <199509082347.AAA02923@iol-mail.iol.it> <42r7f8$mgv@geraldo.cc.utexas.edu> <1995Sep9.111748.2873@rhodes>
newsgroups: rec.games.corewar

>: Yeah, Randy!  Gimme constants! :>

>Sorry, not yet.  I will tell you that the DJNOFF is a negative, not
>too large.  In fact, it is incorrect in what is on the hill.  The
>intention was for her to bomb through until I got back to myself.
>Somehow, I submitted the wrong Porch Swing, and she stops when crossing
>the djn-stream pointer.

Damn, you're evil.  That's ok, I like evil. :->

Interesting, that's exactly what my Porch Swing does...  I believe
the DJNOFF I was using was 1985, but I was using a larger step.
Twenty-something...  I'd prolly be better with a smaller step, I suppose...
I'll work with it.

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: ***deleted for security reasons***
Date: 1995/09/09
Message-ID: <42sgq2$gk1@geraldo.cc.utexas.edu>#1/1
references: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

"Steven C. Morrell" <morrell@math.utah.edu> wrote:
>
>My newsreader crashed on the last two messages that I posted, so I am
>trying out the Stormking list server.  Its feed is about 36 hours before
>my regular feed, so I'm getting tomorrow's news today!
>
Feed?  You mean that you don't get posted messages immediately?

Why not?  Email is always immediate... :/

>>> The easiest defense is to ALWAYS mod your strategy before using it.
>
>Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
>for your warrior to execute.
>
Incrementing afterwards takes more time, just start off with "JMP @1, 0"
and add to it.  Doesn't that make more sense?  Well, maybe not, obviously
I don't know much since I haven't made a succesful pwarrior yet. :)

-John




From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/09
Message-ID: <1995Sep9.111748.2873@rhodes>#1/1
references: <199509082347.AAA02923@iol-mail.iol.it>
newsgroups: rec.games.corewar

John Wilkinson (jwilkinson@mail.utexas.edu) wrote:
: >Please Randy publish the equ and stop that. We have now to think
: >"Pink" ;-)
: >
: Yeah, Randy!  Gimme constants! :>
Sorry, not yet.  I will tell you that the DJNOFF is a negative, not
too large.  In fact, it is incorrect in what is on the hill.  The
intention was for her to bomb through until I got back to myself.
Somehow, I submitted the wrong Porch Swing, and she stops when crossing
the djn-stream pointer.

: -John




From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: Re: Pizza Weirdness!
Date: 1995/09/09
Message-ID: <42scr5$dsa@valhalla.stormking.com>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu>
newsgroups: rec.games.corewar

Michael Constant (mconst@soda.CSUA.Berkeley.EDU) wrote:
: John Wilkinson  <jwilkinson@mail.utexas.edu> wrote:
: >Michael Constant <mconst@csua.berkeley.edu> wrote:
: >>I have a Pentium/90 at home which sits idle for a good portion of the day
: >>and night, and there's a chance that I might get a real net connection
: >>sometime in the near future (I'm using SLIP right now).  If this happens,
: >>I would be happy to run a KotH server on my machine.
: >>

Folks........

	All someone needs to do is come up with a new hill concept, and 
pitch it.  If people show an interest, I'll put it on Stormking.Com, no
problems.  You want a 500 round game, sure, not a problem.  Which hills?
;redcode-94ml (long)?

			Tuc
--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: "Steven C. Morrell" <morrell@math.utah.edu>
Subject: ***deleted for security reasons***
Date: 1995/09/09
Message-ID: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu>#1/1
newsgroups: rec.games.corewar


My newsreader crashed on the last two messages that I posted, so I am
trying out the Stormking list server.  Its feed is about 36 hours before
my regular feed, so I'm getting tomorrow's news today!

If anyone was wondering what my last two messages said, they were mostly
snide remarks or comparing the success of Porch Swing/Withershins Twice-
type programs (there must be a shorter name for these things!) with
imp-stone combo's when they first came out.  But I did have the following
gem for P-space switching protection:


(The poor man's quote mechanism:)
>> From: graham@hal.mathcs.rhodes.edu (Randy Graham)
>> Date: 2 Sep 95 09:34:04 -0500

>> The easiest defense is to ALWAYS mod your strategy before using it.

Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
for your warrior to execute.

Actually, this makes vector-selecting pretty slow unless you can cram lots
of warriors into 100 instructions.


If this post works correctly, I shall post a little on the theory and
implementation of so-called PVamps, including the source to PGP, which I
wrote mostly to show off.  Sorry to keep folks waiting.

Steven
-- 
Steven Morrell                       morrell@math.utah.edu



From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: How to beat P. Swing and W. Thrice (and die in the try :-) -Long-
Date: 1995/09/09
Message-ID: <42r7f8$mgv@geraldo.cc.utexas.edu>#1/1
references: <199509082347.AAA02923@iol-mail.iol.it>
newsgroups: rec.games.corewar

>I don't know your Cotton replicators, happy to give a look if you want
>to send them to me, so I cannot say.
>BTW I have too a replicator able to beat Porch (not Agony, not miss
>Careless); I submitted it to the hill, entering but scoring *far* less
>than Marcia in the total score so I kept it home; I also have a
>replicator better, in home test, than Marcia, I hope I learnt
>something in the last two months :-), but I keep it, should Marcia be
>put off the hill.
>
I'm not sure how Cottons would do in the 94 hill, but on the multiwarrior
94 hill, I believe I've got four different cotton-based programs out of
the ten spots on the hill...

Oh also, I submitted my version of Porch Swing to the hill, with my
guess at the constants, and it scored 109... or maybe 116?  I dunno,
a little over one hundred, and it didn't make it to the hill.

>Please Randy publish the equ and stop that. We have now to think
>"Pink" ;-)
>
Yeah, Randy!  Gimme constants! :>

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/09
Message-ID: <42r76n$mgv@geraldo.cc.utexas.edu>#1/1
references: <9509081553.AA07321@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

>>Can someone explain an affirmative reason for the -F option?
>
>I think Michael and Paul did a very good job of explaining why -F.
>Essentially, it allows you optimize your warrior without the "noise"
>of random score fluctuations.
>
No, that doesn't make logical sense.  In order to get a more representative
picture of the ability of a warrior, you need to test it under varied
situations, not this static example of all possible contingencies.

>>I don't think I'm gonna be submitting the same warrior over and
>>over, in the hope that I'll get a lucky break...  But, I would
>>like to be able to re-submit to look for anomolies in the battle
>>outcomes.
>
>Explain what you mean by "anomalies in the battle outcomes".
>
Well, in any random data set, there will always be values which are
far from representative of the average, or median values.  The same
is the case in Corewars.  I've computed a generalized statement: About
1 in every ten battles I run is 20 or more wins off from the average.

This means that out of the 20 battles that I fight in my attempt to
get on the hill, two of those are totally unrepresentative of my average
score using random placements.  However, with the -F option, I have no
method for determining which battles are anomolies in the data set.

-John




From: max@alcyone.darkside.com (Erik Max Francis)
Subject: Re: High level language for corewar?
Date: 1995/09/10
Message-ID: <7i88aD8w200w@alcyone.darkside.com>#1/1
references: <42toe7$do@nuscc.nus.sg>
newsgroups: rec.games.corewar

lohwengk@iscs.nus.sg (Loh Weng Keong Calvin) writes:

> I mean, right now, corewar is like assembly language.
> But, on top of assembly language, you have things like
> Pascal, C, etc.
> Could the same thing be done for corewar?

It of course could (an ANSI C compiler which generates REDCODE -- 
weird), but one question is how useful that is.  Certainly it would 
facilitate writing larger, more complex, beasties, but another very 
important issue in Corewar is efficiency -- if it runs too slowly or 
wastes time doing unnecessary things, then it will probably get 
nailed.  Compilers can be very good at optimization (C compilers are 
well-known for this), but they can't be perfect, and that might tip 
the balance.

It certainly would be a very interesting project, at least.


Erik Max Francis, &tSftDotIotE // uuwest!alcyone!max, max@alcyone.darkside.com
San Jose, CA, USA // 37 20 07 N  121 53 38 W // GIGO, Omega, Psi // the 4th R!
H.3`S,3,P,3$S,#$Q,C`Q,3,P,3$S,#$Q,3`Q,3,P,C$Q,#(Q.#`-"C`- // 1love // folasade
_Omnia quia sunt, lumina sunt._ // mc2? oo?  Nah. // http://www.spies.com/max/



From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: High level language for corewar?
Date: 1995/09/10
Message-ID: <42vnua$s04@agate.berkeley.edu>#1/1
references: <42toe7$do@nuscc.nus.sg>
newsgroups: rec.games.corewar

Loh Weng Keong Calvin <lohwengk@iscs.nus.sg> wrote:
>How about a high level language for corewar?

It would kind of defeat the purpose.  The fact is that redcode is an
assembly-level language, and all the techniques used in corewar revolve
around that fact.  Try writing an imp-spiral in an HLL!

If you are interested in HLL programming games, you might want to try
C++Robots (a FAQ is posted here periodically).
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: pan0178@comune.bologna.it (MV)
Subject: Re: High level language for corewar?
Date: 1995/09/10
Message-ID: <199509102020.QAA24340@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

>How about a high level language for corewar?
>It would make it more accesible.
>
>I mean, right now, corewar is like assembly language.
>But, on top of assembly language, you have things like
>Pascal, C, etc.
>Could the same thing be done for corewar?
>
>
Yes, maybe a high level language would make CW more accessible but...
I think that hight level cw would be a game very different from RedCode-cw,
it would be simply another game!

If you enjoy yourself with high level programming, maybe you can write (of
course in your preferred language: C, Pascal, Basic(bleah!), etc.) a
compiler who translates your high level warrior in a low level one (written
in RedCode assembly language, of course ;-)  )

Seriously, if you don't like assembly languages, maybe you'll enjoy much
yourself trying C-robots or P-robots...
But I suggest you to try to learn RedCode (it isn't so difficult! ;-)  )

M.Vittuari              pan0178@iperbole.bologna.it




From: erik.hetzner@infoway.com (Erik Hetzner)
Subject: Re: Pizza Weirdness!
Date: 1995/09/10
Message-ID: <95091011492737507@infoway.com>#1/1
newsgroups: rec.games.corewar

  Mconst@soda.Csua.Berkeley,
  In a message on 9 September, you wrote to me :

MC> can access other internet computers at a reasonable speed.  A real net
MC> connection is by no means required for a KotH server; all that's required
MC> is a 24-hour connection.  Unfortunately, it is hard for me to get a 24-
MC> hour connection without actually getting a real net connection.  See belo

MC> Anyway, if I do get one of the above services, I will be happy to run a
MC> KotH server on my computer.

All that is really required for a KotH server, from what I can tell, is a
UUCP connection, which can be had for about 20 a month. Of course, this
will not allow people to do spiffo things like check how they're
doing by WWW, but all that is really needed is mail, and UUCP fits that
bill perfectly.
-erik hetzner  
---
 � ATP/Linux 1.42 � I want to rise with the masses, not from them.



From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: Net Connections and KotH Servers
Date: 1995/09/10
Message-ID: <42vq19$seq@agate.berkeley.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42sh9j$gk1@geraldo.cc.utexas.edu> <42u7v4$acc@agate.berkeley.edu> <42v7kj$bei@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John Wilkinson  <jwilkinson@mail.utexas.edu> wrote:
>mconst@soda.CSUA.Berkeley.EDU (Michael Constant) wrote:
>>Not if you're not using a modem :-)  When I say "real net connection", I
>>mean anything that's available 24hrs/day and faster than a regular modem.
>>
>So what, a leased line?  Like one of those direct ethernet things?
>
>But I thought those cost like $1000/month or something.

Well, a leased line is only a few hundred dollars a month, and if I could
find a friend to split it with me (see footnote for how this would be done)
I could probably afford it.  ISDN would also work, and is somewhat (but not
much) cheaper.

>>ObCorewar: Has anyone noticed how well the system of matrix combat and
>>IC programs in the Shadowrun role-playing game can be modeled by corewar?
>
>What is Shadowrun?  I read your comparison to Corewars, but it seemed to
>assume prior knowledge of some kind...

Shadowrun is a role-playing game made by FASA Corporation.  I was referring
specifically to the part of the rulebook which describes the "matrix" (the
modern computer network in the time period of the game) and how characters
interact in the matrix.  Basically the idea is that if a character wants to
enter the matrix, they connect themselves to a cyberdeck, which they
connect to an access node for the matrix.  The cool part (and the part
that's related to corewar) is that the computer code in their cyberdeck is
accessible to any other person or program in the matrix.  Unfortunately, I
can't really describe the entire system of the matrix here.  If you are
interested, get a copy of the Shadowrun rulebook (2nd edition).  It costs
about $20 and is well worth it -- it's a great game :-)

There's got to be *some* way to get this thread back on topic... but I
must admit I don't see it ;-)  Perhaps there's another newsgroup that might
be better suited for this discussion?
____________
(*) To split a net connection with a friend, I would use my current modem
and data line for a 24-hour connection to my friend's computer, and I would
run any server programs on my computer (for the extra bandwidth).

--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Net Connections and KotH Servers
Date: 1995/09/10
Message-ID: <42v7kj$bei@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu> <42sh9j$gk1@geraldo.cc.utexas.edu> <42u7v4$acc@agate.berkeley.edu>
newsgroups: rec.games.corewar

mconst@soda.CSUA.Berkeley.EDU (Michael Constant) wrote:
>>Interesting.  How would you manage to get faster access with an internet
>>connection?  Aren't you still limited by you modem speed?
>
>Not if you're not using a modem :-)  When I say "real net connection", I
>mean anything that's available 24hrs/day and faster than a regular modem.
>
So what, a leased line?  Like one of those direct ethernet things?

But I thought those cost like $1000/month or something.

>ObCorewar: Has anyone noticed how well the system of matrix combat and
>IC programs in the Shadowrun role-playing game can be modeled by corewar?

What is Shadowrun?  I read your comparison to Corewars, but it seemed to
assume prior knowledge of some kind...

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/10
Message-ID: <42v7dv$bei@geraldo.cc.utexas.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu> <42v4eu$c5c@enterprise.america.com>
newsgroups: rec.games.corewar

daredevl@enterprise.america.com (pagefault) wrote:
>Michael Constant (mconst@soda.CSUA.Berkeley.EDU) wrote:
>: Stefan Strack <stst@idnsun.gpct.Vanderbilt.Edu> wrote:
>You could do it with SLIP, if you wanted to make a subject-line based 
>entry system, so you don't lose your mailbox.  The only overhead over the 
>line itself is the mail transfer which is practically nil.  I don't know 
>how much your P5/90 is compared to their workstation, but in many cases 
>it doesn't even compare (i.e. your P5/90 is slower in many cases)

I've run Corewars on a P90 before.  It took about 45 seconds to run 100
battles in pMARS 0.8.  The old MERCURY program was slightly faster...
maybe 30 seconds or so.

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/10
Message-ID: <42v7ad$bei@geraldo.cc.utexas.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu> <42m4qr$ai@hermes.cair.du.edu> <42u95b$agu@agate.berkeley.edu>
newsgroups: rec.games.corewar

mconst@soda.CSUA.Berkeley.EDU (Michael Constant) wrote:
>>  Wouldn't be simpler to have the program be placed in a different
>>location each battle?  Well, not simpler, but it would be more valid
>>as far as results are concerned.  I might have a great program that is
>>placed to close to a scanner or bomber and therefore lose, even though
>>I should win.  Not that any of my programs are that good...
>
>I think you misunderstood what I said.  The 100 rounds are, in fact, run
>with 100 different starting locations.  It's just that the *sequence* of
>starting locations is fixed, so that if you submit the same program more
>than once it will get exactly the same score.

True, however, with pspace warriors, the very first battle is much more
important than the remaining 99.  Having a fixed spot for battle number 1
goes a long way towards simply fixing the remaining 99.

That is why Pizza should run 10, 100 round battles, and average those
10 scores together...

-John




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: High level language for corewar?
Date: 1995/09/10
Message-ID: <42v74v$bei@geraldo.cc.utexas.edu>#1/1
references: <42toe7$do@nuscc.nus.sg>
newsgroups: rec.games.corewar

lohwengk@iscs.nus.sg (Loh Weng Keong Calvin) wrote:
>How about a high level language for corewar?
>It would make it more accesible.
>
>I mean, right now, corewar is like assembly language.
>But, on top of assembly language, you have things like
>Pascal, C, etc.
>Could the same thing be done for corewar?

Ick!  I hope not... that'd kill the idea.

I mean if you wrote these things in C, then the real test of programming
skill would be to create the most effecient compiler possible...

-John




From: daredevl@enterprise.america.com (pagefault)
Subject: Re: Pizza Weirdness!
Date: 1995/09/10
Message-ID: <42v4eu$c5c@enterprise.america.com>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42oqcd$scv@agate.berkeley.edu>
newsgroups: rec.games.corewar

Michael Constant (mconst@soda.CSUA.Berkeley.EDU) wrote:
: Stefan Strack <stst@idnsun.gpct.Vanderbilt.Edu> wrote:
: >But then again we might start a collection among newsgroup readers to buy a
: >workstation dedicated to corewar :-)

: I have a Pentium/90 at home which sits idle for a good portion of the day
: and night, and there's a chance that I might get a real net connection
: sometime in the near future (I'm using SLIP right now).  If this happens,
: I would be happy to run a KotH server on my machine.

You could do it with SLIP, if you wanted to make a subject-line based 
entry system, so you don't lose your mailbox.  The only overhead over the 
line itself is the mail transfer which is practically nil.  I don't know 
how much your P5/90 is compared to their workstation, but in many cases 
it doesn't even compare (i.e. your P5/90 is slower in many cases)
-- 
     p a g e f a u l t   
    /~-~-~-~-~-~-~-~-~~\
-==|unseen terror - UDIC|==-
    \__c c++ vb xbase__/
       ~~~~~~~~~~~~~~



From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: High level language for corewar?
Date: 1995/09/10
Message-ID: <42toe7$do@nuscc.nus.sg>#1/1
newsgroups: rec.games.corewar

How about a high level language for corewar?
It would make it more accesible.

I mean, right now, corewar is like assembly language.
But, on top of assembly language, you have things like
Pascal, C, etc.
Could the same thing be done for corewar?



From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: Pizza Weirdness!
Date: 1995/09/10
Message-ID: <42u95b$agu@agate.berkeley.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu> <42m4qr$ai@hermes.cair.du.edu>
newsgroups: rec.games.corewar

Robert A. Uhl <ruhl@phoebe.cair.du.edu> wrote:
>Michael Constant <mconst@soda.CSUA.Berkeley.EDU> wrote:
>>This was a decision made quite a while ago.  In fact, Pizza used to have
>>random battle placements.  The change was made for a couple of reasons [...]
>
>  Wouldn't be simpler to have the program be placed in a different
>location each battle?  Well, not simpler, but it would be more valid
>as far as results are concerned.  I might have a great program that is
>placed to close to a scanner or bomber and therefore lose, even though
>I should win.  Not that any of my programs are that good...

I think you misunderstood what I said.  The 100 rounds are, in fact, run
with 100 different starting locations.  It's just that the *sequence* of
starting locations is fixed, so that if you submit the same program more
than once it will get exactly the same score.

If this is still unclear, please email me and I will explain further.
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: mconst@soda.CSUA.Berkeley.EDU (Michael Constant)
Subject: Re: Net Connections and KotH Servers
Date: 1995/09/10
Message-ID: <42u7v4$acc@agate.berkeley.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.Vanderbilt.Edu.noname> <42qbvv$3sa@geraldo.cc.utexas.edu> <42ries$9bh@agate.berkeley.edu> <42sh9j$gk1@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John Wilkinson  <jwilkinson@mail.utexas.edu> wrote:
>Michael Constant <mconst@csua.berkeley.edu> wrote:
>>Well, the answers to those questions are seperate.  I want a real net
>>connection so I can run various servers on my home computer, and so that I
>>can access other internet computers at a reasonable speed.  A real net
>>connection is by no means required for a KotH server; all that's required
>>is a 24-hour connection.  Unfortunately, it is hard for me to get a 24-
>>hour connection without actually getting a real net connection.  See below.
>
>Interesting.  How would you manage to get faster access with an internet
>connection?  Aren't you still limited by you modem speed?

Not if you're not using a modem :-)  When I say "real net connection", I
mean anything that's available 24hrs/day and faster than a regular modem.

>Hmmm... so what does ISDN give you?

ISDN gives you speed.  A standard ISDN line, if used for data only, will
run at 128 Kbps.  This is much better than the max of 28.8 Kbps offered by
modems (and much *much* better than the 14.4 modem I have now :-)).

>I know that I could stay connected 24h/day to UT.  They only charge
>by the day.  It's 12 cents a day... and some charge per megabyte of
>transfer or something...  :/

Ah, that's the problem.  I can't stay connected 24hrs/day, unless I pay
for the service.  And since the service is expensive, and I really want
more than a 24-hour modem connection anyway, I am considering things like
ISDN.

ObCorewar: Has anyone noticed how well the system of matrix combat and
IC programs in the Shadowrun role-playing game can be modeled by corewar?
Imagine the matrix as one big (distributed) MARS system, with each
cyberdeck and each program as a corewar program.  Consider what happens
when a decker is attacked by IC.  The IC program attempts to overwrite
the code in the cyberdeck to crash it (like DAT instructions) or to cause
it to feedback and damage itself (like SPL instructions).  The decker
tries to disable the IC before it disables his cyberdeck.  The idea alone
of modifying another program's code is very corewar-esque.  So is the way
programs and deckers scan for each other.

Oh well, I could continue, but, now that I think of it, this is probably
even more off-topic than the rest of my post :-)  This is what I get for
posting at 1:30am.  Sigh.
--
            Michael Constant (mconst@soda.csua.berkeley.edu)



From: vanet@ix.netcom.com (Robert Dennis )
Subject: WIN CASH - New Web Site
Date: 1995/09/10
Message-ID: <42vnp8$1ti@ixnews4.ix.netcom.com>#1/1
newsgroups: rec.games.computer.puzzle,rec.games.computer.xpilot,rec.games.corewar,rec.games.design,rec.games.diplomacy,rec.games.empire,rec.games.frp.advocacy,rec.games.frp.cyber,rec.games.frp.dnd,rec.games.frp.gurps,rec.games.frp.live-action,rec.games.frp.marketplace,rec.games.frp.misc,rec.games.go,rec.games.int-fiction,rec.games.mecha,rec.games.miniatures,rec.games.miniatures.historical,rec.games.miniatures.misc,rec.games.miniatures.warhammer

Check out:

    http://www.vanet.com

and some of their links, you can win cash in their monthly
GIVEAWAYS just by visiting the site, no gimmicks, U.S.A. only...



From: nmoffitt@cs.usfca.edu (Nicholas M. Moffitt)
Subject: Re: PCROBOTS
Date: 1995/09/10
Message-ID: <42ts6b$gjj@noc.usfca.edu>#1/1
references: <41ceoe$pji@ra.ins.de> <41ln94$7ro@news.snowcrest.net> <41t69o$icm@agate.berkeley.edu> <42d1oe$ar5@news.snowcrest.net> <42q92b$9vg@krone.daimi.aau.dk>
newsgroups: rec.games.corewar

	I'm wondering if there's any other sort of C-robots for the PC.  I'm sick
of using the old 1985 non-cannonical C for my robots.  I don't want the cumbersome
nature of the PBeM server, and I REALLY don't want to use Pascal.  Is there any
other sort of C-robots out there for the PC?  Perhaps one that uses a normal ANSI C
compiler and just gives you header files to work with?  email me please with your
response.

----
The tao is in all programming languages, but try and avoid using COBOL.
	Nick Moffitt  nmoffitt@usfca.edu




From: John Wilkinson <jwilkinson@mail.utexas.edu>
Subject: Re: ***deleted for security reasons***
Date: 1995/09/10
Message-ID: <42te5s$6qh@geraldo.cc.utexas.edu>#1/1
references: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu> <42sgq2$gk1@geraldo.cc.utexas.edu> <1995Sep9.165044.2877@rhodes>
newsgroups: rec.games.corewar

>: >>> The easiest defense is to ALWAYS mod your strategy before using it.
>: >Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
>: >for your warrior to execute.
>: Incrementing afterwards takes more time, just start off with "JMP @1, 0"
>: and add to it.  Doesn't that make more sense?  Well, maybe not, obviously
>: I don't know much since I haven't made a succesful pwarrior yet. :)
>I think he meant, instead of jmp @0, 5, use jmp >0, 5.  This doesn't
>erase the boot pointer, but makes an opponent use a little more
>"thinking" to get you.

Hmmmm...  I don't understand the advantage of using "jmp >0, 5".
I don't know of any program that examines random core ptrs, and then
bombs the pointer that the last pointer pointed to.  That's the
only way I could see it making a difference... :/

Ah well, maybe I misunderstood completely.

-John




From: bremermr@elsun269.cc.purdue.edu (Myer R. Bremer)
Subject: Re: Pizza Weirdness!
Date: 1995/09/11
Message-ID: <431vkt$pha@mozo.cc.purdue.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.vanderbilt.edu.noname> <42oa6q$5l5@charnel.ecst.csuchico.edu> <42q6vm$6jd@hermes.cair.du.edu>
newsgroups: rec.games.corewar

Greetings.

I don't want to mail in 2 files when attempting to kill
my warriors.  I'd rather just type something like:
;redcode-94
;name juliet storm
;author M R Bremer
;kill juliet storm
;killpass xxxxxx
;submitpass yyyyyy

And if an author doesn't submit a password, then he/she 
can operate like before without this password business.

M R Bremer




From: pan0178@comune.bologna.it (MV)
Subject: Re: High level language for corewar?
Date: 1995/09/11
Message-ID: <199509112319.TAA20512@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

>Loh Weng Keong Calvin (lohwengk@iscs.nus.sg) wrote:
>: How about a high level language for corewar?
>: It would make it more accesible.
>
>: I mean, right now, corewar is like assembly language.
>: But, on top of assembly language, you have things like
>: Pascal, C, etc.
>: Could the same thing be done for corewar?
>
>why not try using a couple of processes, a fork() or two and a few kill()'s
>and playing a real core war.
>
>It sends the load through the roof mind you. You also need to make sure
>you don't kill your window manager etc.
>
>I have a bit of code here, but it is very buggy.
>
>
>Maybe I've misunderstood but...
remember that the relative speed of two process running concurrently on a
multiprocess system, is not deterministic!
In other words you can't be sure that the two warrior execute alternatively!

MV
>




From: gibo@morinda.it.ntu.edu.au (Nathan Gibson)
Subject: Re: High level language for corewar?
Date: 1995/09/11
Message-ID: <4312br$d7n@pellew.ntu.edu.au>#1/1
references: <42toe7$do@nuscc.nus.sg>
newsgroups: rec.games.corewar

Loh Weng Keong Calvin (lohwengk@iscs.nus.sg) wrote:
: How about a high level language for corewar?
: It would make it more accesible.

: I mean, right now, corewar is like assembly language.
: But, on top of assembly language, you have things like
: Pascal, C, etc.
: Could the same thing be done for corewar?

why not try using a couple of processes, a fork() or two and a few kill()'s
and playing a real core war.

It sends the load through the roof mind you. You also need to make sure
you don't kill your window manager etc.

I have a bit of code here, but it is very buggy.





From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: PSPACE question
Date: 1995/09/11
Message-ID: <199509111404.PAA00421@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

daredevl@enterprise.america.com (pagefault) wrote


>I have some questions about PSPACE.  What happens with this situation:

>1. You lose, change to strat #2
>2. You lose again w/Strat #2, how does your program know on the next 
>round not to use strat #2?

>I guess my baseline question is if PSPACE stays the same between 
>rounds...? Other than the win/loss that is.

If you save, in some location of the p-space, the strategy used, using
STP, you'll find it unchanged in the next round. 

Unless someone manage to "Brain Wash" you :-), using *your* processes
to modify it.

-- 
>     p a g e f a u l t   


--------
Beppe Bezzi
bezzi@iol.it

Gaudeamus igitur..




From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: KotH!
Date: 1995/09/11
Message-ID: <199509111404.PAA00424@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

John <jwilkinson@mail.utexas.edu> wrote:

>When a warrior is pushed off of the hill, do all of the other warriors
>that are still on the hill have the score that they made against that
>warrior subtracted away?

>It would seem logical that the answer is yes... but just thought I'd
>ask.

I don't know for sure but I made some calculations with hill results;
your score is the average of results against all warriors on hill
including yourself and the warrior pushed out (the 21�)

Obviously the next challenge the 21� will be another, so the score
against the previous one will be discharged.
Even a challenge of a warrior not making on the hill changes the
scores of your warriors.

>-John

-Beppe




From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: Re: Pizza Weirdness!
Date: 1995/09/11
Message-ID: <431b3u$h51@valhalla.stormking.com>#1/1
references: <95091011492737507@infoway.com>
newsgroups: rec.games.corewar

Erik Hetzner (erik.hetzner@infoway.com) wrote:
:   Mconst@soda.Csua.Berkeley,
:   In a message on 9 September, you wrote to me :

: All that is really required for a KotH server, from what I can tell, is a
: UUCP connection, which can be had for about 20 a month. Of course, this
: will not allow people to do spiffo things like check how they're
: doing by WWW, but all that is really needed is mail, and UUCP fits that
: bill perfectly.

	I tend to disagree with that statement.  STORMKING.COM was UUCP
up until a few months ago.  Up until that time people were constantly
posting "I sent my warrior to STORMKING 3 minutes ago, I didn't get a
reply, what gives?!" even though Stefan explained... A FEW TIMES in the 
FAQ that we were UUCP.  This is one of the driving forces that pushed STORMKING
to get a full time link to the Internet.  

			Scott
--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: pk6811s@acad.drake.edu
Subject: Re: Pizza Weirdness!
Date: 1995/09/11
Message-ID: <1995Sep11.154406@acad.drake.edu>#1/1
references: <9509071634.AA07064@idnsun.gpct.vanderbilt.edu.noname> <42oa6q$5l5@charnel.ecst.csuchico.edu> <42q6vm$6jd@hermes.cair.du.edu>
newsgroups: rec.games.corewar

(Robert A. Uhl) writes:
>...
>   Why not add verification?  If mail is sent requesting a ;kill, just
> mail back with a unique passwd.  The user sends it in with a ;rkill or
> something, and it kills the old warrior and deletes the passwd from

However, the usual purpose of ;kill is to remove the earlier version 
of the same program so we don't end up with multiple copies on the Hill.
So when you are submitting tests each one replaces the older one.  It
would be unimaginably hard to do this if the Hill had to wait for
a manual ;ok-to-kill response before finalizing the results of a submission.

Paul Kline
pk6811s@acad.drake.edu



From: John <jwilkinson@mail.utexas.edu>
Subject: Request
Date: 1995/09/11
Message-ID: <43229i$bqn@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

For those people who are participating in the 94 KotH hills, if you are
not going to post the strategy for your warrior, could you at least
say whether or not it uses pspace?  That wouldn't give away what you're
doing, but it'd give a better picture of pspace warriors' success rate...

-John




From: stst@vuse.vanderbilt.edu (Stefan Strack)
Subject: Core War Frequently Asked Questions (rec.games.corewar FAQ)
Date: 1995/09/11
Message-ID: <games/corewar-faq_810825449@rtfm.mit.edu>
newsgroups: rec.games.corewar,rec.answers,news.answers

Archive-name: games/corewar-faq
Last-Modified: 95/08/06
Version: 3.4

   These are the Frequently Asked Questions (and answers) from the Usenet
   newsgroup rec.games.corewar. A plain text version of this document is
   posted every two weeks. The hypertext version is available as
   <URL:http://www.stormking.com/~koth/corewar-faq.html>
     _________________________________________________________________
   
Table of Contents

    1. What is Core War
    2. Is it Core War or Core Wars?
    3. Where can I find more information about Core War?
    4. Core War has changed since Dewdney's articles. Where do I get a
       copy of the current instruction set?
    5. What is ICWS'94? Which simulators support ICWS'94?
    6. What is the ICWS?
    7. What is TCWN?
    8. How do I join?
    9. What is the EBS?
   10. Where are the Core War archives?
   11. Where can I find a Core War system for ...?
   12. I do not have FTP. How do I get all this great stuff?
   13. I do not have access to Usenet. How do I post and receive news?
   14. Are there any Core War related WWW sites?
   15. When is the next tournament?
   16. What is KotH? How do I enter?
   17. Is it DAT 0, 0 or DAT #0, #0? How do I compare to core?
   18. How does SLT (Skip if Less Than) work?
   19. What is the difference between in-register and in-memory
       evaluation?
   20. What does (expression or term of your choice) mean?
   21. Other questions?
       
   
     _________________________________________________________________
   
What is Core War?

   Core War is a game played by two or more programs (and vicariously by
   their authors) written in an assembly language called Redcode and run
   in a virtual computer called MARS (for Memory Array Redcode
   Simulator). The object of the game is to cause all processes of the
   opposing program to terminate, leaving your program in sole posession
   of the machine.
   
   There are Core War systems available for most computer platforms.
   Redcode has been standardized by the ICWS, and is therefore
   transportable between all standard Core War systems.
   
   [ToC]
     _________________________________________________________________
   
Is it "Core War" or "Core Wars"?

   Both terms are used. Early references were to Core War. Later
   references seem to use Core Wars. I prefer "Core War" to refer to the
   game in general, "core wars" to refer to more than one specific
   battle.
   
   [ToC]
     _________________________________________________________________
   
Where can I find more information about Core War?

   Core War was first described in the Core War Guidelines of March, 1984
   by D. G. Jones and A. K. Dewdney of the Department of Computer Science
   at The University of Western Ontario (Canada). Dewdney wrote several
   "Computer Recreations" articles in Scientific American which discussed
   Core War, starting with the May 1984 article. Those articles are
   contained in two anthologies:
   
   Author: Dewdney, A. K.
   Title: The Armchair Universe: An Exploration of Computer Worlds
   Published: New York: W. H. Freeman (c) 1988
   ISBN: 0-7167-1939-8
   Library of Congress Call Number: QA76.6 .D517 1988
   
   
   Author: Dewdney, A. K.
   Title: The Magic Machine: A Handbook of Computer Sorcery
   Published: New York: W. H. Freeman (c) 1990
   ISBN: 0-7167-2125-2 (Hardcover), 0-7167-2144-9 (Paperback)
   Library of Congress Call Number: QA76.6 .D5173 1990
   
   
   A.K. Dewdney's articles are still the most readable introduction to
   Core War, even though the Redcode dialect described in there is no
   longer current.
   
   [ToC]
     _________________________________________________________________
   
Core War has changed since Dewdney's articles. Where do I get a copy of the
current instruction set?

   A draft of the official standard (ICWS'88) is available as
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/standards/re
   dcode-icws-88> . This document is formatted awkwardly and contains
   ambiguous statements. For a more approachable intro to Redcode, take a
   look at Mark Durham's tutorial,
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/tutorial.1>
   and
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/tutorial.2>.
   
   
   Steven Morrell (morrell@math.utah.edu) is preparing a more practically
   oriented Redcode tutorial that discusses different warrior classes
   with lots of example code. Mail him for a preliminary version. Michael
   Constant (mconst@csua.berkeley.edu) is reportedly working on a
   beginner's introduction.
   
   [ToC]
     _________________________________________________________________
   
What is ICWS'94? Which simulators support ICWS'94?

   There is an ongoing discussion about future enhancements to the
   Redcode language. A proposed new standard, dubbed ICWS'94, is
   currently being evaluated. A major change is the addition of
   "instruction modifiers" that allow instructions to modify A-field,
   B-field or both. Also new is a post-increment indirect addressing mode
   and unrestricted opcode and addressing mode combination ("no illegal
   instructions"). ICWS'94 is backwards compatible; i.e. ICWS'88 warriors
   will run correctly on an ICWS'94 system. Take a look at the ICWS'94
   draft
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/documents/icws94.0202.Z
   > for more information. You can try out the new standard by
   submitting warriors to the '94 hills of the KotH servers. Two corewar
   systems currently support ICWS'94, pMARS (many platforms) and Redcoder
   (Mac), both available at ftp.csua.berkeley.edu.
   
   [ToC]
     _________________________________________________________________
   
What is the ICWS?

   About one year after Core War first appeared in Sci-Am, the
   "International Core War Society" (ICWS) was established. Since that
   time, the ICWS has been responsible for the creation and maintenance
   of Core War standards and the running of Core War tournaments. There
   have been six annual tournaments and two standards (ICWS'86 and
   ICWS'88).
   
   [ToC]
     _________________________________________________________________
   
What is TCWN?

   Since March of 1987, "The Core War Newsletter" (TCWN) has been the
   official newsletter of the ICWS. It is published quarterly and recent
   issues are also available as Encapsulated PostScript files.
   
   [ToC]
     _________________________________________________________________
   
How do I join?

   For more information about joining the ICWS (which includes a
   subscription to TCWN), or to contribute an article, review, cartoon,
   letter, joke, rumor, etc. to TCWN, please contact:
   

   Jon Newman
   13824 NE 87th Street
   Redmond, WA 98052-1959
   email: jonn@microsoft.com  (Note: Microsoft has NO affiliation with
                                     Core War.  Jon Newman just happens
                                     to work there, and we want to keep
                                     it that way!)

   Current annual dues are $15.00 in US currency.
   
   [ToC]
     _________________________________________________________________
   
What is the EBS?

   The Electronic Branch Section (EBS) of the ICWS is a group of Core War
   enthusiasts with access to electronic mail. There are no fees
   associated with being a member of the EBS, and members do reap some of
   the benefits of full ICWS membership without the expense. For
   instance, the ten best warriors submitted to the EBS tournament are
   entered into the annual ICWS tournament. All EBS business is conducted
   in the rec.games.corewar newsgroup.
   
   The current goal of the EBS is to be at the forefront of Core War by
   writing and implementing new standards and test suites.
   
   [ToC]
     _________________________________________________________________
   
Where are the Core War archives?

   Many documents such as the guidelines and the ICWS standards along
   with previous tournament Redcode entries and complete Core War systems
   are available via anonymous ftp from
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/systems>
   (128.32.149.19) in the /pub/corewar directories. Also, most of past
   rec.games.corewar postings (including Redcode source listings) are
   archived there. Jon Blow (blojo@csua.berkeley.edu) is the archive
   administrator. When uploading to /pub/corewar/incoming, ask Jon to
   move your upload to the appropriate directory and announce it on the
   net.
   
   Much of what is available on soda is also available on the German
   archive at iraun1.ira.uka.de (129.13.10.90) in the /pub/x11/corewars
   directory.
   
   The plain text version of this FAQ is automatically archived by
   news.answers.
   
   [ToC]
     _________________________________________________________________
   
Where can I find a Core War system for . . . ? 

   Core War systems are available via anonymous ftp from
   ftp.csua.berkeley.edu in the /pub/corewar/systems directory.
   Currently, there are UNIX, IBM PC-compatible, Macintosh, and Amiga
   Core War systems available there. It is a good idea to check
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/incoming> for program
   updates first.
   
   CAUTION! There are many, many Core War systems available which are NOT
   ICWS'88 (or even ICWS'86) compatible available at various archive
   sites other than ftp.csua.berkeley.edu. Generally, the older the
   program - the less likely it will be ICWS compatible.
   
   Reviews of Core War systems would be greatly appreciated in the
   newsgroup and in the newsletter.
   
   Below is a not necessarily complete or up-to-date list of what's
   available at ftp.csua.berkeley.edu:
   

    MADgic41.lzh    - corewar for the Amiga, v4.1
    MAD4041.lzh     - older version?
    MAD50B.lha      - corewar for the Amiga, beta version 5.0
    Redcoder-21.hqx - corewar for the Mac, supports ICWS'88 and '94 (without
                      extensions)
    core-11.hqx     - corewar for the Mac
    core-wars-simulator.hqx - same as core-11.hqx?
    corewar_unix_x11.tar.Z - corewar for UNIX/X-windows, ICWS'86 but
                      not ICWS'88 compatible
    koth31.tar.Z    - corewar for UNIX/X-windows. This program ran the
                      former KotH server at intel.com
    koth.shar.Z     - older version
    kothpc.zip      - port of older version of KotH to the PC
    deluxe20c.tar.Z - corewar for UNIX (broken X-windows or curses) and PC
    mars.tar.Z      - corewar for UNIX, likely not ICWS'88 compatible
    icons.zip       - corewar icons for MS-Windows
    macrored.zip    - a redcode macro-preprocessor (PC)
    c88v49.zip      - PC corewar, textmode display
    mars88.zip      - PC corewar, graphics mode display
    corwp302.zip    - PC corewar, textmode display, slowish
    mercury2.zip    - PC corewar written in assembly, fast!
    mtourn11.zip    - tournament scheduler for mercury (req. 4DOS)
    pmars08s.zip    - portable system, ICWS'88 and '94, runs on UNIX,
                      PC, Mac, Amiga. C source archive
    pmars08s.tar.Z  - same as above
    pmars08.zip    - PC executables with graphics display, req 386+
    macpmars02.sit.hqx - pMARS executable for Mac (port of version 0.2)
                      buggy, no display
    MacpMARS.10.cpt.hqx - port of v0.6 for the Mac, with display and
                      debugger
    MacpMARS.10s.cpt.hqx - C source (MPW, ThinkC) for Mac frontend
    ApMARS03.lha    - pMARS executable for Amiga (port of version 0.3.1)
    wincor11.zip    - MS-Windows system, shareware ($15)

   
   
   [ToC]
     _________________________________________________________________
   
I do not have FTP. How do I get all this great stuff? 

   There is an ftp email server at ftpmail@decwrl.dec.com. Send email
   with a subject and body text of "help" (without the quotes) for more
   information on its usage. If you don't have access to the net at all,
   send me a 3.5 '' diskette in a self-addressed disk mailer with postage
   and I will mail it back with an image of the Core War archives in PC
   format. My address is at the end of this post.
   
   [ToC]
     _________________________________________________________________
   
I do not have access to Usenet. How do I post and receive news?

   To receive rec.games.corewar articles by email, join the COREWAR-L
   list run on the Stormking.Com list processor. To join, send the
   message

        SUB COREWAR-L FirstName LastName

   to listproc@stormking.com. You can send mail to
   corewar-l@stormking.com to post even if you are not a member of the
   list. Responsible for the listserver is Scott J. Ellentuch
   (tuc@stormking.com).
   
   Another server that allows you to post (but not receive) articles is
   available. Email your post to rec-games-corewar@cs.utexas.edu and it
   will be automatically posted for you.
   
   [ToC]
     _________________________________________________________________
   
Are there any Core War related WWW sites?

   You bet. Each of the two KotH sites sport a world-wide web server.
   Stormking's Core War page is <URL:http://www.stormking.com/~koth>;
   pizza's is <URL:http://www.ecst.csuchico.edu/~pizza/koth> . A third
   WWW site is in Koeln, Germany:
   <URL:http://www.khm.uni-koeln.de/~wbi/Index/Smoking/CWBH>. Last but
   not least, Stephen Beitzel's "Unofficial Core War Page" is
   <URL:http://www2.eworld.com/aos/people/beitzel/core.html>.
   All site are in varying stages of construction, so it would be futile
   to list here what they have to offer.
   
   [ToC]
     _________________________________________________________________
   
When is the next tournament?

   The ICWS holds an annual tournament. Traditionally, the deadline for
   entering is the 15th of December. The EBS usually holds a preliminary
   tournament around the 15th of November and sends the top finishers on
   to the ICWS tournament.
   
   Informal double-elimination and other types of tournaments are held
   frequently among readers of the newsgroup; watch there for
   announcements or contact me.
   
   [ToC]
     _________________________________________________________________
   
What is KotH? How do I enter?

   King Of The Hill (KotH) is an ongoing Core War tournament available to
   anyone with email. You enter by submitting via email a Redcode program
   (warrior) with special comment lines. You will receive a reply
   indicating how well your program did against the current top programs
   "on the hill".
   
   There are two styles of KotH tournaments, "classical" and
   "multi-warrior". The "classical" KotH is a one-on-one tournament, that
   is your warrior will play 100 battles against each of the 20 other
   programs currently on the Hill. You receive 3 points for each win and
   1 point for each tie. (The existing programs do not replay each other,
   but their previous battles are recalled.) All scores are updated to
   reflect your battles and all 21 programs are ranked from high to low.
   If you are number 21 you are pushed off the Hill, if you are higher
   than 21 someone else is pushed off.
   
   In "multi-warrior" KotH, all warriors on the hill fight each other at
   the same time. Score calculation is a bit more complex than for the
   one-on-one tournament. Briefly, points are awarded based on how many
   warriors survive until the end of a round. A warrior that survives by
   itself gets more points than a warrior that survives together with
   other warriors. Points are calculated from the formula (W*W-1)/S,
   where W is the total number of warriors and S the number of surviving
   warriors. The pMARS documentation has more information on
   multi-warrior scoring.
   
   The idea for an email-based Core War server came from David Lee. The
   original KotH was developed and run by William Shubert at Intel
   starting in 1991, and discontinued after almost three years of
   service. Currently, KotHs based on Bill's UNIX scripts but offering a
   wider variety of hills are are running at two sites:
   "koth@stormking.com" is maintained by Scott J. Ellentuch
   (tuc@stormking.com) and "pizza@ecst.csuchico.edu" by Thomas H. Davies
   (sd@ecst.csuchico.edu). Up until May '95, the two sites provided
   overlapping services, i.e. the some of the hill types were offered by
   both "pizza" and "stormking". To conserve resources, the different
   hill types are now divided up among the sites. The way you submit
   warriors to both KotHs is pretty much the same. Therefore, the entry
   rules described below apply to both "pizza" and "stormking" unless
   otherwise noted.
   
   Entry rules for King of the Hill Corewar:
   
   1) Write a corewar program. KotH is fully ICWS '88 compatible, EXCEPT
   that a comma (",") is required between two arguments.
   
   2) Put a line starting with ";redcode" (or ";redcode-94, etc., see
   below) at the top of your program. This MUST be the first line.
   Anything before it will be lost. If you wish to receive mail on every
   new entrant, use ";redcode verbose". Otherwise you will only receive
   mail if a challenger makes it onto the hill. Use ";redcode quiet" if
   you wish to receive mail only when you get shoved off the hill. (Also,
   see 5 below).
   
   Additionally, adding ";name " and ";author " will be helpful in the
   performance reports. Do NOT have a line beginning with ";address" in
   your code; this will confuse the mail daemon and you won't get mail
   back.
   
   In addition, it would be nice if you have lines beginning with
   ";strategy" that describe the algorithm you use.
   
   There are currently seven separate hills you can select by starting
   your program with ;redcode-b, ;redcode-94, ;redcode-94x, ;redcode,
   ;redcode-icws, ;redcode-94m or ;redcode-94xm. The former three run at
   "pizza", the latter four at "stormking". More information on these
   hills is listed below.
   
   3) Mail this file to koth@stormking.com or pizza@ecst.csuchico.edu.
   "Pizza" requires a subject of "koth" (use the -s flag on most
   mailers).
   
   4) Within a few minutes you should get mail back telling you whether
   your program assembled correctly or not. If it did assemble correctly,
   sit back and wait; if not, make the change required and re-submit.
   
   5) In an hour or so you should get more mail telling you how your
   program performed against the current top 20 (or 10) programs. If no
   news arrives during that time, don't worry; entries are put in a queue
   and run through the tournament one at a time. A backlog may develop.
   Be patient.
   
   If your program makes it onto the hill, you will get mail every time a
   new program makes it onto the hill. If this is too much mail, you can
   use ";redcode[-??] quiet" when you first mail in your program; then
   you will only get mail when you make it on the top 20 list or when you
   are knocked off. Using ";redcode[-??] verbose" will give you even more
   mail; here you get mail every time a new challenger arrives, even if
   they don't make it onto the top 20 list.
   
   Often programmers want to try out slight variations in their programs.
   If you already have a program named "foo V1.0" on the hill, adding the
   line ";kill foo" to a new program will automatically bump foo 1.0 off
   the hill. Just ";kill" will remove all of your programs when you
   submit the new one. The server kills programs by assigning an
   impossibly low score; it may therefore take another successful
   challenge before a killed program is actually removed from the hill.
   

SAMPLE ENTRY:
;redcode
;name Dwarf
;author A. K. Dewdney
;strategy Throw DAT bombs around memory, hitting every 4th memory cell.
;strategy This program was presented in the first Corewar article.
bomb  DAT   #0
dwarf ADD   #4,    bomb
      MOV   bomb, @bomb
      JMP   dwarf
      END   dwarf          ; Programs start at the first line unless
                           ; an "END start" pseudo-op appears to indicate
                           ; the first logical instruction.  Also, nothing
                           ; after the END instruction will be assembled.

Here are the Specs for the various hills:

ICWS'88 Standard Hill Specs: (Accessed with ";redcode", available at
"stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: ICWS '88

ICWS Annual Tournament Hill Specs: (Accessed with ";redcode-icws",
available at "stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8192 instructions
   max. processes: 8000 per program
         duration: After 100,000 cycles, a tie is declared.
max. entry length: 300
 minimum distance: 300
  instruction set: ICWS '88

ICWS'94 Draft Hill Specs: (Accessed with ";redcode-94", available at
"pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft

ICWS'94 Beginner's Hill Specs: (Accessed with ";redcode-b", available
at "pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft
         max. age: after 100 successful challenges, warriors are retired.

ICWS'94 Experimental (Big) Hill Specs: (Accessed with ";redcode-94x",
available at "pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: extended ICWS '94 Draft

ICWS'94 Draft Multi-Warrior Hill Specs: (Accessed with ";redcode-94m",
available at "stormking")
         hillsize: 10 warriors
           rounds: 200
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft

ICWS'94 Experimental (Big) Multi-Warrior Hill Specs:
(Accessed with ";redcode-94xm", available at "stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: extended ICWS '94 Draft

   If you just want to get a status report without actually challenging
   the hills, send email with ";status" as the message body (and don't
   forget "Subject: koth" for "pizza"). If you send mail to "pizza" with
   "Subject: koth help" you will receive instructions that may be more up
   to date than those contained in this document.
   
   At stormking, a message body with ";help" will return brief
   instructions. If you submit code containing a ";test" line, your
   warrior will be assembled but not actually pitted against the warriors
   on the hill.
   
   All hills run portable MARS (pMARS) version 0.8, a
   platform-independent corewar system available at
   ftp.csua.berkeley.edu.
   
   The '94 and '94x hills allow three experimental opcodes and addressing
   modes currently not covered in the ICWS'94 draft document:

        SEQ - Skip if EQual (synonym for CMP)
        SNE - Skip if Not Equal
        NOP - (No OPeration)

        *   - indirect using A-field as pointer
        {   - predecrement indirect using A-field
        }   - postincrement indirect using A-field

   
   
   [ToC]
     _________________________________________________________________
   
Is it DAT 0, 0 or DAT #0, #0? How do I compare to core? 

   Core is initialized to DAT 0, 0. This is an "illegal" instruction
   under ICWS'88 rules and strictly compliant assemblers (such as KotH or
   pmars -8) will not let you write a DAT 0, 0 instruction - only DAT #0,
   #0. So this begs the question, how to compare something to see if it
   is empty core. The answer is, most likely the instruction before your
   first instruction and the instruction after your last instruction are
   both DAT 0, 0. You can use them, or any other likely unmodified
   instructions, for comparison. Note that under ICWS'94, DAT 0, 0 is a
   legal instruction.
   
   [ToC]
     _________________________________________________________________
   
How does SLT (Skip if Less Than) work?

   SLT gives some people trouble because of the way modular arithmetic
   works. It is important to note that all negative numbers are converted
   to positive numbers before a battles begins. Example: (-1) becomes (M
   - 1) where M is the memory size.
   
   Once you realize that all numbers are treated as positive, it is clear
   what is meant by "less than". It should also be clear that no number
   is less than zero.
   
   [ToC]
     _________________________________________________________________
   
What is the difference between in-register and in-memory evaluation?

   These terms refer to the way instruction operands are evaluated. The
   '88 Redcode standard ICWS'88 is unclear about whether a simulator
   should "buffer" the result of A-operand evaluation before the
   B-operand is evaluated. Simulators that do buffer are said to use
   in-register evaluation, those that don't, in-memory evaluation.
   ICWS'94 clears this confusion by mandating in-register evaluation.
   Instructions that execute differently under these two forms of
   evaluation are MOV, ADD, SUB, MUL, DIV and MOD where the effective
   address of the A-operand is modified by evaluation of the B-operand.
   This is best illustrated by an example:

   L1  mov L2,<L2
   L2  dat #0,#1

   Under in-register evaluation, the 'L2' instruction is saved in a
   buffer before the 'L2' memory location is decremented by evaluation of
   the B-operand of 'L1'. The saved 'dat #0,#1' instruction is then
   written to 'L2', leaving it unchanged.
   
   Under in-memory evaluation, the 'L2' instruction is not buffered and
   thus decremented by evaluation of the B-operand. After execution of
   'L1', 'L2' changes to "dat #0,#0".
   
   [ToC]
     _________________________________________________________________
   
What does (expression or term of your choice) mean?

   Here is a selected glossary of terms. If you have a definition and/or
   term you wish to see here, please send it to me.
   
   (References to an X-like program mean that the term X is derived from
   the specific program X and has become a generic term).
   
   Binary launch
          one of several means to start an imp-spiral running. The
          fastest launch technique, but requires the most code. See also
          JMP/ADD Launch and Vector Launch.
          


    impsize   equ 2667
    example   spl 4               ; extend by adding spl 8, spl 16, etc.
              spl 2
              jmp imp+(0*impsize) ; jmp's execute in order
              jmp imp+(1*impsize)
              spl 2
              jmp imp+(2*impsize)
              jmp imp+(3*impsize)
    imp       mov 0,impsize       ; in '94 use -> mov.i #0,impsize

   Bootstrapping
          Strategy of copying the active portion of the program away from
          the initial location, leaving a decoy behind and making the
          relocated program as small as possible.
          
   B-Scanners
          Scanners which only recognize non-zero B-fields.
          

    example   add #10,scan
    scan      jmz example,10

   C
          Measure of speed, equal to one location per cycle. Speed of
          light.
          
   CMP-Scanner
          A Scanner which uses a CMP instruction to look for opponents.
          

    example   add step,scan
    scan      cmp 10,30
              jmp attack
              jmp example
    step      dat #20,#20

   Color
          Property of bombs making them visible to scanners, causing them
          to attack useless locations, thus slowing them down.
          

    example   dat #100

   Core-Clear
          code that sequentially overwrites core with DAT instructions;
          usually the last part of a program.
          
   Decoys
          Bogus or unused instructions meant to slow down Scanners.
          Typically, DATs with non-zero B-fields.
          
   DJN-Stream (also DJN-Train)
          Using a DJN command to rapidly decrement core locations.
          

    example   ...
              ...
              djn example,<4000

   Dwarf
          the prototypical small bomber.
          
   Gate-busting (also gate-crashing)
          technique to "interweave" a decrement-resistant imp-spiral
          (e.g. MOV 0, 2668) with a standard one to overrun imp-gates.
          
   Hybrids
          warriors that combine two or more of the basic strategies,
          either in sequence (e.g. stone->paper) or in parallel (e.g.
          imp/stone).
          
   Imp
          Program which only uses the MOV instruction.
          

    example   MOV 0, 1
  or
    example   MOV 0, 2
              MOV 0, 2

   Imp-Gate
          A location in core which is bombed or decremented continuously
          so that an Imp can not pass. Also used to describe the
          program-code which maintains the gate.
          

    example   ...
              ...
              SPL 0, <example
              DAT <example, #0

   Imp-Ring
          A minimal Imp-Spiral.
          

    d         EQU (coresize+1)/3
    A         MOV 0,d   ; copy self to B
    B         MOV 0,d   ; copy self to C
    C         MOV 0,d   ; copy self to A+1

   Imp-Spiral
          An Imp-like program with two or more processes supporting each
          other. A three-point spiral, with six processes running in this
          sequence:
          

    d         EQU (coresize+1)/3
    A         MOV 0,d   ; copy self to B
    B         MOV 0,d   ; copy self to C
    C         MOV 0,d   ; copy self to A+1
    A+1       MOV 0,d   ; copy self to B+1
    B+1       MOV 0,d   ; copy self to C+1
    C+1       MOV 0,d   ; copy self to A+2

   Incendiary Bomb
          A type of Stun bomb which creates a SPL 0 carpet.
          

    example   SPL 0, 8
              MOV -1, <-1

   JMP/ADD Launch
          one of several means to start an imp-spiral running. The
          slowest launch technique, but requires the least code. See also
          Binary Launch and Vector Launch.
          

    impsize   equ 2667
    example   spl 1               ; extend by adding more spl 1's
              spl 1
              spl 2
              jmp @0,imp
              add #impsize,-1     ; bump address by impsize after each jmp
              dat #0,#0           ; excess processes must die!
    imp       mov 0,impsize       ; in '94 use -> mov.i #0,impsize

   Mirror
          see reflection.
          
   On-axis/off-axis
          On-axis scanners compare two locations M/2 apart, where M is
          the memory size. Off-axis scanners use some other separation.
          
   Optimal Constants
          (also optima-type constants) Bomb or scan increments chosen to
          cover core most effectively, i.e. leaving gaps of uniform size.
          Programs to calculate optimal constants and lists of optimal
          numbers are available at ftp.csua.berkeley.edu.
          
   Paper
          A Paper-like program. One which replicates itself many times.
          Part of the Scissors (beats) Paper (beats) Stone (beats
          Scissors) analogy.
          
   Pit-Trapper
          (also Slaver, Vampire). A program which enslaves another.
          Usually accomplished by bombing with JMPs to a SPL 0 pit with
          an optional core-clear routine.
          
   Quick Scan
          2c scan of a set group of core locations with bombing if
          anything is found. Both of the following codes snips scan 16
          locations and check for a find. If anything is found, it is
          attacked, otherwise 16 more locations are scanned. Example:
          

     start
     s1 for 8    ;'88 scan
            cmp  start+100*s1, start+100*s1+4000  ;check two locations
            mov  #start+100*s1-found, found  ;they differ so set pointer
     rof
            jmn  attack,     found     ;if we have something, get it
     s2 for 8
            cmp  start+100*(s2+6), start+100*(s2+6)+4000
            mov  #start+100*(s2+6)-found, found
     rof
     found  jmz  moveme,  #0        ;skip attack if qscan found nothing
     attack cmp  @found,  start-1   ;does found points to empty space?
            add  #4000,   found     ;no, so point to correct location
            mov  start-1, @found    ;move a bomb
     moveme jmp  0,       0

   In ICWS'94, the quick scan code is more compact because of the SNE
          opcode:
          


     start       ;'94 scan
     s1 for 4
            sne  start+400*s1, start+400*s1+100  ;check two locations
            seq  start+400*s1+200, start+400*s1+300  ;check two locations
            mov  #start+400*s1-found, found  ;they differ so set pointer
     rof
            jmn  which,   found     ;if we have something, get it
     s2 for 4
            sne  start+400*(s2+4), start+400*(s2+4)+100
            seq  start+400*(s2+4)+200, start+400*(s2+4)+300
            mov  #start+400*(s2+4)-found-100, found
     rof
     found  jmz  moveme,  #0        ;skip attack if qscan found nothing
            add  #100,    -1        ;increment pointer till we get the
     which  jmn  -1,      @found    ;right place
            mov  start-1, @found    ;move a bomb
     moveme jmp  0,       0

   Reflection
          Copy of a program or program part, positioned to make the
          active program invisible to a CMP-scanner.
          
   Replicator
          Generic for Paper. A program which makes many copies of itself,
          each copy also making copies.
          
   Self-Splitting
          Strategy of amplifying the number of processes executing a
          piece of code.
          

    example    SPL 0
    loop       ADD #10, example
               MOV example, @example
               JMP loop

   Scanner
          A program which searches through core for an opponent rather
          than bombing blindly.
          
   Scissors
          A program designed to beat replicators, usually a (B-field
          scanning) vampire. Part of the Paper-Scissors-Stone analogy.
          
   Self-Repair
          Ability of a program to fix it's own code after attack.
          
   Silk
          A replicator which splits off a process to each new copy before
          actually copying the code. This allows it to replicate
          extremely quickly. This technique is only possible under the
          '94 draft, because it requires post-increment indirect
          addressing. Example:
          

                spl     1
                mov.i   -1,     0
                spl     1               ;generate 6 consecutive processes
    silk        spl     3620,   #0      ;split to new copy
                mov.i   >-1,    }-1     ;copy self to new location
                mov.i   bomb,   >2000   ;linear bombing
                mov.i   bomb,   }2042   ;A-indirect bombing for anti-vamp
                jmp     silk,   {silk   ;reset source pointer, make new copy
    bomb        dat.f   >2667,  >5334   ;anti-imp bomb

   Slaver
          see Pit-Trapper.
          
   Stealth
          Property of programs, or program parts, which are invisible to
          scanners, accomplished by using zero B-fields and reflections.
          
   Stone
          A Stone-like program designed to be a small bomber. Part of the
          Paper-Scissors-Stone analogy.
          
   Stun
          A type of bomb which makes the opponent multiply useless
          processes, thus slowing it down. Example is referred to as a
          spl-jmp bomb.
          

    example    spl 0
               jmp -1

   Two-Pass Core-Clear (also spl/dat Core-Clear)
          core clear that fills core first with SPL instructions, then
          with DATs. This is very effective in killing paper and certain
          imp-spiral variations.
          
   Vampire
          see Pit-Trapper.
          
   Vector Launch
          one of several means to start an imp-spiral running. As fast as
          Binary Launch, but requiring much less code. See also JMP/ADD
          Launch and Binary Launch. This example is one form of a Vector
          Launch:
          

    impsize   equ 2667
    example   spl 1                       ; extend by adding more spl 1's
              spl 1
              djn.a   @imp,#0             ; jmp @ a series of pointers
              dat     #0,imp+(3*impsize)
              dat     #0,imp+(2*impsize)
              dat     #0,imp+(1*impsize)
              dat     #0,imp+(0*impsize)
    imp       mov.i   #0,impsize

   
   
   [ToC]
     _________________________________________________________________
   
Other questions?

   Just ask in the rec.games.corewar newsgroup or contact me (address
   below). If you are shy, check out the Core War archives first to see
   if your question has been answered before.
   
   [ToC]
     _________________________________________________________________
   
Credits

   Additions, corrections, etc. to this document are solicited. Thanks in
   particular to the following people who have contributed major portions
   of this document: Paul Kline, Randy Graham.
   Mark Durham wrote the first version of the FAQ.
   
   The rec.games.corewar FAQ is Copyright 1995 and maintained by:
   

Stefan Strack, PhD                          stst@vuse.vanderbilt.edu
Dept. Molecular Physiol. and Biophysics     stst@idnsun.gpct.vanderbilt.edu
Rm. 762, MRB-1                              stracks@vuctrvax.bitnet
Vanderbilt Univ. Medical Center             Voice: +615-322-4389
Nashville, TN 37232-6600, USA               FAX:   +615-322-7236

   
     _________________________________________________________________
   
   $Id: corewar-faq.html,v 3.4 1995/08/06 21:38:09 stst Exp stst $



From: pk6811s@acad.drake.edu
Subject: More Stuff
Date: 1995/09/11
Message-ID: <1995Sep11.071042@acad.drake.edu>#1/1
references: <CMM.0.90.2.810630782.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

"Steven C. Morrell" <morrell@math.utah.edu> writes:
>...
> snide remarks or comparing the success of Porch Swing/Withershins Twice-
> type programs (there must be a shorter name for these things!) with

One-shot scanners.  FlyPaper was a good one.  GammaPaper was a one-shot
bomber.

>...
>>> The easiest defense is to ALWAYS mod your strategy before using it.
> 
> Don't forget to increment afterwards.  JMP @0,0 is a pretty bad instruction
> for your warrior to execute.

This is never a problem using this sequence:

        mod.a   #3,1
        jmp     @0,str1
        dat      0,str2
        dat      0,str3
str1    strategy 1
str2    strategy 2
str3    strategy 3

Whatever your pspacer does, realistically you have about 15-20 cycles
to make up your mind, boot, and get started.  A qscanner will be
bombing you after that.  Expect to see some on a Hill near you.

Paul Kline
pk6811s@acad.drake.edu

Some (well, most) of Pink follows:

;redcode-94
;name Pink
;kill Pink
;author P.Kline
;strategy switching DieHard & core-sweep  (tie or win)

PIN      ; almost forgot to hide this one :-)
rndlmt   equ    ; number of trial rounds - some small number < 20
tscore   equ    ; acceptable trial score - aggressive
pzero    equ    ; somewhere in pspace
prndcnt  equ    (pzero-2)
pscore   equ    (pzero-8)
pstragy  equ    (pzero-14)
presult  equ    0
        
         dat    #gate-10,clr-gate+7    ; just in case clr is decremented
gate     dat    #4000   ,3000
wipe4    dat    #4000   ,clr-gate+7
wipe3    dat    #4000   ,clr-gate+7
wipe2    spl    #6000   ,clr-gate+7    ; redundant wipers
wipe1    spl    #3050   ,clr-gate+7
     for 6
         dat 0,0
     rof
clr      spl    #0,>-20
         mov    @2,>gate
         mov    @1,>gate
         djn.b  -2,{wipe1
     for 40
         dat 0,0
     rof
start    ldp.a  presult,wresult    ; get result of last battle
         sne.a  #-1,wresult        ; check for very first battle
         jmp    goforit
         ldp.a  #prndcnt,wrndcnt   ; get trial round counter
         ldp.a  #pscore ,wscore    ; get trial score
         ldp.a  #pstragy,wstragy   ; get trial strategy
wresult  jmp    @0,tlose
         dat    0 ,twin            ; couldn't we make this scoring system
         dat    0 ,ttie            ; just a little more difficult to manage??
twin     add.a  #3,wscore          ;
ttie     add.a  #1,wscore
tlose    djn.a  goforit,wrndcnt
wscore   slt    #0,#tscore         ; trial score acceptable?
         jmp    reset,}reset+1     ;    yes (grant edge for winning)
         add.a  #1,wstragy         ;    better try something else
                                   ; start a new trial period
reset    mov.a  #0,wscore          ;   set trial score to zero
         mov.a  #rndlmt,wrndcnt    ;   set trial round count to rndlmt
goforit  mod.a  #str1-wstragy,wstragy
wrndcnt  stp.ab #rndlmt+1,#prndcnt ; remember these three
         stp.ab wscore   ,#pscore  ;   "
         stp.ab wstragy  ,#pstragy ;   "
;break
wstragy  jmp    @0,clr
         dat     0,dh
str1     dat     0,0

dh       [insert DieHard here - not exactly a Silk replicator, but
          you could put one here for testing :-]

         dat    1,1                ; prevent Agony's djn overrunning
         dat    1,1                ; before we get started
         end    start




From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - Standard
Date: 1995/09/11
Message-ID: <199509110400.AAA16628@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html

Current Status of the StormKing Industries Standard KotH CoreWar Hill :

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  27/ 11/ 61                 Cannonade              P.Kline     143      96
 2  36/ 31/ 33              Keystone t21              P.Kline     141      83
 3  30/ 21/ 49    Der Zweiter Blitzkrieg      Mike Nonemacher     140      91
 4  28/ 18/ 54   CAPS KEY IS STUCK AGAIN       Steven Morrell     139      62
 5  39/ 40/ 20       Beholder's Eye V1.7         W. Mintardjo     138     140
 6  38/ 37/ 25               Christopher       Steven Morrell     138      61
 7  27/ 15/ 58              Blue Funk 88       Steven Morrell     138      60
 8  28/ 19/ 52                      Test       Wayne Sheppard     137      85
 9  41/ 47/ 12                  bigproba        nandor sieben     136      84
10  25/ 15/ 60           jmp/add crasher         Randy Graham     136      20
11  40/ 44/ 16                 Iron Gate       Wayne Sheppard     135     190
12  35/ 36/ 29              Giskard v0.5           Ken Mitton     135      34
13  38/ 41/ 22              Request v2.0     Brant D. Thomsen     135      32
14  25/ 14/ 61                     Peace            Mr. Jones     135      70
15  27/ 19/ 54                     Hydra      Stephen Linhart     135     170
16  24/ 14/ 61                      ttti        nandor sieben     135      46
17  36/ 37/ 27             Miss Careless           Derek Ross     134       6
18  22/  9/ 69         Imps! Imps! Imps!       Steven Morrell     134     107
19  20/ 10/ 70              Evol Imp v5a            Wilkinson     130       7
20   4/ 28/ 69              Imp Bomber 3        Darren Bergen      80       1

21   4/ 88/  8                   DriveBy     m. scott everard      20       0



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - ICWS Tournament
Date: 1995/09/11
Message-ID: <199509110400.AAA25336@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html


Current Status of the StormKing Industries Annual ICWS Tournament CoreWar Hill:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  62/ 10/ 28                 Cannonade           Paul Kline     214      63
 2  55/ 32/ 13             Miss Careless           Derek Ross     178      17
 3  53/ 38/  8                   Agony T        Stefan Strack     169      64
 4  51/ 35/ 14            Old Tire Swing         Randy Graham     167      20
 5  47/ 29/ 24              Giskard v0.5           Ken Mitton     166      36
 6  49/ 41/ 10                Miss Carry           Derek Ross     156      27
 7  46/ 39/ 15          DoubleStone v0.7    Christoph C. Birk     153       6
 8  46/ 39/ 15                 Maya v1.6    Christoph C. Birk     153      37
 9  42/ 33/ 24                     stone  matthew householder     151      75
10  42/ 37/ 21                warrior 42      stefan roettger     148      76
11  46/ 46/  7                       xtc      stefan roettger     146      68
12  43/ 42/ 15              Slaver v1.1i    Christoph C. Birk     145      24
13  42/ 46/ 12                  Illusion         Randy Graham     138      22
14  37/ 42/ 21                  scissors  matthew householder     133      78
15  36/ 46/ 18                  Cat v2.0           Tim Scheer     125      60
16  34/ 50/ 16             Smartbomb 4.0     Devin Kilminster     118      62
17  24/ 61/ 15                      test    Christoph C. Birk      88       2
18   3/ 36/ 60          Mr Administrator           Derek Ross      70       1
19   3/ 42/ 55          Mr Administrator           Derek Ross      64       8
20   4/  2/  4                      test    Christoph C. Birk      15       3

21   3/  0/  2            Miss Treatment           Derek Ross      10      28



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - Multiwarrior Experimental 94
Date: 1995/09/11
Message-ID: <199509110400.AAA14085@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html


Current Status of the StormKing Industries MultiWarrior Experimental 94 CoreWar Hill:

 #                      Name               Author   Score     Age
 1          TimeScapeX (0.1)       J. Pohjalainen    3550      32
 2                  Lucky 13        Stefan Strack    3105      34
 3                  Paperone          Beppe Bezzi    3097      17
 4                     jaded           M R Bremer    2888       3
 5                      life        Nandor Sieben    2663      33
 6                 lifedwarf        Nandor Sieben    2583       9
 7            Miss Treatment           Derek Ross    2344      12
 8                    Hidden M.C.Diskett Bullfrog    2263       2
 9             Miss Careless           Derek Ross    2222       1
10                 Whirlwind              Bob Uhl    2161      19
11                   Shwing!         T. H. Davies    2148      28
12                   Piggy 3              Bob Uhl    1857       5
13            AB Scanner 2.9         Chris Hodson    1840      21
14                   Piggy 2              Bob Uhl    1785       7
15            Illusion-94/55         Randy Graham    1656      11
16        Miss Understanding           Derek Ross    1616       8
17                Miss Carry           Derek Ross    1556      13
18               Whirlwind 2              Bob Uhl    1367      20
19                   Unknown            Anonymous    1130      10
20                   MuDwarf             G. Eadon     809      22

21                Veeble Jr.         T. H. Davies     785      29



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - MultiWarrior 94
Date: 1995/09/11
Message-ID: <199509110400.AAA26622@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html

Current Status of the StormKing Industries Multiwarrior 94 CoreWar Hill:

 #                      Name               Author   Score     Age
 1                Son of Imp       Steven Morrell    2421      19
 2                  Die Hard              P.Kline    2395       6
 3            75% Cotton v3b            Wilkinson    2286       3
 4                60% Cotton            Wilkinson    2270       2
 5            90% Cotton v5c            Wilkinson    2250       1
 6               Cotton v C1            Wilkinson    2227       5
 7              Silkworm 1.1          Beppe Bezzi    2107      16
 8           TimeScape (1.1)       J. Pohjalainen    2100      64
 9                    test 2              P.Kline    2029      23
10                  Paperone          Beppe Bezzi    2008      41

11             Miss Careless           Derek Ross    1131       0



From: John <jwilkinson@mail.utexas.edu>
Subject: KotH!
Date: 1995/09/11
Message-ID: <42vvuh$s25@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

When a warrior is pushed off of the hill, do all of the other warriors
that are still on the hill have the score that they made against that
warrior subtracted away?

It would seem logical that the answer is yes... but just thought I'd
ask.

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: README
Date: 1995/09/12
Message-ID: <434lqq$94o@geraldo.cc.utexas.edu>#1/1
references: <434jtr$bm2@agate.berkeley.edu>
newsgroups: rec.games.corewar

All info concerning everything (Corewars-related, that is :) may be found at

http://www.stormking.com/~koth/

Get pMARS 0.8, which is also linked from the above site.

-John




From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: PSPACE
Date: 1995/09/12
Message-ID: <434jtr$bm2@agate.berkeley.edu>#1/1
newsgroups: rec.games.corewar

Ummm, what exactly is P-space (what does it do, what opcodes does it use, 
and what is it good for?), and where can I find more information about 
it?  Thanks.





From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: P-Space
Date: 1995/09/12
Message-ID: <434jq7$bls@agate.berkeley.edu>#1/1
newsgroups: rec.games.corewar

Umm, exactly what is p-space (i.e. how does it work, what opcodes does it 
use, and what good is it?), and where can I find information about it?  
Thanks.




From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: P-Space
Date: 1995/09/12
Message-ID: <434jp1$bjv@agate.berkeley.edu>#1/1
newsgroups: rec.games.corewar

Umm, exactly what is p-space (i.e. how does it work, what opcodes does it 
use, and what good is it?), and where can I find information about it?  
Thanks.




From: John <jwilkinson@mail.utexas.edu>
Subject: pSHELL bug
Date: 1995/09/12
Message-ID: <434j60$6ho@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

In pSHELL, I cannot use the file selection menu to change directory
to one of the directories on my drive C.  I suspect that the problem
may be due to the fact that that particular directory contains a little
over 500 files in it.

You need to either fix this problem, or make it easy to type in the
path name of the warrior you want to load.  At the moment, I am forced
to copy warriors I want to use out of that directory, before using
them...  :(

-John




From: "Steven C. Morrell" <morrell@math.utah.edu>
Subject: PGP
Date: 1995/09/13
Message-ID: <CMM.0.90.2.810963847.morrell@jeeves.math.utah.edu>#1/1
newsgroups: rec.games.corewar

Okay, here is PGP.  My philosophy with Porch Swing and Withershins Thrice
is, "If you can't beat 'em, join 'em.  And then beat 'em."  So the
strategy is:

   - Boot away and leave a decoy behind
   - One-shot scan at 89% of c.
   - Multipass core-clear with a STP run thrown in.

Scanning at >80% c is not actually as tough as you might think.  P. Kline
unwound the scanning loop in Taking Names to 80% c.  I just unwound it a
little more.

The STP.f instruction is a little misleading, but since there are no
illegal instructions, why not?  Actually, this was written before I
realized that P-space just holds numbers between 0 and CORESIZE-1, and
not instructions or operand-pairs.  Pity the programmer who wants to
store instructions in P-space (for more reasons than this)!  The .f
operand is converted to the closest thing that makes sense, which is
.b so the STP.f # stores the B-value into where it points to points to.
Or something like that.

Since the core-clear modifies the B-field of the bomb it is carpeting
with, the core is colored to provide a reasonable sweep through P-space
with just one STP instruction, despite all the indirection.  Plus, the
values of munged P-space are somewhat unpredictable, so that the enemy
has a tough time testing whether his P-space was hit.

The code could be polished a little more, but it runs.  As usual, it
plays worst against Agony II.  To fix that, I wrote sdrawkcaB as a more
conservative version of PGP that scanned backwards, core-cleared
backwards, and DJN'ed foreward.  Nice, but the enemy tends to fall off
the SPL carpet onto more useful instructions.


;redcode-94 verbose
;name  PGP
;author  Steven Morrell
;strategy  Encrypt P-space for added security.
;kill  PGP

org boot

offset equ 500
stream equ 250
step equ 12
first equ (20+step)
flag equ 167

; stp equ nop                 ;uncomment to make this run when VERSION<80

x    mov p,p+offset           ;Sloppy!!!  (Just used as pointer)
boot mov }x,>x
     for 16
       mov }x,>x
     rof
     jmp scan+offset

p    dat first,first+step
     dat last-p,p
     stp.i #last-p,}p
     spl #last-p,}p-20
d    spl #step*-2,#step*-2    ;Also sloppy
loop sub d,p
scan sne *p,@p
     sub d,p
     sne *p,@p
     sub d,p
     sne *p,@p
     sub d,p
     sne *p,@p
     djn.f loop,<stream
     mov @1,}p                ;faster than a multi-process clear
     djn.b -1,<d-1            ;but stupid against imp-rings
     jmp -2

last dat {d-2,{d-2 
for 63
     dat {d-2,{d-2
rof

end

Steven
-- 
Steven Morrell                        morrell@math.utah.edu



From: morrell@jeeves.math.utah.edu (Steven C. Morrell)
Subject: Re: Pizza Weirdness!
Date: 1995/09/13
Message-ID: <MORRELL.95Sep12223750@jeeves.math.utah.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu>
newsgroups: rec.games.corewar

In article <42pjo5$eip@geraldo.cc.utexas.edu> jwilkinson@mail.utexas.edu writes:

-- >[discussion of fixed placements]

-- >The effect of fixed placements is to freeze the existing programs on the
-- >Hill in place while the submitter is free to change his boot distance,
-- >step size, internal spacing, etc.  But even when we had random spacing
-- >between submissions the changes in win/loss nearly always averaged out
-- >over the 20 opponents.

-- Indeed, however with random placements, at least I would be able to resubmit
-- my warrior a couple of times, and see which losses and wins were the
-- rule, and which were the exception.  That seems fair...

Place 'em randomly yourself!  Try copying your working program one or two
instructions from where it was before.  Or stick DAT 0,0 instructions at
the start of your warrior to change the spacing.  Especially if you are
doing something P-spaceish.

-- -John

Steven (trying from home)
-- 
Steven Morrell                      morrell@math.utah.edu



From: jimc@mainelink.net
Subject: Re: Request
Date: 1995/09/13
Message-ID: <435fg7$llk@news.mainelink.net>#1/1
references: <43229i$bqn@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

OK.  And for those of us who are new to Corewar, what is 
pspace?




From: graham@hal.mathcs.rhodes.edu (Randy Graham)
Subject: Porch Swing
Date: 1995/09/13
Message-ID: <1995Sep13.202708.2892@rhodes>#1/1
newsgroups: rec.games.corewar

Well, nothing much of interest here.  But, since I said I would
eventually give the EQU lines for Porch Swing, here they are (the
whole thing again, in fact).  she fell off the hill a few days ago.  I
tell you what, with PSpace, Porch and Withers got eaten up, but Agony
is still going VERY strong.  Now there's a warrior everyone should be
clamoring to see.  Anyway, here she is (including the bug that kept
here from perhaps scoring higher).

Randy
-----------------------------------------------------------------
;redcode-94
;name Porch Swing
;kill Porch Swing
;Author Randy Graham
;assert 1
;strategy Swing with a little wider range.
;strategy Now 80% bomb/scan with djn-stream once-thru

STEP    equ    12
EXTRA   equ    4
DJNOFF  equ    (-426-EXTRA)

        dat.f  #gate-10,  step-gate+5
gate    dat.f  #gate-10,  sneer-STEP+1
dat2    dat.f  #gate-20,  step-gate+5
dat1    dat.f  #gate-25,  step-gate+5
site2   spl.a  #gate-30,  step-gate+5
site    spl.a  #gate-40,  step-gate+5
for EXTRA
        dat.f  0,         0
rof

adder   sub.f  sweeper,   sneer
hithigh mov.i  step,      *sneer
hitlow  mov.i  step,      @sneer
sneer   sne.i  @gate+STEP*6-1-EXTRA, *gate+STEP*3-EXTRA  ;so we bomb step
looper  djn.b  adder,     <DJNOFF
setup   add.f  sneer,     gate
sweeper spl.a  #-STEP*4+1,<-STEP*4+1
mover   mov.i  @swinger,  >gate
swinger djn.b  mover,     {site
step    dat.f  <STEP,     <-STEP

end     hithigh



From: nfox@jagor.srce.hr (Nikola Fox - V Gimn.)
Subject: Re: Request
Date: 1995/09/13
Message-ID: <437qgk$db1@nippur.irb.hr>#1/1
references: <43229i$bqn@geraldo.cc.utexas.edu> <435fg7$llk@news.mainelink.net>
newsgroups: rec.games.corewar

jimc@mainelink.net wrote:
: OK.  And for those of us who are new to Corewar, what is 
: pspace?

yeah what IS pspace? And is there a '94 compatible MARS system for the Amiga?


--
<--[jE]--------------------------------------------------------------------.
__/\__| [NiKoLa^FoX a.K.a jUicE] [STuDeNT @ FeR iN ZaGReB^aMiGaN^MuSiCiaN] |
\oO  /| [>E<-MaiL: nfox@jagor.srce.hr] [iNTeReSTS: DeMoSCeNe^MuZaK^CYBeR*^ |
/_- _\| ^GRaFFiTi^aSCii/aNSi^DeZiGN^RaVE^MaNGa^FReeSTYLe^uNDeRGRouND^BeeR^ |
:.\/:.| ^SCi-Fi^CoNTaCTS^RPG/MuDS^BeaViS&BuTTHeaD^NeTSuRFiNG^aNaRCHY^eT-C] |
. :. .| [>CoMPuTeR<:a12oo^o2o@14MHZ^2MBRAM^42oMBHD^1942MoNiToR^144ooMoDeM] |
 .  . `------>[Keyboard not connected error. Press F1 to continue.]<-------'




From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: P-Space
Date: 1995/09/13
Message-ID: <437q3p$gld@lastactionhero.rs.itd.umich.edu>#1/1
references: <434jq7$bls@agate.berkeley.edu>
newsgroups: rec.games.corewar

Peter Young (pyoung1@uclink4.berkeley.edu) wrote:
: Umm, exactly what is p-space (i.e. how does it work, what opcodes does it 
: use, and what good is it?), and where can I find information about it?  
: Thanks.

STP - store to p space
LDP - load from p space

;redcode-94b
;author John K. Lewis
;strategy "Smart" program.  Uses P-Space to remember last fight.
;name Obvious

result	ldp #0,#0	; load location 0 (last result) into the b field 
prog	ldp #1, #0	; load in last program number used (1-4)
	seq #1, -2	; stick with a winner
	add #1, prog	; if we lost try next program
	seq #0,prog
	mov.ba prog, 1	; grab program pointer
	jmp.a 1, 0	; eventual pointer
jmp.a rusty		; index of choices
jmp.a pidge
jmp.a impring

rusty stp #1,#1		;store the number 1 to Pspace location 1

pidge stp #2,#1		;store the number 2 to Pspace location 1

impring stp #3,#1	;store the number 3 to Pspace location 1






From: mneumenth@kernvalley.com
Subject: Corewars Hill Idea
Date: 1995/09/13
Message-ID: <199509132320.TAA26205@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

How about a hill with SHARED P-Space.  Make P-Space largish, and give
each program an equal share by making them have a line like:
PSPACESTART EQU 4000
appended to each warrior just prior to being tested.  Would result in a
two-dimensional win-loss grid:
PAPER->STONE->SCISSOR->PAPER
PSPACE_USER->PSPACE_IDIOT->PSPACE_TRASHER->PSPACE_USER

Mneumenth, cyber-dragon of Earth
mneumenth@kernvalley.com
P.S.  Would post to r.g.c, but posting's down... :-(





From: lewin@netcom.com (Karl Lewin)
Subject: Re: Alzheimers Musings
Date: 1995/09/13
Message-ID: <lewinDEuGLH.Bpr@netcom.com>#1/1
references: <CMM.0.90.2.810963608.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

Steven C. Morrell (morrell@math.utah.edu) wrote:

: Fourth: Tips for protecting your P-space.  As STP's are NOP's in my
: implementation, I am not very authoritative, but I wanted you to
: read all of the rest before I told you this.  Bitter experience shows
: that you should always MOD and ADD #1 to your strategy before you
: JMP @0 it.  Any other hints?

Well, I'm not sure if this really reduces the effects "brain damage" too 
much because my component writing skill are very iffy, but I'm pretty 
sure that this will keep things from getting too out of hand.  Any 
comments/suggestions are very welcome.  

I did manage to put this on the hill (the bottom) for a brief period of 
time with a small stone, multipass clear, and CMP scanner as the 3 
components but I'm pretty sure they were far from optimal.

The overhead is pretty small as you are JMP ing to a component in 9-15 
cycles.

;pSpace equates
;------------------------------------
_RES	EQU	#0
_BAL	EQU	;whatever
_CUR	EQU	;whatever else

;other equates
;------------------------------------
NCOMP	EQU	3	;# of components
PSMARTS EQU	5	;brain damage #
NSMARTS EQU	4	;brain damage #
;------------------------------------
first	ldp.ab	_RES,	last
	ldp.a	_BAL,	balance
	ldp.a	_CUR,	launch

;keep from getting MUCH brain damage
;------------------------------------
	slt.a	#501,	balance
	jmp	plus
	slt.a	#-(NSMARTS+1),balance
	mov.a	#-NSMARTS,balance     
	jmp	last
plus	slt.ab	balance,#(PSMARTS+1)
	mov.a	#PSMARTS,balance      ;PSMARTS could be replaced
                                      ;with 1 maybe on this line
                                      ;to further minimize the 
                                      ;effects of the scrambler
	                              ;same thing goes for -NSMARTS
                                      ;on the mov.a line above

;determine whether to switch or not
;------------------------------------
last  	sne.ab	#1,	#0
	jmp.b	store,	}balance
	sne.ab	#0,	last
	sub.a	#1,	balance
balance	slt.ab	#0	,#501
switch	add.a	#1,	launch
store	mod.a	#NCOMP,	launch
 	stp.ab	launch,	_CUR
     	stp.ab	balance,_BAL

;launch proper strategy
;------------------------------------
launch	jmp	@1,	strat2
	dat	0,	strat3
	dat	0,	strat1
;------------------------------------
strat1
;------------------------------------
strat2
;------------------------------------
strat3
-- 







From: bezzi@iol.it (Beppe Bezzi)
Subject: Re: Pizza Weirdness!
Date: 1995/09/13
Message-ID: <199509130823.JAA16696@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar


Paul Kline wrote

>(Robert A. Uhl) writes:
>>...
>>   Why not add verification?  If mail is sent requesting a ;kill, just
>> mail back with a unique passwd.  The user sends it in with a ;rkill or
>> something, and it kills the old warrior and deletes the passwd from

>However, the usual purpose of ;kill is to remove the earlier version 
>of the same program so we don't end up with multiple copies on the Hill.
>So when you are submitting tests each one replaces the older one.  It
>would be unimaginably hard to do this if the Hill had to wait for
>a manual ;ok-to-kill response before finalizing the results of a submission.

A simple solution, without lots of work, may be not allow at all ;kill
of warriors aged more than 20 (10, 50 ?), or allow it only after
verification, and let it free, as is now, for younger ones.

What we don't want is seeing some stupid guy killing our old veterans
and, in the meantime, be free to test our young creatures. I don't
think Stefan will wake tomorrow and resubmit Agony thinking that
changing boot distance 20 lines would be better, and if someone kills
my new Brain Wash does nothing more than what I'll do tomorrow
submitting the new one.

Should happen someone begins killing all warriors he can many times we
can take some more steps for security. (And maybe try to backtrack him
for a good flame :-)

-Beppe




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Request
Date: 1995/09/13
Message-ID: <435lrq$2v1@geraldo.cc.utexas.edu>#1/1
references: <43229i$bqn@geraldo.cc.utexas.edu> <435fg7$llk@news.mainelink.net>
newsgroups: rec.games.corewar

jimc@mainelink.net wrote:
>OK.  And for those of us who are new to Corewar, what is 
>pspace?
>
Go to http://www.stormking.com/~koth/pmars.html and download pMARS 0.8.

Then read the WHATSNEW file.

-John




From: "Steven C. Morrell" <morrell@math.utah.edu>
Subject: Alzheimers Musings
Date: 1995/09/13
Message-ID: <CMM.0.90.2.810963608.morrell@jeeves.math.utah.edu>#1/1
newsgroups: rec.games.corewar

Judging from the discussion on this group, any help with PVamps would
be helpful.

First: the name.  Vamps are dead for the moment, and may stay dead as
long as we have A-field indirection.  Silk wasn't the first sign that
vamps were in trouble.  Brant Thomsen's Stimpy (I think) was a bomber
that used A-field anti-vamp techniques and overwhelmed the best anti-
anti-vamp anybody has ever come up with.  Perhaps Stimpy can be called
anti-anti-anti-vamp.  But I digress.

The point is, a Malevolent P-Space Modifying Program can take many
forms, except perhaps as a classical vampire.  I have implemented
this sort of thing in paper and one-shot scanners, and hope to have a
stone-imp combo that does this as soon as I can stop talking and get
to work.  Granted, the whole mess was conceived as a vampire strategy,
but the first reference to MPSMP's was by somebody clever whom I forgot
who called it an "Alzheimer's Vampire."  I like the name "Alzheimer's"
and, more importantly, so does my wife.

Second: Things to keep in mind while building an Alzheimer's program.

- As impressive as the 99/ 0/ 1 scores look, you need some other
  strategy to stay on the hill.  For some reason, all of my victims
  tended to fall of the hill as soon as they were attacked.  Perhaps
  because of low scores.   Against safe programs, you may be able to
  counter the switching routines somewhat, but you will have to fight
  against e.g. either Die Hard or Withershins Thrice.  Alzheimer's is
  best implemeted as almost an afterthought.

- There is too much indirection to make a P-space sweep in one
  instruction.  Normal sweeps can use a  MOV bomb,>p   instruction or
  something like that, but  STP #bomb,>p  will generally point  p  to
  a zero B-field, so you might as well use  STP #bomb,#0,  which you
  can't get the enemy to execute soon enough to do any good.

- If your simulator doesn't have P-Space, you can still test the
  conventional part of your program by using the line:

  stp equ nop

  This is most useful for MacPmars users.  Be sure to comment out this
  line when you send your program to the hill!

- You may want to vary the numbers you put into P-space if practical.
  Otherwise, your program generates a "signature" that the enemy may
  adapt to.

Third: How do I implement Alzheimer's in a multi-pass core-clear?

Traditionally, multi-pass core-clears use a SPL/SPL/DAT/DAT sequence
of core-clears.  Well, you've got to ask yourself how many extra enemy
processes the second SPL carpet generates.  Each core-clear takes
about 16000 cycles, so I figure that by the end of the first SPL
carpet, the enemy should be maxed out.  The second core-clear is just
there to let the enemy fester while you clean up any loose ends, so I
just changed it to an STP command with indirection.  The indirection
works in this case, because the first SPL carpet has colored the core
very nicely.  Then a DAT stream will knock out the rest.  As long as
you don't hit yourself, the last DAT stream doesn't really need to be
there, because you won't kill much with DAT's that you didn't kill
with SPL/STP/DAT's.

This reasoning may be wrong, because I am losing more to Marcia
Trionfale than Paul's or Randy's multi-pass clears.

Fourth: Tips for protecting your P-space.  As STP's are NOP's in my
implementation, I am not very authoritative, but I wanted you to
read all of the rest before I told you this.  Bitter experience shows
that you should always MOD and ADD #1 to your strategy before you
JMP @0 it.  Any other hints?


I talk too much.  The next post will have the source for PGP artfully
hidden in it.

Steven
-- 
Steven Morrell              morrell@math.utah.edu



From: tmcs01@cs.auckland.ac.nz (Tim McSweeney)
Subject: What is P-space
Date: 1995/09/13
Message-ID: <tmcs01-1309950927120001@uglq4.cs.auckland.ac.nz>#1/1
newsgroups: rec.games.corewar

as it says What is P-space, I've seen seceraql messages about it but there
is no mention in the faq?

Tim



From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: Abstraction of corewar strategise
Date: 1995/09/14
Message-ID: <4389hs$92m@nuscc.nus.sg>#1/1
newsgroups: rec.games.corewar

I finally realised why I have so much trouble figuring out the
corewar tutorials. I have been thinking about the instructions
as moving bits and bytes from location to location.

Could someone write something about the terms used for the 
strategies discussed in this group, e.g. what is meant by
bombing,etc.
TIA.

--
Calvin Loh



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Alzheimers Musings
Date: 1995/09/14
Message-ID: <437tnv$ks1@geraldo.cc.utexas.edu>#1/1
references: <CMM.0.90.2.810963608.morrell@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

>Fourth: Tips for protecting your P-space.  As STP's are NOP's in my
>implementation, I am not very authoritative, but I wanted you to
>read all of the rest before I told you this.  Bitter experience shows
>that you should always MOD and ADD #1 to your strategy before you
>JMP @0 it.  Any other hints?

For my switching routines, I've been using something like

case: jmp @0, str1
      jmp @0, str2
      jmp @0, str3

I'd ldp my strategy, mod #3 it (or whatever number...), and add it to case.
0's are no problem, and the next two lines could just as easily
be DAT's, but I sleep better with them being JMP's. :)

-John

>Steven Morrell              morrell@math.utah.edu





From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Pizza Weirdness!
Date: 1995/09/14
Message-ID: <437t26$ks1@geraldo.cc.utexas.edu>#1/1
references: <42lcr6$6mn@geraldo.cc.utexas.edu> <42lm66$81e@agate.berkeley.edu>	<42ln57$ebf@geraldo.cc.utexas.edu> <1995Sep7.090716@acad.drake.edu>	<42pjo5$eip@geraldo.cc.utexas.edu> <MORRELL.95Sep12223750@jeeves.math.utah.edu>
newsgroups: rec.games.corewar

>-- >[discussion of fixed placements]
>
>-- >The effect of fixed placements is to freeze the existing programs on the
>-- >Hill in place while the submitter is free to change his boot distance,
>-- >step size, internal spacing, etc.  But even when we had random spacing
>-- >between submissions the changes in win/loss nearly always averaged out
>-- >over the 20 opponents.

>-- Indeed, however with random placements, at least I would be able to resubmit
>-- my warrior a couple of times, and see which losses and wins were the
>-- rule, and which were the exception.  That seems fair...

>Place 'em randomly yourself!  Try copying your working program one or two
>instructions from where it was before.  Or stick DAT 0,0 instructions at
>the start of your warrior to change the spacing.  Especially if you are
>doing something P-spaceish.

>Steven Morrell                      morrell@math.utah.edu

Ugh.  That does not have the same effect.  Changing the dat 0's does not
change the spot that the first instruction excecutes from, if you're already
using all 100 lines.

I can try changing boot spots... but that shouldn't be nescessary in order
to get a better picture of how my warrior is doing, since changing boot
spots CHANGES the warrior... :(

Ah well, I was in second place for a while.  I think AgonyII retook it's
spot earlier today.  Man, that program's been around for a long time...

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: Another PShell bug
Date: 1995/09/15
Message-ID: <43arl1$otm@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Hey, I found yet another bug in pshell.  :)

This one is the most annoying yet... :(  If the paramaters that you pass
to pmars are longer than the DOS prompt limit, then characters get cut off.

Well, I suppose it's really a DOS bug than a pshell bug...  Anyway, you
need to make a temp file for pmars to read the paramaters from, instead
of passing them all from the prompt.

Of course, if anyone knows a way to circumvent this problem, I would
welcome it wholeheartedly!

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Porch Swing
Date: 1995/09/15
Message-ID: <43ares$otm@geraldo.cc.utexas.edu>#1/1
references: <1995Sep13.202708.2892@rhodes>
newsgroups: rec.games.corewar

graham@hal.mathcs.rhodes.edu (Randy Graham) wrote:

>Well, nothing much of interest here.  But, since I said I would
>eventually give the EQU lines for Porch Swing, here they are (the
>whole thing again, in fact).  she fell off the hill a few days ago.  I
>tell you what, with PSpace, Porch and Withers got eaten up, but Agony
>is still going VERY strong.  Now there's a warrior everyone should be
>clamoring to see.  Anyway, here she is (including the bug that kept
>here from perhaps scoring higher).

Thanks from me Randy...  now I'll finally know if my losses to Swing
were an anomoly cause by the -F option... :)

As for Agony II, my programs have been winning against it strongly...
I suppose mine are in the minority for some reason, because it's
still doing so well.

I'm really amazed that a scanner can do so well.  Not a single one of my
programs loses to Rave... :/  But my first attempts all lost big to Agony...

-John




From: bremermr@tweety.ecn.purdue.edu (Myer R. Bremer)
Subject: simple ptest
Date: 1995/09/15
Message-ID: <43c28j$ha9@mozo.cc.purdue.edu>
newsgroups: rec.games.corewar

Greetings.

As usual, success is a total surprise.  I had some time before
class so I cut some stuff out of endpoint and pasted in some
new code from another program.  I left the switching strategy
alone.  Then I just sent it to the hill and took off for classes.
Go figure.

By simple switching, I mean decisions made in < 10 cycles.  Way 
less although I haven't counted exactly.  Lame components isn't
the best choice of words, but the sub-warriors were not specialized
to complement each other and are in fact very simple.  I guess less
IS more.  When I have some time ( which won't be for awhile ) I'll
get around to optimizing everything.  

For your amusement, here are all the scores:
 
Program "Simple Ptest" (length 100) by "M R Bremer"
(contact address "bremermr@ecn.purdue.edu"):
has challenged the ICWS '94 Draft hill.
 
Program "Pink" (length 98) by "P.Kline"
(contact address "PDK@ADMIN.DRAKE.EDU"):
;strategy switching DieHard & core-sweep  (tie or win)
Simple Ptest wins: 42
Pink wins: 30
Ties: 28
 
Program "Armory - A5" (length 93) by "Wilkinson"
(contact address "jwilkinson@mail.utexas.edu"):
;strategy use pspace to go to battle
;strategy v 5 - well, I'm still losing to Brain Wash... I may still
;strategy have a pcode bug. :/
;strategy Major changes.  Hoping for more wins, and less ties...
Simple Ptest wins: 41
Armory - A5 wins: 52
Ties: 7
 
Program "Agony II" (length 99) by "Stefan Strack"
(contact address "stst@idnsun.gpct.Vanderbilt.Edu"):
;strategy Small-interval CMP scanner that bombs with a SPL 0 carpet.
;strategy This is the good old '88 Agony with some '94 enhancements:
;strategy  - boots off decoy to delay (quick)scanners; erases boot pointer
;strategy  - optimized decoy to avoid self-triggering for more than 6000 cycles
;strategy Submitted: Sun Jul 31 14:48:08 CDT 1994
Simple Ptest wins: 53
Agony II wins: 42
Ties: 5
 
Program "Ehni" (length 100) by "Maurizio Vittuari"
(contact address "pan0178@comune.bologna.it"):
;strategy What? Scanners with papers !?!
;strategy By the way at the moment I don't use Pspace...
Simple Ptest wins: 28
Ehni wins: 43
Ties: 29
 
Program "Thermite 1.0" (length 99) by "Robert Macrae"
(contact address "nobody"):
;strategy  Quick-scan -> incendiary bomber.
Simple Ptest wins: 31
Thermite 1.0 wins: 43
Ties: 26
 
Program "Brain Wash" (length 100) by "Beppe Bezzi"
(contact address "bezzi@iol.it"):
;strategy headache, bad thoughts, combat stress ?
;strategy you need a good Brain Wash :-)
;strategy v 07 new strategy
Simple Ptest wins: 40
Brain Wash wins: 41
Ties: 19
 
Program "Cthulhu v2" (length 85) by "Wilkinson"
(contact address "jwilkinson@mail.utexas.edu"):
;strategy Swallow up other warriors using pspace and divine power. :)
Simple Ptest wins: 39
Cthulhu v2 wins: 53
Ties: 8
 
Program "PGP" (length 100) by "Steven Morrell"
(contact address "morrell@math.utah.edu"):
;strategy  Encrypt P-space for added security.
Simple Ptest wins: 70
PGP wins: 26
Ties: 4
 
Program "sdrawkcaB" (length 100) by "Steven Morrell"
(contact address "morrell@math.utah.edu"):
;strategy  One-shot/two-pass
;strategy  Guess which way it scans?
Simple Ptest wins: 46
sdrawkcaB wins: 41
Ties: 13
 
Program "Marcia Trionfale" (length 99) by "Beppe Bezzi"
(contact address "bezzi@iol.it"):
;strategy A solid silk, Timescape like, base 
;strategy tuned for max scissors resistance
;strategy and some bombing add on
Simple Ptest wins: 35
Marcia Trionfale wins: 38
Ties: 27
 
Program "Cthulhu" (length 85) by "Wilkinson"
(contact address "jwilkinson@mail.utexas.edu"):
;strategy Swallow up other warriors using pspace and divine power
Simple Ptest wins: 49
Cthulhu wins: 27
Ties: 24
 
Program "PeepHole" (length 92) by "Randy Graham"
(contact address "graham@colossus.mathcs.rhodes.edu"):
;strategy New version of people.  Combine an imp/stone and a
;strategy stone/scanner.  Uses people's switching routine with
;strategy modifications to prevent specialization by opponents.
Simple Ptest wins: 59
PeepHole wins: 37
Ties: 4
 
Program "Torch t17" (length 42) by "P.Kline"
(contact address "PDK@ADMIN.DRAKE.EDU"):
;strategy very rapid incendiary bombing, core-clear & gate
;strategy t16: added spl/dat clear, removed boot & decoy 
;strategy t17: use djn-stream to keep Timescape & NobodySpecial at bay
Simple Ptest wins: 38
Torch t17 wins: 38
Ties: 24
 
Program "Mr Speculative" (length 16) by "Derek Ross"
(contact address "rossd@arbroath.win-uk.net"):
;strategy Scan, Carpet Bomb, Faster Core Clear ...
;strategy Written in '94 code
Simple Ptest wins: 40
Mr Speculative wins: 45
Ties: 15
 
Program "endpoint ." (length 100) by "M R Bremer"
(contact address "bremermr@ecn.purdue.edu"):
;strategy Very basic p-test. Remember, this is only a test.
Simple Ptest wins: 56
endpoint . wins: 26
Ties: 18
 
Program "Miss Careless" (length 13) by "Derek Ross "
(contact address "rossd@arbroath.win-uk.net"):
;strategy 75% bomb/scan once through core.
;strategy Then SPL carpet followed by SPL/DAT coreclear ...
;strategy Tuned against scanners then tweaked.
;strategy Written in '88 code.
;strategy Submitted: 31 August 95 
Simple Ptest wins: 54
Miss Careless wins: 29
Ties: 17
 
Program "Die Hard" (length 87) by "P.Kline"
(contact address "PDK@ADMIN.DRAKE.EDU"):
;strategy minimal kill, maximal survival
Simple Ptest wins: 0
Die Hard wins: 15
Ties: 85
 
Program "juliet storm " (length 100) by "M R Bremer"
(contact address "bremermr@expert.cc.purdue.edu"):
;strategy bomber--yippee 
;strategy absolutely zero paper protection
Simple Ptest wins: 49
juliet storm wins: 25
Ties: 26
 
Program "HeremScimitar" (length 32) by "A.Ivner,P.Kline"
(contact address "d91andiv@und.ida.liu.se"):
;strategy Fast bomber
Simple Ptest wins: 52
HeremScimitar wins: 38
Ties: 10
 
Program "Commision - PD9C" (length 99) by "Wilkinson"
(contact address "jwilkinson@mail.utexas.edu"):
;strategy use pspace
Simple Ptest wins: 39
Commision - PD9C wins: 26
Ties: 35
 
Program "Simple Ptest" (length 100) by "M R Bremer"
(contact address "bremermr@ecn.purdue.edu"):
;strategy playing that p-space magic
;strategy simple switching--lame components
;strategy I'll have something better after these my exams
Simple Ptest wins: 39
Ties: 17
 
Your overall score: 149.571429
 



From: r22666@zuma.sps.mot.com (Ramon Hontiveros)
Subject: Info req'd
Date: 1995/09/15
Message-ID: <43b01l$n5v@newsgate.sps.mot.com>#1/1
newsgroups: rec.games.corewar

Hi all,

I'm new to this newsgroup but not to corewar, though I haven't been
actively involved in any redcode programming lately. I was wondering
if there's an anon ftp site where I can get info on the "koth hill", 
"p-space", and the other stuff I see posted on this newsgroup. I can't
get to the web home page 'cos my system sits behind a firewall that doesn't
support HTML access (yet).

Thanks :)

-- Ramon 1990 KR-1S (pickled)

p.s. Is the "hill" sort of like a mailing list multi-player corewar scenario?




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: How do I kill with an Imp
Date: 1995/09/17
Message-ID: <43hh2p$7us@geraldo.cc.utexas.edu>#1/1
references: <43hgaf$omb$2@mhade.production.compuserve.com>
newsgroups: rec.games.corewar

Jonathan West <73742.1715@CompuServe.COM> wrote:
>TOday I sent a corewar program to the begginers hill called 
>"He-HE you can kill me!". It needs an improvent even thoguth it 
>placed 13 on "The hill". I am wondering how do I kill an enemys 
>program with an imp?

Well, there are many ways.  The most effective has been the imp
spiral.  Have you visited http://www.stormking.com/~koth ?
It has some tutorials to explain everything...

-John




From: Jonathan West <73742.1715@CompuServe.COM>
Subject: How do I kill with an Imp
Date: 1995/09/17
Message-ID: <43hgaf$omb$2@mhade.production.compuserve.com>#1/1
newsgroups: rec.games.corewar

TOday I sent a corewar program to the begginers hill called 
"He-HE you can kill me!". It needs an improvent even thoguth it 
placed 13 on "The hill". I am wondering how do I kill an enemys 
program with an imp?

-- 
This is a .sig virus. Please copy me into your signature



From: Jonathan West <73742.1715@CompuServe.COM>
Subject: How can I kill with an Imp
Date: 1995/09/17
Message-ID: <43hg6i$omb$1@mhade.production.compuserve.com>#1/1
newsgroups: rec.games.corewar

Today, I sent a corewar program to the begginers hill. Its called 
"He-He you can't kill me

-- 
This is a .sig virus. Please copy me into your signature



From: bremermr@cartoon.ecn.purdue.edu (Myer R. Bremer)
Subject: simple ptest
Date: 1995/09/18
Message-ID: <43ksej$fq3@mozo.cc.purdue.edu>#1/1
newsgroups: rec.games.corewar

Greetings.

Speaking of truly adaptive warriors: simple ptest is a test in
dynamic switching.  It's a very simple switching mechanism, but
I was hoping to catch some of the other pspace programs after
they set their strategies.  And with all this new pspace magic,
can't somebody do someting about Agony.  Sheesh.

M R Bremer    




From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Reducing score variability
Date: 1995/09/18
Message-ID: <9509181851.AA10532@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

jwilkinson@mail.utexas.edu wrote:
>Simply run the battle between
>each warrior 5 or 10 times, with random placements.  That way, the standard
>deviation of battles would be GREATLY reduced.  At the moment, with
>only one 100 round battle per each set of two warriors, you get a very
>poor gauge of a pspace warrior's average performance.

That's a claim without any backing. In fact, I can come up with arguments that
running 5*100 rounds would produce *more* variability than 1*500. Fortunately,
John, you can test your hypothesis at home by using a number of published
warriors and report the results to us. It's a little much to ask Tuc and Thomas
to change the scripts based on a hunch.

-Stefan



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: (no subject)
Date: 1995/09/18
Message-ID: <43kupf$lhl@geraldo.cc.utexas.edu>#1/1
references: <43k2u2$t7l@geraldo.cc.utexas.edu> <43kkos$hmj@lastactionhero.rs.itd.umich.edu>
newsgroups: rec.games.corewar

jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis) wrote:
>
>I think this is a bad idea. Firstly, if I have ten strategies to try out 
>then haveing only five battles would mean I never try all of them.  Even
>with only three strategies in a five battle set I would still need to 
>re-test myself every reset.  This also closes any chance that a new
>breed of dwarf that change thier constants might have to even be 
>developed.
>
>John -

Um, you must have greatly misunderstood.
I'm not exactly sure what you're talking about, but the change I suggest
would not change the strategy that warriors use in ANY WAY WHATSOEVER.

The benefit of this change is that we run 500 rounds of battles against
the two warriors, and group them into 5, 100 round sections, resetting
the pspace in between the 5 battles...

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: simple ptest
Date: 1995/09/18
Message-ID: <43kui6$lhl@geraldo.cc.utexas.edu>#1/1
references: <43ksej$fq3@mozo.cc.purdue.edu>
newsgroups: rec.games.corewar

bremermr@cartoon.ecn.purdue.edu (Myer R. Bremer) wrote:
>Greetings.
>
>Speaking of truly adaptive warriors: simple ptest is a test in
>dynamic switching.  It's a very simple switching mechanism, but
>I was hoping to catch some of the other pspace programs after
>they set their strategies.  And with all this new pspace magic,
>can't somebody do someting about Agony.  Sheesh.

Well, my programs are slowly getting pushed down... but all of them
beat Agony fairly badly...

Ah well, maybe whatever is letting my programs beat Agony, is keeping
them from scoring well elsewhere...  Hmmm...

-John 




From: pk6811s@acad.drake.edu
Subject: Re: (no subject)
Date: 1995/09/18
Message-ID: <1995Sep18.161325@acad.drake.edu>#1/1
references: <43k2u2$t7l@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John <jwilkinson@mail.utexas.edu> writes:

> Don't change the setup of the hill any.  Simply run the battle between
> each warrior 5 or 10 times, with random placements.  That way, the standard
> deviation of battles would be GREATLY reduced.  At the moment, with
> only one 100 round battle per each set of two warriors, you get a very
> poor gauge of a pspace warrior's average performance.

5x100 would reward the pspace approach which chooses a strategy early and
commits to it for the rest of the battles.  A truly adaptive program would
continuously monitor its strategy and switch off a bad performer.

But it would be better than the current setup.  The deviation in results for 
pspace warriors with 1x100 battles is huge.  5 rounds of 100 battles would be
better.  

Paul Kline
pk6811s@acad.drake.edu



From: DURHAM_M@msn.com (Melinda Durham)
Subject: RE: New Addresses For Mark A. Durham
Date: 1995/09/18
Message-ID: <00001f35+00000236@msn.com>#1/1
references: <00001f35+00000235@msn.com>
newsgroups: rec.games.corewar

Oh, my wife is graciously allowing me to share in her MSN account 
until we figure out how to set up separate mailboxes.

Mark A. Durham
durham_m@msn.com



From: John <jwilkinson@mail.utexas.edu>
Subject: (no subject)
Date: 1995/09/18
Message-ID: <43k2u2$t7l@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Don't change the setup of the hill any.  Simply run the battle between
each warrior 5 or 10 times, with random placements.  That way, the standard
deviation of battles would be GREATLY reduced.  At the moment, with
only one 100 round battle per each set of two warriors, you get a very
poor gauge of a pspace warrior's average performance.

Please implement this.  I cannot think of any positive reason to NOT
make this change.

>	Have you put this up on the news group?  If not, put it up and see
>what people think!
>
>				Tuc 

Ok, well there it is.  Come on guys, tell me what you think...
The strategy of all warriors would remain exactly the same, the only
difference would be pspace battles wouldn't be left so much to chance...

-John






From: DURHAM_M@msn.com (Melinda Durham)
Subject: New Addresses For Mark A. Durham
Date: 1995/09/18
Message-ID: <00001f35+00000235@msn.com>#1/1
newsgroups: rec.games.corewar

Hello again everyone! I have moved both in physical space and in cyberspace.
My new addresses are:

	20 Mockingbird Court
	Spartanburg SC 29307-3733
	durham_m@msn.com

Hopefully I will be able to resume more vigorous activity in Core War (and
rec.games.corewar) in the near future. I have to get past "My Fair Lady"
(I am Professor Zoltan Karpathy, that marvelous boy!) first though.

Mark A. Durham
durham_m@msn.com



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Reducing score variability
Date: 1995/09/19
Message-ID: <43n50u$6k3@geraldo.cc.utexas.edu>#1/1
references: <9509181851.AA10532@idnsun.gpct.Vanderbilt.Edu.noname> <1995Sep19.130722@acad.drake.edu>
newsgroups: rec.games.corewar

>> That's a claim without any backing. In fact, I can come up with arguments that
>> running 5*100 rounds would produce *more* variability than 1*500. Fortunately,
>> John, you can test your hypothesis at home by using a number of published
>> warriors and report the results to us. It's a little much to ask Tuc and Thomas
>> to change the scripts based on a hunch.
>
>Here are the results from running 10 rounds of 100 battles each for
>Pink vs. Agony51 and for Torch vs. Agony51.

<results>
>
>Pink is a pspace warrior and Torch is not.  The standard deviation
>is about 20 in both cases.  The difference between the lowest and highest
>scores is 68 for Pink, and 61 for Torch.  The difference between the
>lowest-to-highest number of wins is 21 for Pink and 20 for Torch.
>
My tests came up with an average standard deviation from the mean
number of wins of 10.6 for Fun v.e1 vs Tornado v1.1.
And a average standard deviation from the mean number of wins of
4.68 for Fire Storm vs. Tornado.

Similarly, the biggest gap was 27 wins for Fun and Tornado, and
only an 18 gap for Fire Storm and Tornado.

I've done tests like this over and over, and I keep getting results
like that.  Perhaps the thing is that my dynamic switching procedure
can get set off real easily.  However, I think we'll be seeing more
switchers like mine in the future...

-John




From: pk6811s@acad.drake.edu
Subject: Re: Reducing score variability
Date: 1995/09/19
Message-ID: <1995Sep19.130722@acad.drake.edu>#1/1
references: <9509181851.AA10532@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

(Stefan Strack) writes:
> 
> That's a claim without any backing. In fact, I can come up with arguments that
> running 5*100 rounds would produce *more* variability than 1*500. Fortunately,
> John, you can test your hypothesis at home by using a number of published
> warriors and report the results to us. It's a little much to ask Tuc and Thomas
> to change the scripts based on a hunch.
> 

Here are the results from running 10 rounds of 100 battles each for
Pink vs. Agony51 and for Torch vs. Agony51.

Pink				Torch			
Win	Loss	Tie	Score	|Win	Loss	Tie	Score
-------------------------------------------------------------
19	24	57	114	|47	44	9	150
29	16	55	142	|54	40	6	168
19	19	62	119	|43	50	7	136
25	21	54	129	|36	54	10	118
20	18	62	122	|56	34	10	178
8	42	50	74	|47	43	10	151
22	24	54	120	|56	33	11	179
11	34	55	88	|55	39	6	171
14	35	51	93	|49	40	11	158
15	22	63	108	|49	43	8	155
				|			
mean			110.9	|mean			156.4	
sumofsquares		3790.9	|sumofsquares		3290.4	
variance		421.2	|variance		365.6	
std deviation		20.5	|std deviation		19.1

Pink is a pspace warrior and Torch is not.  The standard deviation
is about 20 in both cases.  The difference between the lowest and highest
scores is 68 for Pink, and 61 for Torch.  The difference between the
lowest-to-highest number of wins is 21 for Pink and 20 for Torch.

Pink continuously monitors the performance of its current strategy,
and since it doesn't really have a strategy to beat Agony, it 
continuously rotates them in a futile attempt to discover a winning one :-(

YMMV

Paul Kline
pk6811s@acad.drake.edu

(did I really type 'sheek' meaning 'chic'? :-)



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: simple ptest
Date: 1995/09/19
Message-ID: <43mt3a$1ba@geraldo.cc.utexas.edu>#1/1
references: <43ksej$fq3@mozo.cc.purdue.edu> <43mjaq$c4b@lastactionhero.rs.itd.umich.edu>
newsgroups: rec.games.corewar

jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis) wrote:

>: can't somebody do someting about Agony.  Sheesh.

>This kills Agony:

>;redcode-94
>;name Agony Killer
>;author John Lewis
>;strategy Jump Jump Jump
>
>	spl 1
>	mov.i	<2,	<1
>	jmp 	@0,	-7	
>	dat 	#0,	#0	

Hey, that's pretty nifty...  I tested it out, and it does amazing well
for such an odd and fragile program...  Also, the small size would make
it a great program to stick in in order to simply kill Agony... cool. :)

Maybe Agony will fall soon...

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: (no subject)
Date: 1995/09/19
Message-ID: <43msp4$v6@geraldo.cc.utexas.edu>#1/1
references: <43k2u2$t7l@geraldo.cc.utexas.edu> <43kkos$hmj@lastactionhero.rs.itd.umich.edu> <43kupf$lhl@geraldo.cc.utexas.edu> <43mj58$c4b@lastactionhero.rs.itd.umich.edu>
newsgroups: rec.games.corewar

>: The benefit of this change is that we run 500 rounds of battles against
>: the two warriors, and group them into 5, 100 round sections, resetting
>: the pspace in between the 5 battles...

>I must have had too much caffeine.  I understand now.  I thought you wanted
>to have five battle resets to the 100 battle set.  (reset p-space after
>every five - 10 battles.)  I now understand you mean to have 5-10 sets
>of 100 battles.  Sorry.  :)

>John - (i think i need to sleep more)

Ah.  Gotcha...  That -would- be stupid...
However, what do you think of my idea?  Is there any reason NOT to implement
this minor change on the current hill?

-John (sleep is fun! :)






From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: simple ptest
Date: 1995/09/19
Message-ID: <43mjaq$c4b@lastactionhero.rs.itd.umich.edu>#1/1
references: <43ksej$fq3@mozo.cc.purdue.edu>
newsgroups: rec.games.corewar

Myer R. Bremer (bremermr@cartoon.ecn.purdue.edu) wrote:
: Greetings.

: Speaking of truly adaptive warriors: simple ptest is a test in
: dynamic switching.  It's a very simple switching mechanism, but
: I was hoping to catch some of the other pspace programs after
: they set their strategies.  And with all this new pspace magic,
: can't somebody do someting about Agony.  Sheesh.

: M R Bremer    

This kills Agony:

;redcode-94
;name Agony Killer
;author John Lewis
;strategy Jump Jump Jump
;
;


	spl 1
	mov.i	<2,	<1
	jmp 	@0,	-7	
	dat 	#0,	#0	






From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: (no subject)
Date: 1995/09/19
Message-ID: <43mj58$c4b@lastactionhero.rs.itd.umich.edu>#1/1
references: <43k2u2$t7l@geraldo.cc.utexas.edu> <43kkos$hmj@lastactionhero.rs.itd.umich.edu> <43kupf$lhl@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John (jwilkinson@mail.utexas.edu) wrote:
: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis) wrote:
: >
: >I think this is a bad idea. Firstly, if I have ten strategies to try out 
: >then haveing only five battles would mean I never try all of them.  Even
: >with only three strategies in a five battle set I would still need to 
: >re-test myself every reset.  This also closes any chance that a new
: >breed of dwarf that change thier constants might have to even be 
: >developed.
: >
: >John -

: Um, you must have greatly misunderstood.
: I'm not exactly sure what you're talking about, but the change I suggest
: would not change the strategy that warriors use in ANY WAY WHATSOEVER.

: The benefit of this change is that we run 500 rounds of battles against
: the two warriors, and group them into 5, 100 round sections, resetting
: the pspace in between the 5 battles...

: -John



I must have had too much caffeine.  I understand now.  I thought you wanted
to have five battle resets to the 100 battle set.  (reset p-space after
every five - 10 battles.)  I now understand you mean to have 5-10 sets
of 100 battles.  Sorry.  :)

John - (i think i need to sleep more)





From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Reducing score variability
Date: 1995/09/19
Message-ID: <43l5os$q9c@geraldo.cc.utexas.edu>#1/1
references: <9509181851.AA10532@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar

stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack) wrote:
>>Simply run the battle between
>>each warrior 5 or 10 times, with random placements.  That way, the standard
>>deviation of battles would be GREATLY reduced.  At the moment, with
>>only one 100 round battle per each set of two warriors, you get a very
>>poor gauge of a pspace warrior's average performance.
>
>That's a claim without any backing. In fact, I can come up with arguments that
>running 5*100 rounds would produce *more* variability than 1*500. Fortunately,
>John, you can test your hypothesis at home by using a number of published
>warriors and report the results to us. It's a little much to ask Tuc and Thomas
>to change the scripts based on a hunch.
>
>-Stefan

Gee, Stefan, holding my feet to the fire are we? :)

Yes, actually, I think 5*100 would produce less variability than 1*500 in
most cases...  It depends, of course, on the nature of the program's
p-switching procedure.  Opening test pswitchers would vary less with 5*100
battles.  Dynamic switchers would reach a happy resonance at some point,
and probably before the 500 mark.

Anyhow, that's not the point...
The point is, I already have done lots of tests, and 5*100 gives a
dramatically better picture of a warriors performance.

I do not recommend switching over to 1*500, simply because that changes
the nature of the game.  If you want a 1*500 hill, I think it should
be seperate from the current hill, whereas a 5*100 hill is in no
way different from the perspective of the redcodes.

-John




From: Derek Ross <rossd@arbroath.win-uk.net>
Subject: Re: Agony killer
Date: 1995/09/20
Message-ID: <245@arbroath.win-uk.net>#1/1
newsgroups: rec.games.corewar

 
John Lewis (jklewis@stimpy.us.itd.umich.edu) wrote:
>This kills Agony:
>
>;redcode-94
>;name Agony Killer
>;author John Lewis
>;strategy Jump Jump Jump
>;
>;
>
>
>       spl 1
>       mov.i   <2,     <1
>       jmp     @0,     -7      
>       dat     #0,     #0      
>


Agony Killer ?  It must be an ANTAGONIST :o}







From: rogbarn@Walden.MO.NET (Roger Barnes)
Subject: What IS Agony?
Date: 1995/09/20
Message-ID: <43pvo0$hos@Twain.MO.NET>#1/1
newsgroups: rec.games.corewar

 As a Corewar newbie, I'm rather curious about Agony.  What is Agony?  
Scanner, Vamp, Stone, whatever?  A simple explanation would be very 
appriciated.  Thanks.

Nate Barnes, using his father's account.




From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: InfoWarCon - Core Wars?
Date: 1995/09/20
Message-ID: <43peos$efd@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

Hi,

	I was on a project and could not break away, but did anyone see Stuart
and Jo at InfoWarCon earlier this month and thier presentation on 
"Core Wars: Practicing Information Warfare in Cyberspace"?

			Tuc

--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: netwalker@linetime.demon.co.uk (Murray Crane)
Subject: Precompiled PMars
Date: 1995/09/20
Message-ID: <643.6471T788T178@linetime.demon.co.uk>#1/1
newsgroups: rec.games.corewar

Reply to: netwalker@linetime.demon.co.uk

Hi everyone,

It's a bit of a long shot, but would anyone be willing to either supply me
with or tell me where I can get hold of a precompiled version of the latest
PMars for any of the following systems (in order of wanting):-

	Amiga, Apple Mac or PC

Any help would be most appreciated.

Cheerio.

+--------------
|  Murray L. Crane - sis2471@sis.port.ac.uk         (until July '96)
|                    netwalker@linetime.demon.co.uk (Polled account)
+--------------
| Bashing it out in Imagine for all the Universe to see
+--------------

"REMIND ME AGAIN HOW THE LITTLE HORSE-SHAPED ONES MOVE."




From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: What is a process?
Date: 1995/09/21
Message-ID: <43qjs2$sib@nuscc.nus.sg>#1/1
newsgroups: rec.games.corewar

I have just gone through the first 3 warriors in Steve Morrell's
corewar tutorial and came across a term I do not really 
understand. What is meant by process?
Is a process a line of instruction?
Or is it the sequence of instructions making up the warrior?
TIA.

--
Calvin Loh



From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: What IS Agony?
Date: 1995/09/21
Message-ID: <43sfs5$agj@lastactionhero.rs.itd.umich.edu>#1/1
references: <43pvo0$hos@Twain.MO.NET>
newsgroups: rec.games.corewar



Roger Barnes (rogbarn@Walden.MO.NET) wrote:
:  As a Corewar newbie, I'm rather curious about Agony.  What is Agony?  
: Scanner, Vamp, Stone, whatever?  A simple explanation would be very 
: appriciated.  Thanks.

: Nate Barnes, using his father's account.

It's an evilly quick scanner and stunner that finds your very quickly
and overwrites you with SPL 0.  ( I think )  and then eventually
it cleans up core with DAT statments.  At least that's what I heard.

John Lewis



From: survey@forefront.princeton.edu (Matt Webster)
Subject: Please Participate in a Survey!
Date: 1995/09/22
Message-ID: <43v2ml$i54@cnn.Princeton.EDU>#1/1
newsgroups: rec.games.corewar

Hello, fellow 'netters.  I am conducting a survey of Internet public opinion.
Please take a few moments and point your World Wide Web browser at 

http://forefront.princeton.edu/rec/263/

The questions will take only a few minutes and will be of great help to my
research, which is purely academic in nature.  Your anonymity is guaranteed and
individual responses will not be shared with other organizations.  Feel free to
email if you have any questions.

Thanks for your time,
Matt Webster

survey@forefront.princeton.edu





From: bremermr@elsun344.cc.purdue.edu (Myer R. Bremer)
Subject: cmp scanners
Date: 1995/09/23
Message-ID: <441qc2$l8e@mozo.cc.purdue.edu>#1/1
newsgroups: rec.games.corewar

Greetings.

To use a slt to keep from attacking yourself, the scanning
engine could look like:

add 
cmp
slt
djn

Less lines, plus a djn stream.  Check out rave or iron gate.

M R Bremer
 



From: everard@infi.net (M. Scott Everard)
Subject: Re: What is a process?
Date: 1995/09/23
Message-ID: <441nga$1b2@allnews.infi.net>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

In article <440gar$nm0@geraldo.cc.utexas.edu>, John <jwilkinson@mail.utexas.edu> says:
>
>>>I have just gone through the first 3 warriors in Steve Morrell's
>>>corewar tutorial and came across a term I do not really 
>>>understand. What is meant by process?
>>>Is a process a line of instruction?
>>>Or is it the sequence of instructions making up the warrior?
>
>>>Calvin Loh
>
>>A process is an instance of a program in execution.
>
>What a vague answer! :/
>
Accurate though.

(jeez, there's a critic on every corner!)

















From: John <jwilkinson@mail.utexas.edu>
Subject: Re: What is a process?
Date: 1995/09/23
Message-ID: <440gar$nm0@geraldo.cc.utexas.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net>
newsgroups: rec.games.corewar

>>I have just gone through the first 3 warriors in Steve Morrell's
>>corewar tutorial and came across a term I do not really 
>>understand. What is meant by process?
>>Is a process a line of instruction?
>>Or is it the sequence of instructions making up the warrior?

>>Calvin Loh

>A process is an instance of a program in execution.

What a vague answer! :/

A process is like a seperate program, launched by the warrior.
Basically, Corewars programs have the ability to do multitasking.
One program may launch other programs, which will be excecuted on
a time-share basis.

-John




From: Politics@usa.com
Subject: Proposed State & Federal Regulations for the INTERNET!
Date: 1995/09/23
Message-ID: <440cmn$9fo@newsie.wis.com>#1/1
newsgroups: rec.games.corewar

  My name is Scott Glasrud, and I am running for the New Mexico State Senate
during the 1996 elections. One of the reasons I have chosen to run is to combat 
the 
proposed state and federal regulations of the Internet.  As you know, the 
Internet 
was never designed to be regulated!  It was designed to allow communications in
the event of anuclear war or a major catastrophe. I OPPOSE REGULATION, and if 
elected
will fight to preserve your constitutional rights. HOWEVER, I NEED YOUR HELP!

I am asking each person who reeives this message to send $5.00 to the 
Scott Glasrud Campaign Committee.  If we pull together, we CAN protect our first
amendment rights!  HELP ME show the politicians the POWER behind this
important NETWORK.  Please send contributions to:

                                    The Scott Glasrud Campaign Committee
                                    11024 Montgomery Blvd. NE, Suite 179
                                         Albuquerque, New Mexico  87111



Thank you!




From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: Re: warrior question
Date: 1995/09/23
Message-ID: <441ha4$p3s@agate.berkeley.edu>#1/1
references: <441620$air@enterprise.america.com>
newsgroups: rec.games.corewar

Having never done this before, and only seen one instance of this, I'll 
attempt to answer your question anyways.  You stick it in the scanner.  
Basically, it should go something like this.

add.ab   #10, $1          ; update where scanning
cmp.ab   @10, @20         ; scan
jmp      $-2              ; nothing found, so go update scanner
slt.a    $10, $-2         ; Skip if the value scanned > progam length
jmp      $-4              ; just scanned program, so go update scanner

Anyways, you'd stick your bomb in next.  This bit of code should work, 
although, I'm not too sure how effective it'll be.





From: daredevl@enterprise.america.com (pagefault)
Subject: warrior question
Date: 1995/09/23
Message-ID: <441620$air@enterprise.america.com>#1/1
newsgroups: rec.games.corewar

How does one make a scanner skip itself when scanning/bombing?  I heard 
you use SLT, does the SLT go in the scanner or the bomb?
-- 
pagefault - daredevl@america.com
UnseenTerror Dragon -=[UDIC]=-

"Fear of corrupting the mind of the younger generation is the loftiest form of cowardice" - Holbrook Jackson



From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Re: What IS Agony
Date: 1995/09/23
Message-ID: <9509230622.AA11953@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

Agony II is a CMP-scanner that bombs with a SPL carpet to stun, followed by a
simple DAT core clear and imp-gate. It's similar to Rave, but really just a
version of Agony 3.1 (posted > 2 yrs ago) that boots off a large decoy which
is optimized to to avoid triggering Agony's own scan until 6000 cycles into
the round. It does well against one-shot scanners like Withershins/Porch
because it boots just ahead of the decoy and finishes them off while they're
busy core clearing. I don't really know why Agony does so well right now.
I'll post source when Agony reaches age 1000 or is bumped off, whichever
comes first.

-Stefan



From: everard@infi.net (M. Scott Everard)
Subject: Re: What is a process?
Date: 1995/09/23
Message-ID: <43vq0s$3mb@allnews.infi.net>#1/1
references: <43qjs2$sib@nuscc.nus.sg>
newsgroups: rec.games.corewar

In article <43qjs2$sib@nuscc.nus.sg>, lohwengk@iscs.nus.sg (Loh Weng Keong Calvin) says:
>
>I have just gone through the first 3 warriors in Steve Morrell's
>corewar tutorial and came across a term I do not really 
>understand. What is meant by process?
>Is a process a line of instruction?
>Or is it the sequence of instructions making up the warrior?
>TIA.
>
>--
>Calvin Loh

A process is an instance of a program in execution.

















From: max@alcyone.darkside.com (Erik Max Francis)
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <w9gyBD1w200w@alcyone.darkside.com>#1/1
references: <442ouq$1ai@nuscc.nus.sg>
newsgroups: rec.games.corewar

lohwengk@iscs.nus.sg (Loh Weng Keong Calvin) writes:

> What happens to the original code?

Nothing.  The idea with having multiple processes is just that a given 
"team" (side) can own more than one instruction pointer.  Just as in 
single-process Corewar each team has one instruction pointer, and 
these instruction pointers execute "simultaneously," with multiple- 
process Corewar each team can have more than one instruction pointer.  
New instruction pointers are created with the SPL instruction, and 
when an process executes a bad instruction, that process dies, but the 
other processes continue operating normally.  When there are no 
processes left, that team loses.


Erik Max Francis, &tSftDotIotE // uuwest!alcyone!max, max@alcyone.darkside.com
San Jose, CA, USA // 37 20 07 N  121 53 38 W // GIGO, Omega, Psi // the 4th R!
H.3`S,3,P,3$S,#$Q,C`Q,3,P,3$S,#$Q,3`Q,3,P,C$Q,#(Q.#`-"C`- // 1love // folasade
_Omnia quia sunt, lumina sunt._ // mc2? oo?  Nah. // http://www.spies.com/max/



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <4448n1$8df@geraldo.cc.utexas.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg>
newsgroups: rec.games.corewar

>: A process is like a seperate program, launched by the warrior.
>: Basically, Corewars programs have the ability to do multitasking.
>: One program may launch other programs, which will be excecuted on
>: a time-share basis.
>
>What happens to the original code?
>
>--
>Calvin Loh

Nothing, the code of the program stays exactly the same.

In fact, in order for the program to launch another process,
it must take care to launch the process to a core location
which contains excecutable code.

-John





From: John <jwilkinson@mail.utexas.edu>
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <4448jv$8df@geraldo.cc.utexas.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <441nga$1b2@allnews.infi.net>
newsgroups: rec.games.corewar

>>>>understand. What is meant by process?
>>
>>>A process is an instance of a program in execution.
>>
>>What a vague answer! :/
>>
>Accurate though.
>
>(jeez, there's a critic on every corner!)

Oh come on, I wouldn't want to scare off newbies by giving them
vague answers that they couldn't possibly decipher...

The more on the hill, the merrier. :)

-John







From: Derek Ross <rossd@arbroath.win-uk.net>
Subject: Multiple postings
Date: 1995/09/24
Message-ID: <444e3q$40a@gwen.ibmpcug.co.uk>#1/1
newsgroups: rec.games.corewar

Thank you Netscape.  Why do simple things never run smoothly :(

Aaargh 

Derek




From: Derek Ross <rossd@arbroath.win-uk.net>
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <444dor$40a@gwen.ibmpcug.co.uk>#1/1
references: <43qjs2$sib@nuscc.nus.sg>
newsgroups: rec.games.corewar

>I have just gone through the first 3 warriors in Steve Morrell's
>corewar tutorial and came across a term I do not really 
>understand. What is meant by process?
>Is a process a line of instruction?
>Or is it the sequence of instructions making up the warrior?
>TIA.
>
>--
>Calvin Loh
>

What is meant by process ? Hmm... let's think about sex...

A sex manual may describe "how to do it" but it's not the same as 
"actually doing it". The sex manual is a list of instructions or 
"program". The sex is a sequence of actions or "process". 

Similarly for computers a program describes "how to do it".  When the 
computer "actually does it" that's the process.  We talk about multiple 
processes when we have two or more computers carrying out a process each 
or one computer carrying out two or more processes simultaneously.  

In redcode the SPL instruction makes the MARS computer start executing a 
second list of instructions without stopping the execution of the first 
list.  So it appears that two different parts of the core war program 
are being executed at the same time.  In fact you can carry on using SPL 
to start new processes until the MARS computer reaches its limit of 
simultaneous execution.

Hope that this makes it clearer.

Cheers

Derek






From: Derek Ross <rossd@arbroath.win-uk.net>
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <444dl8$400@gwen.ibmpcug.co.uk>#1/1
references: <43qjs2$sib@nuscc.nus.sg>
newsgroups: rec.games.corewar

>I have just gone through the first 3 warriors in Steve Morrell's
>corewar tutorial and came across a term I do not really 
>understand. What is meant by process?
>Is a process a line of instruction?
>Or is it the sequence of instructions making up the warrior?
>TIA.
>
>--
>Calvin Loh
>

What is meant by process ? Hmm... let's think about sex...

A sex manual may describe "how to do it" but it's not the same as 
"actually doing it". The sex manual is a list of instructions or 
"program". The sex is a sequence of actions or "process". 

Similarly for computers a program describes "how to do it".  When the 
computer "actually does it" that's the process.  We talk about multiple 
processes when we have two or more computers carrying out a process each 
or one computer carrying out two or more processes simultaneously.  

In redcode the SPL instruction makes the MARS computer start executing a 
second list of instructions without stopping the execution of the first 
list.  So it appears that two different parts of the core war program 
are being executed at the same time.  In fact you can carry on using SPL 
to start new processes until the MARS computer reaches its limit of 
simultaneous execution.

Hope that this makes it clearer.

Cheers

Derek






From: Derek Ross <rossd@arbroath.win-uk.net>
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <444dh7$3u4@gwen.ibmpcug.co.uk>#1/1
references: <43qjs2$sib@nuscc.nus.sg>
newsgroups: rec.games.corewar

>I have just gone through the first 3 warriors in Steve Morrell's
>corewar tutorial and came across a term I do not really 
>understand. What is meant by process?
>Is a process a line of instruction?
>Or is it the sequence of instructions making up the warrior?
>TIA.
>
>--
>Calvin Loh
>

What is meant by process ? Hmm... let's think about sex...

A sex manual may describe "how to do it" but it's not the same as 
"actually doing it". The sex manual is a list of instructions or 
"program". The sex is a sequence of actions or "process". 

Similarly for computers a program describes "how to do it".  When the 
computer "actually does it" that's the process.  We talk about multiple 
processes when we have two or more computers carrying out a process each 
or one computer carrying out two or more processes simultaneously.  

In redcode the SPL instruction makes the MARS computer start executing a 
second list of instructions without stopping the execution of the first 
list.  So it appears that two different parts of the core war program 
are being executed at the same time.  In fact you can carry on using SPL 
to start new processes until the MARS computer reaches its limit of 
simultaneous execution.

Hope that this makes it clearer.

Cheers

Derek






From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: Re: What is a process?
Date: 1995/09/24
Message-ID: <442ouq$1ai@nuscc.nus.sg>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John (jwilkinson@mail.utexas.edu) wrote:

: A process is like a seperate program, launched by the warrior.
: Basically, Corewars programs have the ability to do multitasking.
: One program may launch other programs, which will be excecuted on
: a time-share basis.

What happens to the original code?

--
Calvin Loh



From: pp000738@interramp.com
Subject: Free classified advertising database on the Web
Date: 1995/09/25
Message-ID: <20524.pp000738@ipiws-001.interramp.com>#1/1
newsgroups: rec.games.corewar

OVERDRIVE

Overdrive has a Worldwide Classified Advertising
Database that is available on the Worldwide Web.
It works just like a newspaper classified advertising
section and is FREE for all to use. You can post an
advertisement of your own, update and delete it.

Sections include:
   Announcements and Services
   Training and Education
   Employment
   Financial
   Pets and Livestock
   Computer & Communications Merchandise
   General Merchandise
   Real Estate for Rent
   Real Estate for Sale
   Manufactured Homes
   Transportation

When you enter an Ad, you are prompted for a password.
When the record is created, you receive the record
id for your ad. Later, you may update or delete that
ad using the combination of your password and record
id.

Web travelers can search for Ads based on Classified
Section, price, manufacturer, model, etc.

OVERDRIVE can be found at http://www.ipworld.com




From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - Standard
Date: 1995/09/25
Message-ID: <199509250400.AAA15287@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html

Current Status of the StormKing Industries Standard KotH CoreWar Hill :

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  31/ 28/ 40          Yop La Boum v2.1         P.E.M & E.C.     134       1
 2  22/ 12/ 65                 Cannonade              P.Kline     133     101
 3  25/ 19/ 56   CAPS KEY IS STUCK AGAIN       Steven Morrell     132      67
 4  34/ 37/ 29              Giskard v0.5           Ken Mitton     132      39
 5  27/ 23/ 50    Der Zweiter Blitzkrieg      Mike Nonemacher     131      96
 6  36/ 41/ 24               Christopher       Steven Morrell     130      66
 7  33/ 36/ 31              Keystone t21              P.Kline     129      88
 8  22/ 17/ 61              Blue Funk 88       Steven Morrell     128      65
 9  36/ 43/ 21              Request v2.0     Brant D. Thomsen     128      37
10  23/ 20/ 57                     Hydra      Stephen Linhart     126     175
11  35/ 44/ 21       Beholder's Eye V1.7         W. Mintardjo     126     145
12  24/ 22/ 54                      Test       Wayne Sheppard     126      90
13  37/ 48/ 15                 Iron Gate       Wayne Sheppard     126     195
14  20/ 15/ 65                     Peace            Mr. Jones     126      75
15  37/ 50/ 13                  bigproba        nandor sieben     125      89
16  20/ 16/ 64                      ttti        nandor sieben     124      51
17  32/ 39/ 29             Miss Careless           Derek Ross     124      11
18  17/ 10/ 73         Imps! Imps! Imps!       Steven Morrell     124     112
19  20/ 17/ 62           jmp/add crasher         Randy Graham     124      25
20  15/ 11/ 74              Evol Imp v5a            Wilkinson     119      12

21  18/ 18/ 65              WhirlWind-88         Randy Graham     117       2



From: pk6811s@acad.drake.edu
Subject: Re: What is a process?
Date: 1995/09/25
Message-ID: <1995Sep25.084133@acad.drake.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg> <4448n1$8df@geraldo.cc.utexas.edu> <4456md$f7v@nuscc.nus.sg>
newsgroups: rec.games.corewar

(Loh Weng Keong Calvin) writes:
>...
> What happens if an enemy warrior trashes the code? Or it trashes its own
> code?

Now that would be unfortunate.  To be avoided at all cost :-)

Think of core as a long list of instructions, all of which can be read from
or written to.  Some of these can be legally executed, some are 'illegal'.
Executing an illegal instruction causes your process to terminate.

To start the battle, your and your opponents' programs are loaded somewhere
in core, and begin taking turns executing - a 'cycle' is one instruction
exectuted for every player.

Every player is given one process to start.  If your process dies by
executing an illegal instruction - you lose the battle.  However, one
of the strategies seen is to make copies of your program and SPL to them.
SPL (split) tells the interpreter to start a second process for you 'over
there' in core.  All your processes are in a list called the 'process queue'.
And every time you get a turn exactly one of them executes.  The interpreter
works its way through your process queue so that all your processes get
executed, but they have to share your turns.  You don't get more time than
anyone else, but you can scatter yourself around core, and since you don't
lose until _all_ your processes die you are harder to kill.

There are other strategies to survival, such as being very small or very
fast, but replication can be a good one.

For analogy think of a cat having 'nine lives'.  If your program has nine
processes it won't lose until all of them are killed.

Paul Kline
pk6811s@acad.drake.edu



From: ruhl@phoebe.cair.du.edu (Robert A. Uhl)
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/25
Message-ID: <446v1r$6p0@hermes.cair.du.edu>#1/1
references: <mreddy-2509951116240001@mreddy.comp.glam.ac.uk>
newsgroups: comp.ai.alife,rec.games.corewar

In article <mreddy-2509951116240001@mreddy.comp.glam.ac.uk>,
Mike Reddy <mreddy@glam.ac.uk> wrote:
>
>Sorry, but we cannot distribute the apps as Simon only had a student
>licence for the software. The sources and project files are available
>from:
>
>ftp://ftp.comp.glam.ac.uk/pub/staff/mreddy/

  Actually, you are allowed to distribute the program; the studetn
license allows shareware, so PD distribution shouldn't be any problem
whatsoever.  Also, the above address is invalid.  What is the proper
address?

-- 
+------------------------------------------------------------------------+
| Bob Uhl | Spectre                  | `En touto nika' +                 |
| U of D  | Baron Robert von Raetzin | http://mercury.cair.du.edu/~ruhl/ |
+------------------------------------------------------------------------+



From: bremermr@elsun269.cc.purdue.edu (Myer R. Bremer)
Subject: Agony II
Date: 1995/09/25
Message-ID: <446t9k$7gt@mozo.cc.purdue.edu>#1/1
newsgroups: rec.games.corewar

:I think Agony does so well because the CMP command is the fastest
:way to scan core.  While stones are fast,(1C for each attack) the CMP scanner
:does it's scanning at 2C.  (the CMP instruction scans TWO  locations
:in core at a time.)

:By the way, just how far ahead does Agony boot itself? (ahead meaning
:positive or negative locations in core?)

:John Lewis

I don't think Stefan is about to reveal to us his exact boot
distance.  Nice try, though.

M R Bremer





From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: What IS Agony
Date: 1995/09/25
Message-ID: <446gdb$dvl@lastactionhero.rs.itd.umich.edu>#1/1
references: <9509230622.AA11953@idnsun.gpct.Vanderbilt.Edu.noname>
newsgroups: rec.games.corewar





Stefan Strack (stst@idnsun.gpct.Vanderbilt.Edu) wrote:
: Agony II is a CMP-scanner that bombs with a SPL carpet to stun, followed by a
: simple DAT core clear and imp-gate. It's similar to Rave, but really just a
: version of Agony 3.1 (posted > 2 yrs ago) that boots off a large decoy which
: is optimized to to avoid triggering Agony's own scan until 6000 cycles into
: the round. It does well against one-shot scanners like Withershins/Porch
: because it boots just ahead of the decoy and finishes them off while they're
: busy core clearing. I don't really know why Agony does so well right now.
: I'll post source when Agony reaches age 1000 or is bumped off, whichever
: comes first.

: -Stefan


I think Agony does so well because the CMP command is the fastest
way to scan core.  While stones are fast,(1C for each attack) the CMP scanner
does it's scanning at 2C.  (the CMP instruction scans TWO  locations
in core at a time.)

By the way, just how far ahead does Agony boot itself? (ahead meaning
positive or negative locations in core?)

John Lewis



From: mreddy@glam.ac.uk (Mike Reddy)
Subject: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/25
Message-ID: <mreddy-2509951116240001@mreddy.comp.glam.ac.uk>#1/1
newsgroups: comp.ai.alife,rec.games.corewar

After tracking down my old student, I can now make the source code (for
Code Warrior) available for RedGen. RedGen was a final year project to
develop an Alife simulation using Corewar as the environment and the
fitness function. The student, Simon Jones has now graduated and the
dissertation was entitled "Evolution of competitive Core War warriors
using Crossover". The results were partial and not statistically analysed,
but I thought that people might like to reproduce the work. The interface
is bitty and for best performance you should use a PowerPC, though source
works for 68K Macs as well.

Sorry, but we cannot distribute the apps as Simon only had a student
licence for the software. The sources and project files are available
from:

ftp://ftp.comp.glam.ac.uk/pub/staff/mreddy/

and three formats are available

redgen.sea, redgen.sea.hqx and redgen.sea.hqx.gz

You can log on anonymously. Hope you find it interesting.


Yours Mike Reddy
--
Email: mreddy@glam.ac.uk CU-Seeme: 193.63.130.40 (On Request)
Web: http://www.comp.glam.ac.uk/pages/staff/mreddy/mreddy.html
Snail: J228, Dept. of Computer Studies, University of Glamorgan, Pontypridd, Mid Glamorgan. CF37 1DL Wales, UK. +44 1443 482 240 Fax: +44 1443 482 715



From: max@alcyone.darkside.com (Erik Max Francis)
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/25
Message-ID: <F5T1BD4w200w@alcyone.darkside.com>#1/1
references: <446v1r$6p0@hermes.cair.du.edu>
newsgroups: comp.ai.alife,rec.games.corewar

ruhl@phoebe.cair.du.edu (Robert A. Uhl) writes:

> In article <mreddy-2509951116240001@mreddy.comp.glam.ac.uk>,
> Mike Reddy <mreddy@glam.ac.uk> wrote:
> >
> >Sorry, but we cannot distribute the apps as Simon only had a student
> >licence for the software. The sources and project files are available
> >from:
> >
> >ftp://ftp.comp.glam.ac.uk/pub/staff/mreddy/
> 
>   Actually, you are allowed to distribute the program; the studetn
> license allows shareware, so PD distribution shouldn't be any problem
> whatsoever.  Also, the above address is invalid.  What is the proper
> address?

Ahem!  There's a world of difference between shareware (or even 
freeware!) and public domain.


Erik Max Francis, &tSftDotIotE // uuwest!alcyone!max, max@alcyone.darkside.com
San Jose, CA, USA // 37 20 07 N  121 53 38 W // GIGO, Omega, Psi // the 4th R!
H.3`S,3,P,3$S,#$Q,C`Q,3,P,3$S,#$Q,3`Q,3,P,C$Q,#(Q.#`-"C`- // 1love // folasade
_Omnia quia sunt, lumina sunt._ // mc2? oo?  Nah. // http://www.spies.com/max/



From: KOTH Tourney Server <koth>
Subject: SKI-ICWS: Status - MultiWarrior 94
Date: 1995/09/25
Message-ID: <199509250400.AAA05824@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

			Weekly Status 

See up to the second scores at http://www.stormking.com/~koth
	Check it out for the latest in Core War information
*FAQ* is at http://www.stormking.com/~koth/corewar-faq.html

Current Status of the StormKing Industries Multiwarrior 94 CoreWar Hill:

 #                      Name               Author   Score     Age
 1                Son of Imp       Steven Morrell    2111      19
 2                  Die Hard              P.Kline    2111       6
 3            75% Cotton v3b            Wilkinson    2111       3
 4                  Paperone          Beppe Bezzi    2084      41
 5                60% Cotton            Wilkinson    2083       2
 6            90% Cotton v5c            Wilkinson    2075       1
 7              Silkworm 1.1          Beppe Bezzi    2074      16
 8           TimeScape (1.1)       J. Pohjalainen    2062      64
 9               Cotton v C1            Wilkinson    2038       5
10                    test 2              P.Kline    2019      23

11                 ShadowImp     Robert J. Street    1657       0



From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: Re: What is a process?
Date: 1995/09/26
Message-ID: <4481lo$73i@nuscc.nus.sg>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg> <4448n1$8df@geraldo.cc.utexas.edu> <4456md$f7v@nuscc.nus.sg> <1995Sep25.084133@acad.drake.edu>
newsgroups: rec.games.corewar

pk6811s@acad.drake.edu wrote:

: Think of core as a long list of instructions, all of which can be read from
: or written to.  Some of these can be legally executed, some are 'illegal'.
: Executing an illegal instruction causes your process to terminate.


: Every player is given one process to start.  If your process dies by
: executing an illegal instruction - you lose the battle.  However, one
: of the strategies seen is to make copies of your program and SPL to them.
: SPL (split) tells the interpreter to start a second process for you 'over
: there' in core.  All your processes are in a list called the 'process queue'.
: And every time you get a turn exactly one of them executes.  The interpreter
: works its way through your process queue so that all your processes get
: executed, but they have to share your turns.  You don't get more time than
: anyone else, but you can scatter yourself around core, and since you don't
: lose until _all_ your processes die you are harder to kill.

So I should use 'mov' to copy my code elsewhere, then use 'spl' to that
address. Then what does 'jmp' do? I've seen code where 'jmp' is at the
tail of the code and goes to the head of the code. Is this an endless
loop where the process is always executing in the same address space?

Can someone trace the movement in space of an imp, for example? Steve
Morrell's tutorial seems to trace only the instruction pointer's movement.

Thanks very much for all the replies so far.
--
Calvin Loh



From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: Re: What is a process?
Date: 1995/09/26
Message-ID: <44817t$73i@nuscc.nus.sg>#1/1
references: <442ouq$1ai@nuscc.nus.sg> <w9gyBD1w200w@alcyone.darkside.com>
newsgroups: rec.games.corewar

Erik Max Francis (max@alcyone.darkside.com) wrote:
: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin) writes:

: > What happens to the original code?

: Nothing.  The idea with having multiple processes is just that a given 
: "team" (side) can own more than one instruction pointer.  Just as in 
: single-process Corewar each team has one instruction pointer, and 
: these instruction pointers execute "simultaneously," with multiple- 
: process Corewar each team can have more than one instruction pointer.  
: New instruction pointers are created with the SPL instruction, and 
: when an process executes a bad instruction, that process dies, but the 
: other processes continue operating normally.  When there are no 
: processes left, that team loses.

Does that mean that I win if I can junk all the enemy code floating around?
TIA.
--
Calvin Loh



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Minor PIZZA change
Date: 1995/09/26
Message-ID: <44980j$luj@geraldo.cc.utexas.edu>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu> <HAN.95Sep26100340@uvea.imag.fr>
newsgroups: rec.games.corewar

>When you "ask" a question like that and don't get any response, it's
>very likely either:
>
>1. everybody feels like you: ambivolent, or
>2. people are thinking about it.
>(or 3. Scott is busy right now :-) )
>
>P-space is still very young. I think new experiments would be
>welcome. 

Thank you for acknoweledging my existence.  That's a step in the right
direction...! :)

Well, then think everybody.  Puzzle 'till your puzzlers are sore.

-John








From: stst@idnsun.gpct.Vanderbilt.Edu (Stefan Strack)
Subject: Re:  Minor PIZZA change
Date: 1995/09/26
Message-ID: <9509261818.AA13252@idnsun.gpct.Vanderbilt.Edu.noname>#1/1
newsgroups: rec.games.corewar

Jon Wilkinson (jwilkinson@mail.utexas.edu) wrote:
>Well, I requested long ago that TUC implemenet of 5x100 battle setup...

As Tuc hinted to earlier, rounds will go from 100 to 250 on all stormking
and pizza hills by Oct. 1, except for the -94m hill which will be raised 
from 200 to 500 rounds. This is a compromise between decreasing the score
variability and keeping the turn-around time acceptable.

Thomas Davies, maintainer of the pizza hills, will clone the -94 hill so 
that all submission go to a hill with a 5*50 rounds setup as well, but only
if there's sufficient interest. So, if you would like to test the effect of
breaking up battles, *email* him (sd@ecst.csuchico.edu) or me.

-Stefan



From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: Minor PIZZA change
Date: 1995/09/26
Message-ID: <199509261452.PAA00711@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

>
>Well, I requested long ago that TUC implemenet of 5x100 battle setup...
>
>He said he wanted to see what the newgroup thought.
>
>I posted the idea.
>
>I got very little response...
>
I see no real need to change from 100 battles to 5*100 or 1*500, but I have
nothing against; apart that testing on my machine will take 5 times more,
and five times half an hour for my 12 battles times 100 rounds benchmark is
_a_lot_ more.

I'll go on testing with but 100 battles as I do now.


>
>-John





From: tuc@news.stormking.com (Scott J. Ellentuch)
Subject: Re: Minor PIZZA change
Date: 1995/09/26
Message-ID: <448p8c$bt9@valhalla.stormking.com>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John (jwilkinson@mail.utexas.edu) wrote:
: Well, I requested long ago that TUC implemenet of 5x100 battle setup...


	Um, John, not to pick a nit.... But I run Stormking.Com ....
And as for upping the battles, (Stefan and Thos will probably kill me but)
WATCH THIS SPACE FOR AN IMPORTANT ANNOUNCEMENT SOON REGARDING THIS.....
(Perhaps, lets say, before October 1)

				Tuc
--
* ---   ---  |Scott J. Ellentuch, Pres  | The Telecom Security Group       *
*  | |  |    |Comm. and Comp. Security  | (Storm King family of Companies) *
*  | |__|__  |Consultants and Engineers | P.O. Box 69, Newburgh, NY 12551  *



From: han@imag.fr (Jay "Thierry" Han)
Subject: Re: Minor PIZZA change
Date: 1995/09/26
Message-ID: <HAN.95Sep26100340@uvea.imag.fr>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

In article <447mrh$n5e@geraldo.cc.utexas.edu> John <jwilkinson@mail.utexas.edu> writes:

> Well, I requested long ago that TUC implemenet of 5x100 battle setup...
> He said he wanted to see what the newgroup thought.
> I posted the idea.
> I got very little response...
> Is that good or bad?  Could you guys at least vote or something?  Just
> reply yes or no, or mail TUC or something...  I dunno.
> If eveybody thinks this is a bad idea, please tell ME, because I can't
> think of any reason not to do this...

When you "ask" a question like that and don't get any response, it's
very likely either:

1. everybody feels like you: ambivolent, or
2. people are thinking about it.
(or 3. Scott is busy right now :-) )

P-space is still very young. I think new experiments would be
welcome. 

-=< Jay "Thierry" Han >=-                        Jay.Han@imag.fr 
Bull-IMAG/Systemes. 2, av. Vignate. ZI Mayencin II, 38610 Gieres.
                          Tel: +33 76.63.48.50. Fax: 76.54.76.15.
Perso: 61, rue Thiers. 38000 Grenoble.               76.46.11.26.



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: one-pass scanners
Date: 1995/09/26
Message-ID: <449gnv$roo@geraldo.cc.utexas.edu>#1/1
references: <1995Sep26.112750@acad.drake.edu>
newsgroups: rec.games.corewar

>One-pass scanning has been a cyclical strategy.  Someone puts one up and
>does very well.  Others copy.  Then people add boot-and-decoy and drive
>the one-passers off.  Then people strip out the b&d to save startup time.
>Then one-passers appear again.  The first one-passers were b-scanners,
>then cmp-scanners, now mov/cmp-scanners, each reincarnation being faster
>than the last, but all with the same Achilles' heel.
>-- 
>Paul Kline
>pk6811s@acad.drake.edu

Indeed, I've noticed similar oddness...  The thing is, I'm estimating
about 4-7 -successful- challenges to the Hill per day.  This makes for
very rapid evolution... which does have the interesting side effect
of having very little point spread between the top and bottom warrior.
At the moment, the King is less than 20 points above the last place
warrior.  That's only about 15% differnce...

Perhaps, instead of instituting the 5x100 system, we should add more
spaces to the hill?  This would slow down the rapid change of the
hill, and allow good warriors to stay on longer?

Something to think about.  At least, -I'll- be thinking about it. :)

-John




From: stst@vuse.vanderbilt.edu (Stefan Strack)
Subject: Core War Frequently Asked Questions (rec.games.corewar FAQ)
Date: 1995/09/26
Message-ID: <games/corewar-faq_812133890@rtfm.mit.edu>
newsgroups: rec.games.corewar,rec.answers,news.answers

Archive-name: games/corewar-faq
Last-Modified: 95/08/06
Version: 3.4

   These are the Frequently Asked Questions (and answers) from the Usenet
   newsgroup rec.games.corewar. A plain text version of this document is
   posted every two weeks. The hypertext version is available as
   <URL:http://www.stormking.com/~koth/corewar-faq.html>
     _________________________________________________________________
   
Table of Contents

    1. What is Core War
    2. Is it Core War or Core Wars?
    3. Where can I find more information about Core War?
    4. Core War has changed since Dewdney's articles. Where do I get a
       copy of the current instruction set?
    5. What is ICWS'94? Which simulators support ICWS'94?
    6. What is the ICWS?
    7. What is TCWN?
    8. How do I join?
    9. What is the EBS?
   10. Where are the Core War archives?
   11. Where can I find a Core War system for ...?
   12. I do not have FTP. How do I get all this great stuff?
   13. I do not have access to Usenet. How do I post and receive news?
   14. Are there any Core War related WWW sites?
   15. When is the next tournament?
   16. What is KotH? How do I enter?
   17. Is it DAT 0, 0 or DAT #0, #0? How do I compare to core?
   18. How does SLT (Skip if Less Than) work?
   19. What is the difference between in-register and in-memory
       evaluation?
   20. What does (expression or term of your choice) mean?
   21. Other questions?
       
   
     _________________________________________________________________
   
What is Core War?

   Core War is a game played by two or more programs (and vicariously by
   their authors) written in an assembly language called Redcode and run
   in a virtual computer called MARS (for Memory Array Redcode
   Simulator). The object of the game is to cause all processes of the
   opposing program to terminate, leaving your program in sole posession
   of the machine.
   
   There are Core War systems available for most computer platforms.
   Redcode has been standardized by the ICWS, and is therefore
   transportable between all standard Core War systems.
   
   [ToC]
     _________________________________________________________________
   
Is it "Core War" or "Core Wars"?

   Both terms are used. Early references were to Core War. Later
   references seem to use Core Wars. I prefer "Core War" to refer to the
   game in general, "core wars" to refer to more than one specific
   battle.
   
   [ToC]
     _________________________________________________________________
   
Where can I find more information about Core War?

   Core War was first described in the Core War Guidelines of March, 1984
   by D. G. Jones and A. K. Dewdney of the Department of Computer Science
   at The University of Western Ontario (Canada). Dewdney wrote several
   "Computer Recreations" articles in Scientific American which discussed
   Core War, starting with the May 1984 article. Those articles are
   contained in two anthologies:
   
   Author: Dewdney, A. K.
   Title: The Armchair Universe: An Exploration of Computer Worlds
   Published: New York: W. H. Freeman (c) 1988
   ISBN: 0-7167-1939-8
   Library of Congress Call Number: QA76.6 .D517 1988
   
   
   Author: Dewdney, A. K.
   Title: The Magic Machine: A Handbook of Computer Sorcery
   Published: New York: W. H. Freeman (c) 1990
   ISBN: 0-7167-2125-2 (Hardcover), 0-7167-2144-9 (Paperback)
   Library of Congress Call Number: QA76.6 .D5173 1990
   
   
   A.K. Dewdney's articles are still the most readable introduction to
   Core War, even though the Redcode dialect described in there is no
   longer current.
   
   [ToC]
     _________________________________________________________________
   
Core War has changed since Dewdney's articles. Where do I get a copy of the
current instruction set?

   A draft of the official standard (ICWS'88) is available as
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/standards/re
   dcode-icws-88> . This document is formatted awkwardly and contains
   ambiguous statements. For a more approachable intro to Redcode, take a
   look at Mark Durham's tutorial,
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/tutorial.1>
   and
   <URL:http://ftp.csua.berkeley.edu/pub/corewar/documents/tutorial.2>.
   
   
   Steven Morrell (morrell@math.utah.edu) is preparing a more practically
   oriented Redcode tutorial that discusses different warrior classes
   with lots of example code. Mail him for a preliminary version. Michael
   Constant (mconst@csua.berkeley.edu) is reportedly working on a
   beginner's introduction.
   
   [ToC]
     _________________________________________________________________
   
What is ICWS'94? Which simulators support ICWS'94?

   There is an ongoing discussion about future enhancements to the
   Redcode language. A proposed new standard, dubbed ICWS'94, is
   currently being evaluated. A major change is the addition of
   "instruction modifiers" that allow instructions to modify A-field,
   B-field or both. Also new is a post-increment indirect addressing mode
   and unrestricted opcode and addressing mode combination ("no illegal
   instructions"). ICWS'94 is backwards compatible; i.e. ICWS'88 warriors
   will run correctly on an ICWS'94 system. Take a look at the ICWS'94
   draft
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/documents/icws94.0202.Z
   > for more information. You can try out the new standard by
   submitting warriors to the '94 hills of the KotH servers. Two corewar
   systems currently support ICWS'94, pMARS (many platforms) and Redcoder
   (Mac), both available at ftp.csua.berkeley.edu.
   
   [ToC]
     _________________________________________________________________
   
What is the ICWS?

   About one year after Core War first appeared in Sci-Am, the
   "International Core War Society" (ICWS) was established. Since that
   time, the ICWS has been responsible for the creation and maintenance
   of Core War standards and the running of Core War tournaments. There
   have been six annual tournaments and two standards (ICWS'86 and
   ICWS'88).
   
   [ToC]
     _________________________________________________________________
   
What is TCWN?

   Since March of 1987, "The Core War Newsletter" (TCWN) has been the
   official newsletter of the ICWS. It is published quarterly and recent
   issues are also available as Encapsulated PostScript files.
   
   [ToC]
     _________________________________________________________________
   
How do I join?

   For more information about joining the ICWS (which includes a
   subscription to TCWN), or to contribute an article, review, cartoon,
   letter, joke, rumor, etc. to TCWN, please contact:
   

   Jon Newman
   13824 NE 87th Street
   Redmond, WA 98052-1959
   email: jonn@microsoft.com  (Note: Microsoft has NO affiliation with
                                     Core War.  Jon Newman just happens
                                     to work there, and we want to keep
                                     it that way!)

   Current annual dues are $15.00 in US currency.
   
   [ToC]
     _________________________________________________________________
   
What is the EBS?

   The Electronic Branch Section (EBS) of the ICWS is a group of Core War
   enthusiasts with access to electronic mail. There are no fees
   associated with being a member of the EBS, and members do reap some of
   the benefits of full ICWS membership without the expense. For
   instance, the ten best warriors submitted to the EBS tournament are
   entered into the annual ICWS tournament. All EBS business is conducted
   in the rec.games.corewar newsgroup.
   
   The current goal of the EBS is to be at the forefront of Core War by
   writing and implementing new standards and test suites.
   
   [ToC]
     _________________________________________________________________
   
Where are the Core War archives?

   Many documents such as the guidelines and the ICWS standards along
   with previous tournament Redcode entries and complete Core War systems
   are available via anonymous ftp from
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/systems>
   (128.32.149.19) in the /pub/corewar directories. Also, most of past
   rec.games.corewar postings (including Redcode source listings) are
   archived there. Jon Blow (blojo@csua.berkeley.edu) is the archive
   administrator. When uploading to /pub/corewar/incoming, ask Jon to
   move your upload to the appropriate directory and announce it on the
   net.
   
   Much of what is available on soda is also available on the German
   archive at iraun1.ira.uka.de (129.13.10.90) in the /pub/x11/corewars
   directory.
   
   The plain text version of this FAQ is automatically archived by
   news.answers.
   
   [ToC]
     _________________________________________________________________
   
Where can I find a Core War system for . . . ? 

   Core War systems are available via anonymous ftp from
   ftp.csua.berkeley.edu in the /pub/corewar/systems directory.
   Currently, there are UNIX, IBM PC-compatible, Macintosh, and Amiga
   Core War systems available there. It is a good idea to check
   <URL:ftp://ftp.csua.berkeley.edu/pub/corewar/incoming> for program
   updates first.
   
   CAUTION! There are many, many Core War systems available which are NOT
   ICWS'88 (or even ICWS'86) compatible available at various archive
   sites other than ftp.csua.berkeley.edu. Generally, the older the
   program - the less likely it will be ICWS compatible.
   
   Reviews of Core War systems would be greatly appreciated in the
   newsgroup and in the newsletter.
   
   Below is a not necessarily complete or up-to-date list of what's
   available at ftp.csua.berkeley.edu:
   

    MADgic41.lzh    - corewar for the Amiga, v4.1
    MAD4041.lzh     - older version?
    MAD50B.lha      - corewar for the Amiga, beta version 5.0
    Redcoder-21.hqx - corewar for the Mac, supports ICWS'88 and '94 (without
                      extensions)
    core-11.hqx     - corewar for the Mac
    core-wars-simulator.hqx - same as core-11.hqx?
    corewar_unix_x11.tar.Z - corewar for UNIX/X-windows, ICWS'86 but
                      not ICWS'88 compatible
    koth31.tar.Z    - corewar for UNIX/X-windows. This program ran the
                      former KotH server at intel.com
    koth.shar.Z     - older version
    kothpc.zip      - port of older version of KotH to the PC
    deluxe20c.tar.Z - corewar for UNIX (broken X-windows or curses) and PC
    mars.tar.Z      - corewar for UNIX, likely not ICWS'88 compatible
    icons.zip       - corewar icons for MS-Windows
    macrored.zip    - a redcode macro-preprocessor (PC)
    c88v49.zip      - PC corewar, textmode display
    mars88.zip      - PC corewar, graphics mode display
    corwp302.zip    - PC corewar, textmode display, slowish
    mercury2.zip    - PC corewar written in assembly, fast!
    mtourn11.zip    - tournament scheduler for mercury (req. 4DOS)
    pmars08s.zip    - portable system, ICWS'88 and '94, runs on UNIX,
                      PC, Mac, Amiga. C source archive
    pmars08s.tar.Z  - same as above
    pmars08.zip    - PC executables with graphics display, req 386+
    macpmars02.sit.hqx - pMARS executable for Mac (port of version 0.2)
                      buggy, no display
    MacpMARS.10.cpt.hqx - port of v0.6 for the Mac, with display and
                      debugger
    MacpMARS.10s.cpt.hqx - C source (MPW, ThinkC) for Mac frontend
    ApMARS03.lha    - pMARS executable for Amiga (port of version 0.3.1)
    wincor11.zip    - MS-Windows system, shareware ($15)

   
   
   [ToC]
     _________________________________________________________________
   
I do not have FTP. How do I get all this great stuff? 

   There is an ftp email server at ftpmail@decwrl.dec.com. Send email
   with a subject and body text of "help" (without the quotes) for more
   information on its usage. If you don't have access to the net at all,
   send me a 3.5 '' diskette in a self-addressed disk mailer with postage
   and I will mail it back with an image of the Core War archives in PC
   format. My address is at the end of this post.
   
   [ToC]
     _________________________________________________________________
   
I do not have access to Usenet. How do I post and receive news?

   To receive rec.games.corewar articles by email, join the COREWAR-L
   list run on the Stormking.Com list processor. To join, send the
   message

        SUB COREWAR-L FirstName LastName

   to listproc@stormking.com. You can send mail to
   corewar-l@stormking.com to post even if you are not a member of the
   list. Responsible for the listserver is Scott J. Ellentuch
   (tuc@stormking.com).
   
   Another server that allows you to post (but not receive) articles is
   available. Email your post to rec-games-corewar@cs.utexas.edu and it
   will be automatically posted for you.
   
   [ToC]
     _________________________________________________________________
   
Are there any Core War related WWW sites?

   You bet. Each of the two KotH sites sport a world-wide web server.
   Stormking's Core War page is <URL:http://www.stormking.com/~koth>;
   pizza's is <URL:http://www.ecst.csuchico.edu/~pizza/koth> . A third
   WWW site is in Koeln, Germany:
   <URL:http://www.khm.uni-koeln.de/~wbi/Index/Smoking/CWBH>. Last but
   not least, Stephen Beitzel's "Unofficial Core War Page" is
   <URL:http://www2.eworld.com/aos/people/beitzel/core.html>.
   All site are in varying stages of construction, so it would be futile
   to list here what they have to offer.
   
   [ToC]
     _________________________________________________________________
   
When is the next tournament?

   The ICWS holds an annual tournament. Traditionally, the deadline for
   entering is the 15th of December. The EBS usually holds a preliminary
   tournament around the 15th of November and sends the top finishers on
   to the ICWS tournament.
   
   Informal double-elimination and other types of tournaments are held
   frequently among readers of the newsgroup; watch there for
   announcements or contact me.
   
   [ToC]
     _________________________________________________________________
   
What is KotH? How do I enter?

   King Of The Hill (KotH) is an ongoing Core War tournament available to
   anyone with email. You enter by submitting via email a Redcode program
   (warrior) with special comment lines. You will receive a reply
   indicating how well your program did against the current top programs
   "on the hill".
   
   There are two styles of KotH tournaments, "classical" and
   "multi-warrior". The "classical" KotH is a one-on-one tournament, that
   is your warrior will play 100 battles against each of the 20 other
   programs currently on the Hill. You receive 3 points for each win and
   1 point for each tie. (The existing programs do not replay each other,
   but their previous battles are recalled.) All scores are updated to
   reflect your battles and all 21 programs are ranked from high to low.
   If you are number 21 you are pushed off the Hill, if you are higher
   than 21 someone else is pushed off.
   
   In "multi-warrior" KotH, all warriors on the hill fight each other at
   the same time. Score calculation is a bit more complex than for the
   one-on-one tournament. Briefly, points are awarded based on how many
   warriors survive until the end of a round. A warrior that survives by
   itself gets more points than a warrior that survives together with
   other warriors. Points are calculated from the formula (W*W-1)/S,
   where W is the total number of warriors and S the number of surviving
   warriors. The pMARS documentation has more information on
   multi-warrior scoring.
   
   The idea for an email-based Core War server came from David Lee. The
   original KotH was developed and run by William Shubert at Intel
   starting in 1991, and discontinued after almost three years of
   service. Currently, KotHs based on Bill's UNIX scripts but offering a
   wider variety of hills are are running at two sites:
   "koth@stormking.com" is maintained by Scott J. Ellentuch
   (tuc@stormking.com) and "pizza@ecst.csuchico.edu" by Thomas H. Davies
   (sd@ecst.csuchico.edu). Up until May '95, the two sites provided
   overlapping services, i.e. the some of the hill types were offered by
   both "pizza" and "stormking". To conserve resources, the different
   hill types are now divided up among the sites. The way you submit
   warriors to both KotHs is pretty much the same. Therefore, the entry
   rules described below apply to both "pizza" and "stormking" unless
   otherwise noted.
   
   Entry rules for King of the Hill Corewar:
   
   1) Write a corewar program. KotH is fully ICWS '88 compatible, EXCEPT
   that a comma (",") is required between two arguments.
   
   2) Put a line starting with ";redcode" (or ";redcode-94, etc., see
   below) at the top of your program. This MUST be the first line.
   Anything before it will be lost. If you wish to receive mail on every
   new entrant, use ";redcode verbose". Otherwise you will only receive
   mail if a challenger makes it onto the hill. Use ";redcode quiet" if
   you wish to receive mail only when you get shoved off the hill. (Also,
   see 5 below).
   
   Additionally, adding ";name " and ";author " will be helpful in the
   performance reports. Do NOT have a line beginning with ";address" in
   your code; this will confuse the mail daemon and you won't get mail
   back.
   
   In addition, it would be nice if you have lines beginning with
   ";strategy" that describe the algorithm you use.
   
   There are currently seven separate hills you can select by starting
   your program with ;redcode-b, ;redcode-94, ;redcode-94x, ;redcode,
   ;redcode-icws, ;redcode-94m or ;redcode-94xm. The former three run at
   "pizza", the latter four at "stormking". More information on these
   hills is listed below.
   
   3) Mail this file to koth@stormking.com or pizza@ecst.csuchico.edu.
   "Pizza" requires a subject of "koth" (use the -s flag on most
   mailers).
   
   4) Within a few minutes you should get mail back telling you whether
   your program assembled correctly or not. If it did assemble correctly,
   sit back and wait; if not, make the change required and re-submit.
   
   5) In an hour or so you should get more mail telling you how your
   program performed against the current top 20 (or 10) programs. If no
   news arrives during that time, don't worry; entries are put in a queue
   and run through the tournament one at a time. A backlog may develop.
   Be patient.
   
   If your program makes it onto the hill, you will get mail every time a
   new program makes it onto the hill. If this is too much mail, you can
   use ";redcode[-??] quiet" when you first mail in your program; then
   you will only get mail when you make it on the top 20 list or when you
   are knocked off. Using ";redcode[-??] verbose" will give you even more
   mail; here you get mail every time a new challenger arrives, even if
   they don't make it onto the top 20 list.
   
   Often programmers want to try out slight variations in their programs.
   If you already have a program named "foo V1.0" on the hill, adding the
   line ";kill foo" to a new program will automatically bump foo 1.0 off
   the hill. Just ";kill" will remove all of your programs when you
   submit the new one. The server kills programs by assigning an
   impossibly low score; it may therefore take another successful
   challenge before a killed program is actually removed from the hill.
   

SAMPLE ENTRY:
;redcode
;name Dwarf
;author A. K. Dewdney
;strategy Throw DAT bombs around memory, hitting every 4th memory cell.
;strategy This program was presented in the first Corewar article.
bomb  DAT   #0
dwarf ADD   #4,    bomb
      MOV   bomb, @bomb
      JMP   dwarf
      END   dwarf          ; Programs start at the first line unless
                           ; an "END start" pseudo-op appears to indicate
                           ; the first logical instruction.  Also, nothing
                           ; after the END instruction will be assembled.

Here are the Specs for the various hills:

ICWS'88 Standard Hill Specs: (Accessed with ";redcode", available at
"stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: ICWS '88

ICWS Annual Tournament Hill Specs: (Accessed with ";redcode-icws",
available at "stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8192 instructions
   max. processes: 8000 per program
         duration: After 100,000 cycles, a tie is declared.
max. entry length: 300
 minimum distance: 300
  instruction set: ICWS '88

ICWS'94 Draft Hill Specs: (Accessed with ";redcode-94", available at
"pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft

ICWS'94 Beginner's Hill Specs: (Accessed with ";redcode-b", available
at "pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft
         max. age: after 100 successful challenges, warriors are retired.

ICWS'94 Experimental (Big) Hill Specs: (Accessed with ";redcode-94x",
available at "pizza")
         hillsize: 20 warriors
           rounds: 100
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: extended ICWS '94 Draft

ICWS'94 Draft Multi-Warrior Hill Specs: (Accessed with ";redcode-94m",
available at "stormking")
         hillsize: 10 warriors
           rounds: 200
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: extended ICWS '94 Draft

ICWS'94 Experimental (Big) Multi-Warrior Hill Specs:
(Accessed with ";redcode-94xm", available at "stormking")
         hillsize: 20 warriors
           rounds: 100
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: extended ICWS '94 Draft

   If you just want to get a status report without actually challenging
   the hills, send email with ";status" as the message body (and don't
   forget "Subject: koth" for "pizza"). If you send mail to "pizza" with
   "Subject: koth help" you will receive instructions that may be more up
   to date than those contained in this document.
   
   At stormking, a message body with ";help" will return brief
   instructions. If you submit code containing a ";test" line, your
   warrior will be assembled but not actually pitted against the warriors
   on the hill.
   
   All hills run portable MARS (pMARS) version 0.8, a
   platform-independent corewar system available at
   ftp.csua.berkeley.edu.
   
   The '94 and '94x hills allow three experimental opcodes and addressing
   modes currently not covered in the ICWS'94 draft document:

        SEQ - Skip if EQual (synonym for CMP)
        SNE - Skip if Not Equal
        NOP - (No OPeration)

        *   - indirect using A-field as pointer
        {   - predecrement indirect using A-field
        }   - postincrement indirect using A-field

   
   
   [ToC]
     _________________________________________________________________
   
Is it DAT 0, 0 or DAT #0, #0? How do I compare to core? 

   Core is initialized to DAT 0, 0. This is an "illegal" instruction
   under ICWS'88 rules and strictly compliant assemblers (such as KotH or
   pmars -8) will not let you write a DAT 0, 0 instruction - only DAT #0,
   #0. So this begs the question, how to compare something to see if it
   is empty core. The answer is, most likely the instruction before your
   first instruction and the instruction after your last instruction are
   both DAT 0, 0. You can use them, or any other likely unmodified
   instructions, for comparison. Note that under ICWS'94, DAT 0, 0 is a
   legal instruction.
   
   [ToC]
     _________________________________________________________________
   
How does SLT (Skip if Less Than) work?

   SLT gives some people trouble because of the way modular arithmetic
   works. It is important to note that all negative numbers are converted
   to positive numbers before a battles begins. Example: (-1) becomes (M
   - 1) where M is the memory size.
   
   Once you realize that all numbers are treated as positive, it is clear
   what is meant by "less than". It should also be clear that no number
   is less than zero.
   
   [ToC]
     _________________________________________________________________
   
What is the difference between in-register and in-memory evaluation?

   These terms refer to the way instruction operands are evaluated. The
   '88 Redcode standard ICWS'88 is unclear about whether a simulator
   should "buffer" the result of A-operand evaluation before the
   B-operand is evaluated. Simulators that do buffer are said to use
   in-register evaluation, those that don't, in-memory evaluation.
   ICWS'94 clears this confusion by mandating in-register evaluation.
   Instructions that execute differently under these two forms of
   evaluation are MOV, ADD, SUB, MUL, DIV and MOD where the effective
   address of the A-operand is modified by evaluation of the B-operand.
   This is best illustrated by an example:

   L1  mov L2,<L2
   L2  dat #0,#1

   Under in-register evaluation, the 'L2' instruction is saved in a
   buffer before the 'L2' memory location is decremented by evaluation of
   the B-operand of 'L1'. The saved 'dat #0,#1' instruction is then
   written to 'L2', leaving it unchanged.
   
   Under in-memory evaluation, the 'L2' instruction is not buffered and
   thus decremented by evaluation of the B-operand. After execution of
   'L1', 'L2' changes to "dat #0,#0".
   
   [ToC]
     _________________________________________________________________
   
What does (expression or term of your choice) mean?

   Here is a selected glossary of terms. If you have a definition and/or
   term you wish to see here, please send it to me.
   
   (References to an X-like program mean that the term X is derived from
   the specific program X and has become a generic term).
   
   Binary launch
          one of several means to start an imp-spiral running. The
          fastest launch technique, but requires the most code. See also
          JMP/ADD Launch and Vector Launch.
          


    impsize   equ 2667
    example   spl 4               ; extend by adding spl 8, spl 16, etc.
              spl 2
              jmp imp+(0*impsize) ; jmp's execute in order
              jmp imp+(1*impsize)
              spl 2
              jmp imp+(2*impsize)
              jmp imp+(3*impsize)
    imp       mov 0,impsize       ; in '94 use -> mov.i #0,impsize

   Bootstrapping
          Strategy of copying the active portion of the program away from
          the initial location, leaving a decoy behind and making the
          relocated program as small as possible.
          
   B-Scanners
          Scanners which only recognize non-zero B-fields.
          

    example   add #10,scan
    scan      jmz example,10

   C
          Measure of speed, equal to one location per cycle. Speed of
          light.
          
   CMP-Scanner
          A Scanner which uses a CMP instruction to look for opponents.
          

    example   add step,scan
    scan      cmp 10,30
              jmp attack
              jmp example
    step      dat #20,#20

   Color
          Property of bombs making them visible to scanners, causing them
          to attack useless locations, thus slowing them down.
          

    example   dat #100

   Core-Clear
          code that sequentially overwrites core with DAT instructions;
          usually the last part of a program.
          
   Decoys
          Bogus or unused instructions meant to slow down Scanners.
          Typically, DATs with non-zero B-fields.
          
   DJN-Stream (also DJN-Train)
          Using a DJN command to rapidly decrement core locations.
          

    example   ...
              ...
              djn example,<4000

   Dwarf
          the prototypical small bomber.
          
   Gate-busting (also gate-crashing)
          technique to "interweave" a decrement-resistant imp-spiral
          (e.g. MOV 0, 2668) with a standard one to overrun imp-gates.
          
   Hybrids
          warriors that combine two or more of the basic strategies,
          either in sequence (e.g. stone->paper) or in parallel (e.g.
          imp/stone).
          
   Imp
          Program which only uses the MOV instruction.
          

    example   MOV 0, 1
  or
    example   MOV 0, 2
              MOV 0, 2

   Imp-Gate
          A location in core which is bombed or decremented continuously
          so that an Imp can not pass. Also used to describe the
          program-code which maintains the gate.
          

    example   ...
              ...
              SPL 0, <example
              DAT <example, #0

   Imp-Ring
          A minimal Imp-Spiral.
          

    d         EQU (coresize+1)/3
    A         MOV 0,d   ; copy self to B
    B         MOV 0,d   ; copy self to C
    C         MOV 0,d   ; copy self to A+1

   Imp-Spiral
          An Imp-like program with two or more processes supporting each
          other. A three-point spiral, with six processes running in this
          sequence:
          

    d         EQU (coresize+1)/3
    A         MOV 0,d   ; copy self to B
    B         MOV 0,d   ; copy self to C
    C         MOV 0,d   ; copy self to A+1
    A+1       MOV 0,d   ; copy self to B+1
    B+1       MOV 0,d   ; copy self to C+1
    C+1       MOV 0,d   ; copy self to A+2

   Incendiary Bomb
          A type of Stun bomb which creates a SPL 0 carpet.
          

    example   SPL 0, 8
              MOV -1, <-1

   JMP/ADD Launch
          one of several means to start an imp-spiral running. The
          slowest launch technique, but requires the least code. See also
          Binary Launch and Vector Launch.
          

    impsize   equ 2667
    example   spl 1               ; extend by adding more spl 1's
              spl 1
              spl 2
              jmp @0,imp
              add #impsize,-1     ; bump address by impsize after each jmp
              dat #0,#0           ; excess processes must die!
    imp       mov 0,impsize       ; in '94 use -> mov.i #0,impsize

   Mirror
          see reflection.
          
   On-axis/off-axis
          On-axis scanners compare two locations M/2 apart, where M is
          the memory size. Off-axis scanners use some other separation.
          
   Optimal Constants
          (also optima-type constants) Bomb or scan increments chosen to
          cover core most effectively, i.e. leaving gaps of uniform size.
          Programs to calculate optimal constants and lists of optimal
          numbers are available at ftp.csua.berkeley.edu.
          
   Paper
          A Paper-like program. One which replicates itself many times.
          Part of the Scissors (beats) Paper (beats) Stone (beats
          Scissors) analogy.
          
   Pit-Trapper
          (also Slaver, Vampire). A program which enslaves another.
          Usually accomplished by bombing with JMPs to a SPL 0 pit with
          an optional core-clear routine.
          
   Quick Scan
          2c scan of a set group of core locations with bombing if
          anything is found. Both of the following codes snips scan 16
          locations and check for a find. If anything is found, it is
          attacked, otherwise 16 more locations are scanned. Example:
          

     start
     s1 for 8    ;'88 scan
            cmp  start+100*s1, start+100*s1+4000  ;check two locations
            mov  #start+100*s1-found, found  ;they differ so set pointer
     rof
            jmn  attack,     found     ;if we have something, get it
     s2 for 8
            cmp  start+100*(s2+6), start+100*(s2+6)+4000
            mov  #start+100*(s2+6)-found, found
     rof
     found  jmz  moveme,  #0        ;skip attack if qscan found nothing
     attack cmp  @found,  start-1   ;does found points to empty space?
            add  #4000,   found     ;no, so point to correct location
            mov  start-1, @found    ;move a bomb
     moveme jmp  0,       0

   In ICWS'94, the quick scan code is more compact because of the SNE
          opcode:
          


     start       ;'94 scan
     s1 for 4
            sne  start+400*s1, start+400*s1+100  ;check two locations
            seq  start+400*s1+200, start+400*s1+300  ;check two locations
            mov  #start+400*s1-found, found  ;they differ so set pointer
     rof
            jmn  which,   found     ;if we have something, get it
     s2 for 4
            sne  start+400*(s2+4), start+400*(s2+4)+100
            seq  start+400*(s2+4)+200, start+400*(s2+4)+300
            mov  #start+400*(s2+4)-found-100, found
     rof
     found  jmz  moveme,  #0        ;skip attack if qscan found nothing
            add  #100,    -1        ;increment pointer till we get the
     which  jmn  -1,      @found    ;right place
            mov  start-1, @found    ;move a bomb
     moveme jmp  0,       0

   Reflection
          Copy of a program or program part, positioned to make the
          active program invisible to a CMP-scanner.
          
   Replicator
          Generic for Paper. A program which makes many copies of itself,
          each copy also making copies.
          
   Self-Splitting
          Strategy of amplifying the number of processes executing a
          piece of code.
          

    example    SPL 0
    loop       ADD #10, example
               MOV example, @example
               JMP loop

   Scanner
          A program which searches through core for an opponent rather
          than bombing blindly.
          
   Scissors
          A program designed to beat replicators, usually a (B-field
          scanning) vampire. Part of the Paper-Scissors-Stone analogy.
          
   Self-Repair
          Ability of a program to fix it's own code after attack.
          
   Silk
          A replicator which splits off a process to each new copy before
          actually copying the code. This allows it to replicate
          extremely quickly. This technique is only possible under the
          '94 draft, because it requires post-increment indirect
          addressing. Example:
          

                spl     1
                mov.i   -1,     0
                spl     1               ;generate 6 consecutive processes
    silk        spl     3620,   #0      ;split to new copy
                mov.i   >-1,    }-1     ;copy self to new location
                mov.i   bomb,   >2000   ;linear bombing
                mov.i   bomb,   }2042   ;A-indirect bombing for anti-vamp
                jmp     silk,   {silk   ;reset source pointer, make new copy
    bomb        dat.f   >2667,  >5334   ;anti-imp bomb

   Slaver
          see Pit-Trapper.
          
   Stealth
          Property of programs, or program parts, which are invisible to
          scanners, accomplished by using zero B-fields and reflections.
          
   Stone
          A Stone-like program designed to be a small bomber. Part of the
          Paper-Scissors-Stone analogy.
          
   Stun
          A type of bomb which makes the opponent multiply useless
          processes, thus slowing it down. Example is referred to as a
          spl-jmp bomb.
          

    example    spl 0
               jmp -1

   Two-Pass Core-Clear (also spl/dat Core-Clear)
          core clear that fills core first with SPL instructions, then
          with DATs. This is very effective in killing paper and certain
          imp-spiral variations.
          
   Vampire
          see Pit-Trapper.
          
   Vector Launch
          one of several means to start an imp-spiral running. As fast as
          Binary Launch, but requiring much less code. See also JMP/ADD
          Launch and Binary Launch. This example is one form of a Vector
          Launch:
          

    impsize   equ 2667
    example   spl 1                       ; extend by adding more spl 1's
              spl 1
              djn.a   @imp,#0             ; jmp @ a series of pointers
              dat     #0,imp+(3*impsize)
              dat     #0,imp+(2*impsize)
              dat     #0,imp+(1*impsize)
              dat     #0,imp+(0*impsize)
    imp       mov.i   #0,impsize

   
   
   [ToC]
     _________________________________________________________________
   
Other questions?

   Just ask in the rec.games.corewar newsgroup or contact me (address
   below). If you are shy, check out the Core War archives first to see
   if your question has been answered before.
   
   [ToC]
     _________________________________________________________________
   
Credits

   Additions, corrections, etc. to this document are solicited. Thanks in
   particular to the following people who have contributed major portions
   of this document: Paul Kline, Randy Graham.
   Mark Durham wrote the first version of the FAQ.
   
   The rec.games.corewar FAQ is Copyright 1995 and maintained by:
   

Stefan Strack, PhD                          stst@vuse.vanderbilt.edu
Dept. Molecular Physiol. and Biophysics     stst@idnsun.gpct.vanderbilt.edu
Rm. 762, MRB-1                              stracks@vuctrvax.bitnet
Vanderbilt Univ. Medical Center             Voice: +615-322-4389
Nashville, TN 37232-6600, USA               FAX:   +615-322-7236

   
     _________________________________________________________________
   
   $Id: corewar-faq.html,v 3.4 1995/08/06 21:38:09 stst Exp stst $



From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: Agony II
Date: 1995/09/26
Message-ID: <4493d5$aod@lastactionhero.rs.itd.umich.edu>#1/1
references: <446t9k$7gt@mozo.cc.purdue.edu>
newsgroups: rec.games.corewar

Myer R. Bremer (bremermr@elsun269.cc.purdue.edu) wrote:
: :I think Agony does so well because the CMP command is the fastest
: :way to scan core.  While stones are fast,(1C for each attack) the CMP scanner
: :does it's scanning at 2C.  (the CMP instruction scans TWO  locations
: :in core at a time.)

: :By the way, just how far ahead does Agony boot itself? (ahead meaning
: :positive or negative locations in core?)

: :John Lewis

: I don't think Stefan is about to reveal to us his exact boot
: distance.  Nice try, though.

: M R Bremer

I was hoping he read his news really early (before coffee).

:)





From: John <jwilkinson@mail.utexas.edu>
Subject: Minor PIZZA change
Date: 1995/09/26
Message-ID: <447mrh$n5e@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Well, I requested long ago that TUC implemenet of 5x100 battle setup...

He said he wanted to see what the newgroup thought.

I posted the idea.

I got very little response...

Is that good or bad?  Could you guys at least vote or something?  Just
reply yes or no, or mail TUC or something...  I dunno.

If eveybody thinks this is a bad idea, please tell ME, because I can't
think of any reason not to do this...

Ah well, please respond.

-John




From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/26
Message-ID: <447s6m$hh0@agate.berkeley.edu>#1/1
references: <mreddy-2509951116240001@mreddy.comp.glam.ac.uk> <446v1r$6p0@hermes.cair.du.edu>
newsgroups: comp.ai.alife,rec.games.corewar

The address worked for me this time around.  Maybe the server was down or 
something.




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: What is a process?
Date: 1995/09/26
Message-ID: <447mdq$n5e@geraldo.cc.utexas.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <441nga$1b2@allnews.infi.net> <4448jv$8df@geraldo.cc.utexas.edu> <447cd2$ks6@allnews.infi.net>
newsgroups: rec.games.corewar

-Okay, ya gotta gimme another chance...
-A process is a program in execution.  The execution of a process must
-progress in a sequential fashion (unless otherwise directed by the next
-instruciton in the sequence. ie. a branch or skip)  
-A process is more than the program code.  It also includes the current
-activity, as represented by the value of the program counter and the contents
-of the processor's registers (except in the case of Core Wars where registers
-aren't used), as well as a data section containing global variables.  
-A process is an active entity, with a program counter specifying the next
-instruction to be executed.
-More than one process can be associated with the same program and they
-are considered separate execution sequences.  Once launched, they're on
-their own.  One program can 'spawn' one or more processes.  It's like
-turtle eggs hatching.  (okay, bad example)
-
-Scott

Except for the turtle egg comment, that was grade a material. :)

You get a gold star! <John sticks a gold star on Scott's letter>


-John




From: pk6811s@acad.drake.edu
Subject: Re: Minor PIZZA change
Date: 1995/09/27
Message-ID: <1995Sep27.105948@acad.drake.edu>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

<jwilkinson@mail.utexas.edu> writes:
> 
> I posted the idea.
> 
> I got very little response...

Take two aspirin, go to bed repeating the phrase: "it's just a game..."

:-)

It is great to have some enthusisatic (and successful!) participants.

But what you are proposing (5x100 rounds) is not convincingly necessary
for the health of CoreWar.  It is a change in the environment that may
favor one class of programs.  Is that good?  Most people are probably
waiting to see how well pspacers perform in the existing environ before
changing it to favor them.  Also, if pspace is such a huge advantage,
then maybe it would be a mistake to change the rules to make it even stronger.
For now, let this be the Achille's heel of pspace - high variability in
outcome - and see how various authors deal with it; maybe it can be overcome..

In real life, 'ingelligent' life forms sometimes make bad decisions.

$.02 from:

Paul Kline
pk6811s@acad.drake.edu



From: Beppe Bezzi <bezzi@iol.it>
Subject: P-space
Date: 1995/09/27
Message-ID: <199509272225.XAA12771@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

It's more than a month we have p-space, let me try to make some consideration:

P-warrior are *not* dominating the hill; currently in the 94 hill are half
and half with standard warriors, and new versions of some 'old friend' is
reappeared scoring well. 

Two warrior classes, stones and replicators, have disappeared, but scanners
seem to be well alive and Thermite, the quick scanner, is scoring better
than ever.

Vampires have come back. Maybe because none has put anti-vamp component in
new warriors, maybe because Magnus did a very good work and he did, (I know
myVamp but I cannot give hints without permission :-) they are here, and
vamps are wonderful brainwashing tools :-)

A new class of warriors, (not sure they are a new class but I think yes) the
p-attackers is appeared soon after the first p-warriors. I know for sure
that there are not only Braiwash ad Brain Vamp to scramble p-space. If you
want to know who are they submit a p-warrior configured so as it always
switches on the same component and then again the same but make it read from
p-space where to switch; you'll note some difference in scoring.
P-scramblers, or brainwashers are weaker than p-warriors against non p-,
because you have to pay something, time or space, to do a thing their
enemies are indifferent at, but can score well against p-, expecially if
they have sophisticated swithcers and specialized components. More on this
in another article, together with Brain Wash code and some tips on brainwashing.
BTW I'm curious to hear if Brain Wash has changed your approach to p-space;
I think yes, because new warriors are less sensitive than older, but I would
like to hear others opinions.

Now I have tired you enough of empty speaking, let me go and look at results
of Jack, *not* a brainwasher, and then write something about Brain Wash.

-Beppe Bezzi

<bezzi@iol.it>




From: pan0178@comune.bologna.it (MV)
Subject: Re: What is a process?
Date: 1995/09/27
Message-ID: <199509271406.KAA17120@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

>-Okay, ya gotta gimme another chance...
>-A process is a program in execution.  The execution of a process must
>-progress in a sequential fashion (unless otherwise directed by the next
>-instruciton in the sequence. ie. a branch or skip)  
>-A process is more than the program code.  It also includes the current
>-activity, as represented by the value of the program counter and the contents
>-of the processor's registers (except in the case of Core Wars where registers
>-aren't used), as well as a data section containing global variables.  
>-A process is an active entity, with a program counter specifying the next
>-instruction to be executed.
>-More than one process can be associated with the same program and they
>-are considered separate execution sequences.  Once launched, they're on
>-their own.  One program can 'spawn' one or more processes.  It's like
>-turtle eggs hatching.  (okay, bad example)
>-
>-Scott
>
>Except for the turtle egg comment, that was grade a material. :)
>
>You get a gold star! <John sticks a gold star on Scott's letter>
>
>
>-John
>
Yeah! But you forgot to tell us the difference between threads (also known
as light processes) and heavy processes (also known simply as processes).

;-)
I'm just kiddin' ;-P

Maurizio




From: ruhl@phoebe.cair.du.edu (Robert A. Uhl)
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/27
Message-ID: <44c5n4$dhu@hermes.cair.du.edu>#1/1
references: <446v1r$6p0@hermes.cair.du.edu> <F5T1BD4w200w@alcyone.darkside.com>
newsgroups: comp.ai.alife,rec.games.corewar

In article <F5T1BD4w200w@alcyone.darkside.com>,
Erik Max Francis <max@alcyone.darkside.com> wrote:
>
>Ahem!  There's a world of difference between shareware (or even 
>freeware!) and public domain.

  But if the compiler company allows shareware, which costs, then what
would it have against PD, which is free, or freeware?  PD does not
mean program modifiable, but modification of resources (with a Mac, a
program can easily be translated, new pictures added &c.) and source
modification if it is included.  The compiler company should have no
problem with this.

-- 
+------------------------------------------------------------------------+
| Bob Uhl | Spectre                  | `En touto nika' +                 |
| U of D  | Baron Robert von Raetzin | http://mercury.cair.du.edu/~ruhl/ |
+------------------------------------------------------------------------+



From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: one-pass scanners
Date: 1995/09/27
Message-ID: <44bscl$a1l@lastactionhero.rs.itd.umich.edu>#1/1
references: <1995Sep26.112750@acad.drake.edu> <449gnv$roo@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

That's an interesting idea.  How much longer would it take to have a hill
with 30 warriors on it?

One of the things that I noticed was that I could get to the top of the
begginer hill, but that warrior would not get on the regular hill.  This
makes it hard for me to fix the problems with this warrior.

John Lewis



John (jwilkinson@mail.utexas.edu) wrote:
: >One-pass scanning has been a cyclical strategy.  Someone puts one up and
: >does very well.  Others copy.  Then people add boot-and-decoy and drive
: >the one-passers off.  Then people strip out the b&d to save startup time.
: >Then one-passers appear again.  The first one-passers were b-scanners,
: >then cmp-scanners, now mov/cmp-scanners, each reincarnation being faster
: >than the last, but all with the same Achilles' heel.
: >-- 
: >Paul Kline
: >pk6811s@acad.drake.edu

: Indeed, I've noticed similar oddness...  The thing is, I'm estimating
: about 4-7 -successful- challenges to the Hill per day.  This makes for
: very rapid evolution... which does have the interesting side effect
: of having very little point spread between the top and bottom warrior.
: At the moment, the King is less than 20 points above the last place
: warrior.  That's only about 15% differnce...

: Perhaps, instead of instituting the 5x100 system, we should add more
: spaces to the hill?  This would slow down the rapid change of the
: hill, and allow good warriors to stay on longer?

: Something to think about.  At least, -I'll- be thinking about it. :)

: -John




From: jklewis@stimpy.us.itd.umich.edu (John Kipling Lewis)
Subject: Re: Minor PIZZA change
Date: 1995/09/27
Message-ID: <44bkb1$62p@lastactionhero.rs.itd.umich.edu>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

I think that if this lessens the chance of fluke p-space problems
and result then I am for it.  From what I have heard there would be
very little increase in time for results to return...  sounds like
a no lose situation.

John Lewis (hey John, maybe we should start using our last names)
:)

John (jwilkinson@mail.utexas.edu) wrote:
: Well, I requested long ago that TUC implemenet of 5x100 battle setup...

: He said he wanted to see what the newgroup thought.

: I posted the idea.

: I got very little response...

: Is that good or bad?  Could you guys at least vote or something?  Just
: reply yes or no, or mail TUC or something...  I dunno.

: If eveybody thinks this is a bad idea, please tell ME, because I can't
: think of any reason not to do this...

: Ah well, please respond.

: -John




From: mreddy@glam.ac.uk (Mike Reddy)
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/27
Message-ID: <mreddy-2709951126120001@mreddy.comp.glam.ac.uk>#1/1
references: <mreddy-2509951116240001@mreddy.comp.glam.ac.uk> <446v1r$6p0@hermes.cair.du.edu>
newsgroups: comp.ai.alife,rec.games.corewar

In article <446v1r$6p0@hermes.cair.du.edu>, ruhl@phoebe.cair.du.edu
(Robert A. Uhl) wrote:

>   Actually, you are allowed to distribute the program; the studetn
> license allows shareware, so PD distribution shouldn't be any problem
> whatsoever.

Can someone confirm whether the license does allow software distribution?
If you can put things in the public domain, then I will of course add the
apps to the FTP site (ftp://ftp.comp.glam.ac.uk/pub/staff/mreddy/) as well
as the sources.


Yours Mike Reddy
--
Email: mreddy@glam.ac.uk CU-Seeme: 193.63.130.40 (On Request)
Web: http://www.comp.glam.ac.uk/pages/staff/mreddy/mreddy.html
Snail: J228, Dept. of Computer Studies, University of Glamorgan, Pontypridd, Mid Glamorgan. CF37 1DL Wales, UK. +44 1443 482 240 Fax: +44 1443 482 715



From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Minor PIZZA change
Date: 1995/09/27
Message-ID: <44agrr$lg0@geraldo.cc.utexas.edu>#1/1
references: <199509261452.PAA00711@iol-mail.iol.it>
newsgroups: rec.games.corewar

>I see no real need to change from 100 battles to 5*100 or 1*500, but I have
>nothing against; apart that testing on my machine will take 5 times more,
>and five times half an hour for my 12 battles times 100 rounds benchmark is
>_a_lot_ more.
>
>I'll go on testing with but 100 battles as I do now.

Why would your testing at home be effected by the 5x100 change?

I usually run at least 1000 rounds or so per warrior before I up
it to the Hill... and I wasn't planning on running 5000 rounds,
because the hill does 5 times more...

-John




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: Minor PIZZA change
Date: 1995/09/27
Message-ID: <44agn3$lg0@geraldo.cc.utexas.edu>#1/1
references: <447mrh$n5e@geraldo.cc.utexas.edu> <448p8c$bt9@valhalla.stormking.com>
newsgroups: rec.games.corewar

>	Um, John, not to pick a nit.... But I run Stormking.Com ....
>And as for upping the battles, (Stefan and Thos will probably kill me but)
>WATCH THIS SPACE FOR AN IMPORTANT ANNOUNCEMENT SOON REGARDING THIS.....
>(Perhaps, lets say, before October 1)
>
>				Tuc

Um, ok... I don't know what you're talking about, but my peepers are peeled.

-John







From: lohwengk@iscs.nus.sg (Loh Weng Keong Calvin)
Subject: Re: What is a process?
Date: 1995/09/28
Message-ID: <44cv59$9fb@nuscc.nus.sg>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg> <4448n1$8df@geraldo.cc.utexas.edu> <4456md$f7v@nuscc.nus.sg> <1995Sep25.084133@acad.drake.edu> <4481lo$73i@nuscc.nus.sg>
newsgroups: rec.games.corewar

With my newly acquired knowledge, I meditated over Steve Morrell's tutorial
last night and the light dawned on me. Finally understand the first 4
warriors. The details of the 5th one, 'worm' still escapes me, but should
not be unsolvable.
Thanks to everyone who helped me.
--
Calvin Loh



From: arthur@camba.com
Subject: Re: What is a process?
Date: 1995/09/28
Message-ID: <44dffj$rfa@news.voicenet.com>#1/1
references: <43qjs2$sib@nuscc.nus.sg>
newsgroups: rec.games.corewar

On 21 Sep 1995 02:49:38 GMT, lohwengk@iscs.nus.sg (Loh Weng Keong
Calvin) wrote:

// I have just gone through the first 3 warriors in Steve Morrell's
// corewar tutorial and came across a term I do not really 
// understand. What is meant by process?
// Is a process a line of instruction?
// Or is it the sequence of instructions making up the warrior?
// TIA.
// 
// --
// Calvin Loh

Where would I be able to find this tutorial?
-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: 2.6.2

mQCNAzBSf0wAAAEEALyWxjwL2QVqWJu4Zzw7WDpbG4tEvDIdSt6Gd3QMQYMXA/KZ
5FmyMLwoNxNJanLzqcFHoMohBtAfcMdtpZ5zWdQXoDu+ikIE5g1zQ9HCBJuV07dJ
s/YkZLFet0oIVt/z4q0hCiTQaB0SxlsHNSF+kWFup4VYtpzIpE3O7o+VLaRNAAUR
tBRBcnRodXIgQS4gVmFuZGVyYmlsdA==
=5Af2
-----END PGP PUBLIC KEY BLOCK-----




From: John <jwilkinson@mail.utexas.edu>
Subject: Re: P-space
Date: 1995/09/28
Message-ID: <44f8m7$sv3@geraldo.cc.utexas.edu>#1/1
references: <199509272225.XAA12771@iol-mail.iol.it> <1995Sep28.104806@acad.drake.edu>
newsgroups: rec.games.corewar

>Here's an approach to counter brainwashing:
>
>           ldp.a #pstragy,wstragy
>           mod.a #10,wstragy
>wstragy    jmp  @0,str1
>           dat   0,str2
>           dat   0,str3
>       for 6
>           dat   0,str1    ; preferred anti-brainwash strategy
>       rof
>
>If you are brainwashed by a random number or zero, then the mod.a #10 will pick
>str1 most of the time.  Hopefully str1 is a strategy that works well
>against the brainwasher :-)
>
I think that's a bad idea.  If str1 is good against the brainwasher, then the
switching mechanism will find that str1 is good, and it will use str1.

However, if str1 is bad, you are stuck using it over and over and over... in
a loop, with only 2/9 chance of getting out.

Logically, this approach would seem to only hurt your chances?

Comments?

>One pspace strategy that is *not* working as well as I had thought is pStone.
>pStone switches a fast stone&coreclear with Die Hard with this logic:
>
>    if stone loses switch to Die Hard
>    if Die Hard wins switch to stone
>
>Basically the stone beats most of the stand-still, non-pspacing opponents
>like Agony, Torch, and Taking Names.  And hides under Die Hard when it
>loses.  Since Die Hard can only beat a stand-still program, the stone waits
>until then to reappear.  Sort of a crawdad lifestyle.
>
>Unfortunately the other pspacers adapt to select replicators over stand-still
>strategies and pStone winds up with a lot of ties, holding its score down.
>Looks like a simple play-it-safe strategy won't quite cut it.

Why are you switching Die Hard and a stone?  Die Hard seams to beat all the same
things that a stone beats.  That seems like a fooish choice...  Logically, you
would want to run a scanner or scissors to kill things that Die Hard can't kill,
right? I mean, that is what Pink was doing... And it was riding on top of the hill
until you switched it up and pulled it off...

-- 
-John K. Wilkinson-











From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: Re: What is a process?
Date: 1995/09/28
Message-ID: <44ep28$5av@agate.berkeley.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg> <4448n1$8df@geraldo.cc.utexas.edu> <4456md$f7v@nuscc.nus.sg> <1995Sep25.084133@acad.drake.edu> <4481lo$73i@nuscc.nus.sg>
newsgroups: rec.games.corewar

Yeah, some warrior's will keep on looping... note, however, they can 
change the data they are acting on in the meantime, so they are not 
always doing exactly the same thing.  A dwarf's a pretty good example of 
a program that keeps looping.  An imp just copies itself into the next 
core space, which it proceeds to execute the next cycle.  Oh, where's the 
tutorial?  I could use something like that.




From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: P-space
Date: 1995/09/28
Message-ID: <199509290128.CAA01797@iol-mail.iol.it>
newsgroups: rec.games.corewar

Paul Kline wrote:
>Beppe Bezzi <bezzi@iol.it> writes:
>>...
>> P-warrior are *not* dominating the hill; currently in the 94 hill are half
>> and half with standard warriors, and new versions of some 'old friend' is
>> reappeared scoring well. 
>
>I'm _guessing_ this is due to the one-pass scan strategy used in most
>of the p-warriors submitted so far(?)  It is easily defeated by non-pspacers.
>
Maybe it's true. The problem with different kind of scanners is the big
'decoy' of others warrior's component you have to deal with. Doing this with
a one shot scanner is much easier. Avoiding the scanner, as I did in my
p-warriors, is even easier :-)

>> Two warrior classes, stones and replicators, have disappeared, but scanners
>> seem to be well alive and Thermite, the quick scanner, is scoring better
>> than ever.
>
>Replicators were just about driven off by Withershins Thrice, Porch Swing,
>and similar.  ...
Marcia Trionfale was driven off when the entry level of the hill raised to
near 135, a score nearly inpossible for a pure replicator that cannot score
more than 100 points against others replicators, is subject to scanners and
has no more stones to eat. I have an idea of a replicator able to partially
overcome the problem, but too little time to develop it; hope to appear soon
on the hill with Marcia II

>Thermite feeds on large pspace opponents, frequently
>catching them during startup.  A good qscanner can locate a large opponent
>in under 20 cycles about half the time, and under 10 cycles about one-fourth
>the time.  10 cycles = (20 compares) * (stepsize of 100) = 2000 core locations
>sparsely scanned.
>
I know that, what's strange is that no others have appeared; I tried one
with bad results. The scanning was OK, but after; what to do if you found
something, carpet bombing, bombing through a and b fields? and what if not?
never seen one before to have any idea. 

>> Vampires have come back. Maybe because none has put anti-vamp component in
>
>Vampires were driven off by Silk-style replicators using a-field indirect
>bombing, which is very efficient anti-vamp strategy.  But most of those
>are gone.  
:-(
>
>> A new class of warriors, (not sure they are a new class but I think yes) the
>> p-attackers is appeared soon after the first p-warriors. I know for sure
>>...
>
>One of the neat things about brainwashing is that you don't have to win
>the battle to be sucessful.  The opponent is 'washed' for the next battle.
:-)))
>
>Here's an approach to counter brainwashing:
>
>           ldp.a #pstragy,wstragy
>           mod.a #10,wstragy
>wstragy    jmp  @0,str1
>           dat   0,str2
>           dat   0,str3
>       for 6
>           dat   0,str1    ; preferred anti-brainwash strategy
>       rof
>
>If you are brainwashed by a random number or zero, then the mod.a #10 will pick
>str1 most of the time.  Hopefully str1 is a strategy that works well
>against the brainwasher :-)
>
That's fine if you know how you are brainwashed. I'm near to publish Brain
Wash, (I'd have done it now but this thread is cooler, and I hope BW will
survive till tomorrow ;-) so I can tell that BW uses a paper module with a
Bwashing pit, BVamp is even better because uses both vamp and paper to
brainwash. Difficult to tune against both, I hope :-)

Another thing I noted, and implemented in BW was to use *small* numbers as
brainwashing soap. BW uses something beetween 5 and 8, don't remember ;-),
that works very well with p-warriors like Pink checking the strategy for
some rounds; using smaller numbers is even better, a 1 or 2 can foul every
kind of switcher; I'll use it in next version. BTW during test I noticed
some warriors did better when braiwashed than when Bw was off; we have a lot
to learn on how to use p-space :-)

>One pspace strategy that is *not* working as well as I had thought is pStone.
>pStone switches a fast stone&coreclear with Die Hard with this logic:
>
>    if stone loses switch to Die Hard
>    if Die Hard wins switch to stone
>
>Basically the stone beats most of the stand-still, non-pspacing opponents
>like Agony, Torch, and Taking Names.  And hides under Die Hard when it
>loses.  Since Die Hard can only beat a stand-still program, the stone waits
>until then to reappear.  Sort of a crawdad lifestyle.
                                    ^^^^^^^
What's a crawdad ? Just hope nothing sheek ;-) (got crazy to figure out what
you meant)

>Unfortunately the other pspacers adapt to select replicators over stand-still
>strategies and pStone winds up with a lot of ties, holding its score down.
>Looks like a simple play-it-safe strategy won't quite cut it.
>
Braiwash uses a very aggressive switching:
        keep on every winning strategy, Tornado included
        switch after a tie to the next, list include Firestorm, paper, and a
sweep, not Tornado
        switch to Tornado if paper lose (I'm facing a scanner and Tornado is
a near optimal killer)
        brainwash, with paper, if any other module lost. (I hope to meet a
paper so as washing is complete)

Jack in the box plays safe instead:
        keep on wins and tie
        switch after a loss

Doing so I protect Tornado from Die Hard & co, I crush scanners, and I score
average high against others.
I'm surprised a so simple switcher, 8 lines 6/7 cycles, with so weak
components, none scored good alone, can score so well; I'm just happy to
have been king for some moments.

>Paul Kline
-Beppe Bezzi

<bezzi@iol.it>




From: everard@infi.net (M. Scott Everard)
Subject: Re: What is a process?
Date: 1995/09/28
Message-ID: <44fa2n$t46@allnews.infi.net>#1/1
references: <199509271406.KAA17120@valhalla.stormking.com>
newsgroups: rec.games.corewar

In article <199509271406.KAA17120@valhalla.stormking.com>, pan0178@comune.bologna.it (MV) says:
>
>>-Okay, ya gotta gimme another chance...
>>-A process is a program in execution.  The execution of a process must
>>-progress in a sequential fashion (unless otherwise directed by the next
>>-instruciton in the sequence. ie. a branch or skip)  
>>-A process is more than the program code.  It also includes the current
>>-activity, as represented by the value of the program counter and the contents
>>-of the processor's registers (except in the case of Core Wars where registers
>>-aren't used), as well as a data section containing global variables.  
>>-A process is an active entity, with a program counter specifying the next
>>-instruction to be executed.
>>-More than one process can be associated with the same program and they
>>-are considered separate execution sequences.  Once launched, they're on
>>-their own.  One program can 'spawn' one or more processes.  It's like
>>-turtle eggs hatching.  (okay, bad example)
>>-
>>-Scott
>>
>>Except for the turtle egg comment, that was grade a material. :)
>>
>>You get a gold star! <John sticks a gold star on Scott's letter>
>>
>>
>>-John
>>
>Yeah! But you forgot to tell us the difference between threads (also known
>as light processes) and heavy processes (also known simply as processes).
>
>;-)
>I'm just kiddin' ;-P
>
>Maurizio
>


Okay now... don't get me started! 
I can lecture for days on end! :)



































From: Peter Young <pyoung1@uclink4.berkeley.edu>
Subject: Re: What is a process?
Date: 1995/09/28
Message-ID: <44ep2e$5eg@agate.berkeley.edu>#1/1
references: <43qjs2$sib@nuscc.nus.sg> <43vq0s$3mb@allnews.infi.net> <440gar$nm0@geraldo.cc.utexas.edu> <442ouq$1ai@nuscc.nus.sg> <4448n1$8df@geraldo.cc.utexas.edu> <4456md$f7v@nuscc.nus.sg> <1995Sep25.084133@acad.drake.edu> <4481lo$73i@nuscc.nus.sg>
newsgroups: rec.games.corewar

Yeah, some warrior's will keep on looping... note, however, they can 
change the data they are acting on in the meantime, so they are not 
always doing exactly the same thing.  A dwarf's a pretty good example of 
a program that keeps looping.  An imp just copies itself into the next 
core space, which it proceeds to execute the next cycle.  Oh, where's the 
tutorial?  I could use something like that.




From: pk6811s@acad.drake.edu
Subject: Re: P-space
Date: 1995/09/28
Message-ID: <1995Sep28.104806@acad.drake.edu>#1/1
references: <199509272225.XAA12771@iol-mail.iol.it>
newsgroups: rec.games.corewar

Beppe Bezzi <bezzi@iol.it> writes:
>...
> P-warrior are *not* dominating the hill; currently in the 94 hill are half
> and half with standard warriors, and new versions of some 'old friend' is
> reappeared scoring well. 

I'm _guessing_ this is due to the one-pass scan strategy used in most
of the p-warriors submitted so far(?)  It is easily defeated by non-pspacers.

> Two warrior classes, stones and replicators, have disappeared, but scanners
> seem to be well alive and Thermite, the quick scanner, is scoring better
> than ever.

Replicators were just about driven off by Withershins Thrice, Porch Swing,
and similar.  Thermite feeds on large pspace opponents, frequently
catching them during startup.  A good qscanner can locate a large opponent
in under 20 cycles about half the time, and under 10 cycles about one-fourth
the time.  10 cycles = (20 compares) * (stepsize of 100) = 2000 core locations
sparsely scanned.

> Vampires have come back. Maybe because none has put anti-vamp component in
> new warriors, maybe because Magnus did a very good work and he did, (I know

Vampires were driven off by Silk-style replicators using a-field indirect
bombing, which is very efficient anti-vamp strategy.  But most of those
are gone.  

> A new class of warriors, (not sure they are a new class but I think yes) the
> p-attackers is appeared soon after the first p-warriors. I know for sure
>...

One of the neat things about brainwashing is that you don't have to win
the battle to be sucessful.  The opponent is 'washed' for the next battle.

Here's an approach to counter brainwashing:

           ldp.a #pstragy,wstragy
           mod.a #10,wstragy
wstragy    jmp  @0,str1
           dat   0,str2
           dat   0,str3
       for 6
           dat   0,str1    ; preferred anti-brainwash strategy
       rof

If you are brainwashed by a random number or zero, then the mod.a #10 will pick
str1 most of the time.  Hopefully str1 is a strategy that works well
against the brainwasher :-)

One pspace strategy that is *not* working as well as I had thought is pStone.
pStone switches a fast stone&coreclear with Die Hard with this logic:

    if stone loses switch to Die Hard
    if Die Hard wins switch to stone

Basically the stone beats most of the stand-still, non-pspacing opponents
like Agony, Torch, and Taking Names.  And hides under Die Hard when it
loses.  Since Die Hard can only beat a stand-still program, the stone waits
until then to reappear.  Sort of a crawdad lifestyle.

Unfortunately the other pspacers adapt to select replicators over stand-still
strategies and pStone winds up with a lot of ties, holding its score down.
Looks like a simple play-it-safe strategy won't quite cut it.

Paul Kline
pk6811s@acad.drake.edu



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's Beginner Hill
Date: 1995/09/28
Message-ID: <44ek0h$inq@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server Beginner Hill:
(accessed with ";redcode-b")

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.
  instruction set: ICWS '94 Draft

Last challenge: Wed Sep 27 11:17:59 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  57/  4/ 39                   Rabbits    Maurizio Vittuari     209      49
 2  52/  3/ 44                   uhhhhum      Magnus Paulsson     201      38
 3  53/ 34/ 13            Mythicon v1.1a             G. Eadon     172       6
 4  47/ 29/ 24                  Eft v1.2    Maurizio Vittuari     164      83
 5  43/ 33/ 25                   Obvious        John K. Lewis     153      99
 6  38/  2/ 31                   Rabbits    Maurizio Vittuari     144      56
 7  26/ 13/ 61                 ShadowImp     Robert J. Street     138      21
 8  37/ 37/ 27                    Test-D             G. Eadon     136       1
 9  37/ 45/ 18                 Maya v2.0    Christoph C. Birk     130      25
10  28/ 34/ 38                    double          Kurt Franke     122       9
11  30/ 35/ 16            CoreClear v0.3            Anonymous     105      71
12   8/ 17/ 75           JumpCrawler .21               KDavis      99      16
13   8/ 17/ 75             JumpCrawler 2               KDavis      99      18
14  30/ 65/  5         Pebbles v1.2 (9+)    Christoph C. Birk      95      51
15  16/ 37/ 48                  REVENGE!        Jonathan West      95      14
16   2/ 21/ 76    He-He can't kill me! 2        Jonathan West      83      15
17   4/ 27/ 69         ShadowSlaver v.13     Robert J. Street      81      17
18  12/ 46/ 42             Testing 1 2 3          Kurt Franke      79      10
19  13/ 65/ 22                    Jumper           John Lewis      61       2
20  10/  3/  1                Emc2 v1.0B    Maurizio Vittuari      32      82

21   2/  2/  1                    Test-C             G. Eadon       8       3

A new experiment will be unleashed October 1st.  By popular demand,
all hills will be increased from 100 to 250 round battles.  Also,
the -94x hill will be replaced with a new 5*50 round hill.  Submissions
to this hill will be mirrored from the -94 hill, no special switches
will be necessary.

--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "koth help" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's Experimental '94 Hill
Date: 1995/09/28
Message-ID: <44ek0c$inp@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server Experimental '94 Hill:
(accessed with ";redcode-94x")

Hill Specs:
         coresize: 55440
   max. processes: 10000
         duration: after 500,000 cycles, a tie is declared.
max. entry length: 200
 minimum distance: 200
  instruction set: ICWS '94 Draft

Last challenge: Sun Sep  3 07:39:19 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  46/ 31/ 23            Illusion-94/55         Randy Graham     161       6
 2  43/ 31/ 26                Ptest 01 X          Beppe Bezzi     155       2
 3  32/ 21/ 47                   Aleph 1              Jay Han     144      51
 4  40/ 37/ 23               Stimpy v2.0     Brant D. Thomsen     144      44
 5  32/ 24/ 43                  Lucky 13        Stefan Strack     141     195
 6  42/ 46/ 11          I Can't Drive 55      Randy W. Graham     139      13
 7  29/ 22/ 49                 Blue Funk       Steven Morrell     135      43
 8  29/ 23/ 48 Der Zweite Blitzkrieg - 9      Mike Nonemacher     135     151
 9  30/ 26/ 44             Variation G-1              Jay Han     134     153
10  38/ 42/ 20                   Squeeky         Randy Graham     133      10
11  37/ 40/ 23                  ivscan6b            J.Layland     133      53
12  36/ 40/ 25        Miss Understanding           Derek Ross     132       4
13  27/ 23/ 50              NotSoBigImps        James Layland     130      49
14  19/  7/ 74          TimeScapeX (0.1)       J. Pohjalainen     130      15
15  36/ 42/ 22             Miss Careless           Derek Ross     129       1
16  35/ 41/ 24                     Fscan              Jay Han     129      37
17  39/ 51/ 10 still testing... big vers      Mike Nonemacher     128      14
18  37/ 51/ 12                 Rave B4.1        Stefan Strack     124     150
19  23/ 23/ 54    Marcia Trionfale 0.2 x          Beppe Bezzi     124       5
20  37/ 51/ 11                      Test        Stefan Strack     124      25

21  32/ 48/ 20              Genocide 94x      Mike Nonemacher     116      19

A new experiment will be unleashed October 1st.  By popular demand,
all hills will be increased from 100 to 250 round battles.  Also,
the -94x hill will be replaced with a new 5*50 round hill.  Submissions
to this hill will be mirrored from the -94 hill, no special switches
will be necessary.

--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "koth help" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: pizza@ecst.csuchico.edu (Internet Pizza Server)
Subject: Current Standings for Pizza's '94 Pizza Hill
Date: 1995/09/28
Message-ID: <44ek09$ino@charnel.ecst.csuchico.edu>#1/1
newsgroups: rec.games.corewar


For up to the second scores and info, visit the West Coast's premiere
corewar and KotH web site: http://www.ecst.csuchico.edu/~pizza/koth

Current Status of the Internet Pizza Server ICWS '94 Draft Hill:
(accessed with ";redcode-94")

Hill Specs:
         coresize: 8000
   max. processes: 8000
         duration: after 80,000 cycles, a tie is declared.
max. entry length: 100
 minimum distance: 100
  instruction set: ICWS '94 Draft

Last challenge: Thu Sep 28 09:40:20 PDT 1995 

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  36/ 30/ 33                 Torch t18              P.Kline     142       8
 2  37/ 34/ 29              Thermite 1.0        Robert Macrae     139     571
 3  35/ 30/ 35                   Eureka!    Maurizio Vittuari     139       2
 4  31/ 23/ 46           Jack in the box          Beppe Bezzi     138       3
 5  33/ 29/ 38                       Phq    Maurizio Vittuari     137     110
 6  33/ 32/ 34                        TR    Maurizio Vittuari     134      33
 7  39/ 44/ 17                  Agony II        Stefan Strack     134     848
 8  35/ 36/ 28              Simple Ptest           M R Bremer     134     138
 9  32/ 31/ 37                Brain Vamp  B.Bezzi, M.Paulsson     134      29
10  39/ 47/ 14                Bevo! v.1d       John Wilkinson     132     103
11  38/ 44/ 18              Taking Names              P.Kline     131      36
12  34/ 38/ 28               myVamp v3.5             Paulsson     130      68
13  33/ 37/ 30                   Fun v.f    John K. Wilkinson     129     111
14  15/  1/ 84                  Die Hard              P.Kline     129      93
15  24/ 20/ 56                    pStone              P.Kline     127       4
16  31/ 35/ 35                Fun v.g2d2    John K. Wilkinson     127      27
17  32/ 38/ 30               Armory - A5            Wilkinson     126     147
18  31/ 36/ 34             Brain Wash 11          Beppe Bezzi     126     113
19  35/ 45/ 20                 sdrawkcaB       Steven Morrell     126     163
20  13/ 12/ 76             theMystery1.5             Paulsson     113       1

21   0/  0/  5             theMystery1.4             Paulsson       5       6

A new experiment will be unleashed October 1st.  By popular demand,
all hills will be increased from 100 to 250 round battles.  Also,
the -94x hill will be replaced with a new 5*50 round hill.  Submissions
to this hill will be mirrored from the -94 hill, no special switches
will be necessary.

--
-------------+--------------------------------------------------------------
SPL   0, <-5 | Brought to you by the Internet Pizza Server!
MOV   1, >-6 |  
DAT >-7,  12 | Mail with a subject of "koth help" to pizza@ecst.csuchico.edu
END          | or try <http://www.ecst.csuchico.edu/~pizza/koth/> for info!



From: pan0178@comune.bologna.it (MV)
Subject: Re: (no subject)
Date: 1995/09/29
Message-ID: <199509292223.SAA18587@valhalla.stormking.com>#1/1
newsgroups: rec.games.corewar

>Hey, hey!  With the long-awaited arival of a replicator, my Bevo has
finally gotten
>some grass to munch on. :)  It actually pulled ahead of Agony!  Whoohoo!
>
>I feel just like Darth Vader when he killed Obi Wan... :)
>
>Of course, in the movie, Obi Wan came back more powerful than Darth could
imagine...
>Uh oh, that's a bad omen...
>
>-John
>
mmmhhh...
Every reference to my poor Rabbits is merely wanted! :-(
Never mind John 'Darth Vader' ! 
May the farce be with you!
...in the meantime I'll kill my warrior... :-P  
(hoping someday my Rabbits will drink your Bevo! ;-)

Maurizio




From: Beppe Bezzi <bezzi@iol.it>
Subject: Brain Wash
Date: 1995/09/29
Message-ID: <199509291935.UAA05787@iol-mail.iol.it>
newsgroups: rec.games.corewar

One month after their appearance on the hill, p-warriors are not dominating
it, even if they have such a powerful tool like p-space that should allow
them to score optimal results against all warrior classes. The reasons for
this may be many, I like to think one of these is Brain Wash.

Brain Wash uses p-space in active mode, using a simple switching routine to
try opposing enemy with best component, and uses p-scrambling to attack
enemy p-space not to do real damage, a simple MOD is enough to protect your
warrior, but to force it switching to a wrong strategy.

BW can switch beetween five attack modules: Firestorm, general purpose
incendiary bomber, paper, multipass sweep, Tornado, used only when paper
loses and the brainwasher that's the paper module with a brainwashing pit
attached; using a smart switching I can fit all in 100 lines and still have
some spacers.

The switcher is very simple: 
-Win    :-)     OK go on with same.
-Tie    :-|     Change to next; Firestorm or paper only.
-Loss   :-(
        -Paper or brainwasher lost -> enemy scanner -> Tornado bomber.
        -Firestorm or Tornado lost -> enemy paper or stone imp -> Brainwasher.

Simple and self braiwashing resistent.
Now the code:
;-----------------------
;redcode-94
;name Brain Wash 11
;author Beppe Bezzi
;strategy headache, bad thoughts, combat stress ?
;strategy you need a good Brain Wash :-)
;strategy v 11 little changes
;assert CORESIZE == 8000
;kill Brain Wash

org     think                   ;think first. My grandma told me often :-) 
_RES    equ     #0              ;last result
_STR    equ     #               ;-strategy in use
soap    equ     6               ;to put in opponents p-space

count   equ     750
step    equ     55               
away    equ                     ;boot distance
bcount  equ     #533
dest1   equ     3740
dest2   equ     1860

firest
bootgt  mov     gate,   away
	mov     gate+2, away 
booincr mov     incr,   away+23
	mov     bombm,  away+23

boot    mov     djmp,   away+8
	mov     clr,    <boot
	mov     bombs,  <boot
bjmp    mov     jump,   <boot
	mov     adder,  <boot
	mov     stone,  <boot
	mov     mov2,   <boot
	mov     mov1,   <boot
	jmp     @boot


gate    dat     #1,     #30
	dat     #1,     #30
	spl     #1,     #30
	dat     0,0
	dat     0,0
	dat     0,0
start   
mov1    mov     bombm,   @stone
mov2    mov     bombs,   *stone
stone   mov     2*step+1,*(2*step)+2
adder   add     incr    ,stone           
jump    djn.b   start,  #count 
bombs   spl     #step,  #-step+1
clr     mov     @1,     >gate
djmp    djn.b   clr,    {gate+2

b for 12 
	dat     0,0
rof

incr    dat     2*step, 2*step      
bombm   mov     step,   >step           

for 0
note that Firestorm, sweep and Tornado share some code and boot. This allows
me to save lot of space
Needs some work, done in part with Brain Vamp, and there is a small bug in
Tornado clear, but works.
rof
;--Tornado bomber setup
bomber  mov.ab  bcount, jump
	add.f   bombm,  incr
	mov.i   bombm2, bombm
	mov.i   bstone, stone
	stp.ab  #3,     _STR    ;if wins continue
	jmp     bootgt
bstone  mov     *step+2, *(2*step)+2
bombm2  mov     step,   1       ;This works very well; thanx Paul

;----Sweep setup
sweep
	mov.i   bjmp1,  bjmp
	mov.ab  #away-3,booincr         
	jmp     bootgt
bjmp1   jmp     @-3

;-- Think module
think
res     ldp.ab  _RES,   #0
str     ldp.a   _STR,   str1    ;load strategy in use
	sne.ab  #0,     res     ;check result
	jmp     bwash1          ;lost go for brainwash
	jmn.a   win,    @res    ;win ?
tie     add.a   #1,     str1    ;tie change strat
	mod.a   #3,     str1    ;secure result no Tornado
win     stp.ab  str1,   _STR
str1    jmp     @0,     paper   ;0
	dat     0,      firest  ;1
	dat     0,      sweep   ;2
	dat     0,      bomber  ;3

;-- Brainwashing and paper module
bwash1  slt.ab  str1,   #4      ;just brainwashed ?
	jmp     bomber          ;yes and lost, go for bomber
	jmz.a   bomber, str1    ;paper lost? yes -> bomber
bwash   stp.ab  #0,     _STR    ;no -> bwasher

        spl     1,      <1000   ;dobles processes in use to include bwash pit
paper
        spl     1,      <300    ;\
        spl     1,      <300    ;-> generate 7 consecutive processes
        mov     -1,     0

silk    spl     @0,     <dest1  ;split to new copy
        mov.i   }-1,    >-1     ;copy self to new location
        mov     bomba,  }1001
        mov     bomba,  }2001
        mov     {silk,  <silk2
silk2   jmp     @0,     >dest2
bomba   dat     <-2666, {1      ;multi use bomb, against imps and clears.
Bad with both :-(

brain   stp.ab  #soap,  _STR    ;fixed value seems to work better
wash    spl     brain,  <brain
        mov     bomba,  <brain
        spl     brain,  <brain
        mov     bomba,  <brain
        jmp     brain

for MAXLENGTH-CURLINE-5
	dat     0,0
rof
for 5
	dat     1,1            ;anti djn stream decoy
rof
end
;------------

Brain Wash works very well against static switchers (first test all
startegies then choose the best to finish the match) well against some
dynamic switchers (those testing for some round and switching if results are
not good enough) and fine against quick switchers (like BW itself) Not so
good against standard warriors, tied Agony 42/44/14 and HeremScimitar
44/45/11 (scoring high :-), and lost to both Torch 34/43/23 and Thermite
31/39/30
Now is going down, people are taking counter measures.

Last minute note: he is dead :-( I kept the article to add some tips on
brainwashing, then began discussion on it in another thread and now ...

He will raise back stronger than ever.

I'll publish it anymore, keeping EQUs secret is because some are the same of
Brain Vamp, I started so and they remained. Next BW they will differ.

It's a lot of time I don't see whole warriors published in the newsgroup,
and very many entered the hill, hope to be the first of row.


-Beppe Bezzi

<bezzi@iol.it>




From: Beppe Bezzi <bezzi@iol.it>
Subject: Warrior aging
Date: 1995/09/29
Message-ID: <199509291026.LAA00553@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

We are starting a new Pizza -94 hill with 5*50 rond matches; that could be
good for trying a different aging system.

Now if I send a new version of a warrior, killing the old one, every warrior
ages one even if none risked a bit to be pushed off; that's not IMO fair. A
warrior should age only if it survives a real attack to for having seen an
enemy replace another for a turnover.
We can make a warrior age only if the outgoing warrior has a score >100, in
such a way even if one kills some test warriors, clearing some slots, the
next warriors entering, without a fight, won't age till the next *real*
challenge.

I know there is not enough time for a debate, but we can try it; the new
hill has been made to compare score variability with different number of
rounds, we can make it a test for new aging system too.




From: jwilkinson@mail.utexas.edu
Subject: (no subject)
Date: 1995/09/29
Message-ID: <44h5tq$8ml@geraldo.cc.utexas.edu>#1/1
newsgroups: rec.games.corewar

Hey, hey!  With the long-awaited arival of a replicator, my Bevo has finally gotten
some grass to munch on. :)  It actually pulled ahead of Agony!  Whoohoo!

I feel just like Darth Vader when he killed Obi Wan... :)

Of course, in the movie, Obi Wan came back more powerful than Darth could imagine...
Uh oh, that's a bad omen...

-John




From: user@portal.dx.net (mneumenth)
Subject: Re: Warrior aging
Date: 1995/09/29
Message-ID: <44icaa$9bh@portal.dx.net>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it> <44ibs2$8n7@portal.dx.net>
newsgroups: rec.games.corewar

Please forgive my last post.  I meant the old warriors age, not 
score(repeated slaping and yelping sounds can be heard).

Mneumenth




From: user@portal.dx.net (mneumenth)
Subject: Re: Brain Wash
Date: 1995/09/29
Message-ID: <44ic6g$96a@portal.dx.net>#1/1
references: <199509291935.UAA05787@iol-mail.iol.it>
newsgroups: rec.games.corewar

I guess my idea of a shared P-Space was liked.  I remember that STANDARD 
P-Space isn't shared, and can't be scrambled without an enemy process.  
Or did I TOTALLY miss-read that original Protected-Space announcement 
post... And if you don't get that first sentence, than my newsreader 
wasn't the only hungry one that day.

Mneumenth, cyborg dragon of Earth, helper of Dominia and Doom newbies, 
and collector of music CD's by unknown artists and of movie soundtracks.
mneumenth@kernvalley.com





From: user@portal.dx.net (mneumenth)
Subject: Re: Warrior aging
Date: 1995/09/29
Message-ID: <44ibs2$8n7@portal.dx.net>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it>
newsgroups: rec.games.corewar

:In article <199509291026.LAA00553@iol-mail.iol.it>,
:Beppe Bezzi  <bezzi@iol.it> wrote:
:>We are starting a new Pizza -94 hill with 5*50 rond matches; that could be
:>good for trying a different aging system.
:>
:>Now if I send a new version of a warrior, killing the old one, every 
:warrior
:>ages one even if none risked a bit to be pushed off; that's not IMO fair. A
:>warrior should age only if it survives a real attack to for having seen an
:>enemy replace another for a turnover.
:>We can make a warrior age only if the outgoing warrior has a score >100, in
:>such a way even if one kills some test warriors, clearing some slots, the
:>next warriors entering, without a fight, won't age till the next *real*
:>challenge.
:>
:>I know there is not enough time for a debate, but we can try it; the new
:>hill has been made to compare score variability with different number of
:>rounds, we can make it a test for new aging system too.
:>

How about this, too.  If the name is the same, and there is a version 
statement, give the new version the old one's score.  I've had a couple 
of Beginner-hill warriors that I didn't upgrade, because I didn't want to 
have an "unfair advantage."  I decided against this, even though the new 
ones were VASTLY superior, because I figured that that's like submitting 
a warrior multiple times.  Do it enough, and the competition gets knocked 
off.

Mneumenth, cyborg dragon of Earth, helper of Dominia and Doom newbies, 
and collector of music CD's by unknown bands and of movie soundtracks.
mneumenth@kernvalley.com





From: jwilkinson@mail.utexas.edu
Subject: Re: (no subject)
Date: 1995/09/29
Message-ID: <44i02v$ruh@geraldo.cc.utexas.edu>#1/1
references: <199509292223.SAA18587@valhalla.stormking.com>
newsgroups: rec.games.corewar

pan0178@comune.bologna.it (MV) wrote:
>>Hey, hey!  With the long-awaited arival of a replicator, my Bevo has finally gotten
>>some grass to munch on. :)  It actually pulled ahead of Agony!  Whoohoo!
>>
>>I feel just like Darth Vader when he killed Obi Wan... :)
>>
>>Of course, in the movie, Obi Wan came back more powerful than Darth could imagine...
>>Uh oh, that's a bad omen...
>>
>Every reference to my poor Rabbits is merely wanted! :-(
>Never mind John 'Darth Vader' ! 
>May the farce be with you!

I don't know about the farce, but I see your schwarz is as big as mine!

>...in the meantime I'll kill my warrior... :-P  
>(hoping someday my Rabbits will drink your Bevo! ;-)

ACK!  I checked the standings!  You -did- kill Rabbits!

And now Agony has pulled ahead again. :(

Ah well... Bevo had his day in the sun.  I'm hoping that more replicators will
appear soon... Bevo is still hungry. :)

-John-





From: jwilkinson@mail.utexas.edu
Subject: Re: Warrior aging
Date: 1995/09/29
Message-ID: <44hvme$ruh@geraldo.cc.utexas.edu>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it>
newsgroups: rec.games.corewar

>We can make a warrior age only if the outgoing warrior has a score >100, in
>such a way even if one kills some test warriors, clearing some slots, the
>next warriors entering, without a fight, won't age till the next *real*
>challenge.
>
>I know there is not enough time for a debate, but we can try it; the new
>hill has been made to compare score variability with different number of
>rounds, we can make it a test for new aging system too.

Sorry to burst your bubble there, Beppe.  But, according to your system,
no warrior would ever make it past age 0. :)

However, I agree, if the warrior which fell of the Hill has been 'kill'ed,
then no other warriors should age.

-John-







From: jwilkinson@mail.utexas.edu
Subject: Re: P-space
Date: 1995/09/29
Message-ID: <44hjse$j3t@geraldo.cc.utexas.edu>#1/1
references: <199509272225.XAA12771@iol-mail.iol.it> <1995Sep28.104806@acad.drake.edu> <44f8m7$sv3@geraldo.cc.utexas.edu> <1995Sep29.115231@acad.drake.edu>
newsgroups: rec.games.corewar

>> I think that's a bad idea.  If str1 is good against the brainwasher, then the
>> switching mechanism will find that str1 is good, and it will use str1.
>> 
>> However, if str1 is bad, you are stuck using it over and over and over... in
>> a loop, with only 2/9 chance of getting out.

>Well, this code only gives some protection against random-number brainwashing.
>
>The idea is that if you are brainwashed, you can still retain some control
>over what strategy you select.  If that strategy works well then great,
>if not, then too bad.  But without some built-in protection, random
>brainwashing causes you to select randomly among your strategies, bad
>and good.

Ah, but if your warrior contained a single strategy that could beat the
brainwasher, then as soon as the brainwasher brainwashed you into using
that strategy, then you would then beat the brainwasher... and it couldn't
wash you any more?

If you've got a warrior that you're likely to switch to after a brainwash,
then after the washer gets you, it would be likely to get you again and again.

Hmmm... I don't see what condition it would be helpful under... do you?

>An intelligent brainwasher could probe your strategies by washing your
>pspace with zero, one, two, etc., and see how you perform against 
>his own programs.  Then select a washing number and appropriate strategy
>for the rest of the battles.

Oooh... that'd be nifty.  Of course, would it be -successful-?

>A simpler model would make washing a natural part of all your strategies,
>using zero as a wash number.  This would force your opponent to act like
>a non-intelligent program with whatever strategy he has at zero.  He 
>could respond by putting something safe like Die Hard at zero - even
>better he could put Die Hard in nine positions and something else in
>the other one, making it hard for you to force him out :-)  I feel
>like I'm rambling...

Exactly, if I had Die Hard in my corner, that's what I'd do. :)

-John-










From: pk6811s@acad.drake.edu
Subject: Re: P-space
Date: 1995/09/29
Message-ID: <1995Sep29.115231@acad.drake.edu>#1/1
references: <199509272225.XAA12771@iol-mail.iol.it> <1995Sep28.104806@acad.drake.edu> <44f8m7$sv3@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

John <jwilkinson@mail.utexas.edu> writes:

> I think that's a bad idea.  If str1 is good against the brainwasher, then the
> switching mechanism will find that str1 is good, and it will use str1.
> 
> However, if str1 is bad, you are stuck using it over and over and over... in
> a loop, with only 2/9 chance of getting out.

Well, this code only gives some protection against random-number brainwashing.

The idea is that if you are brainwashed, you can still retain some control
over what strategy you select.  If that strategy works well then great,
if not, then too bad.  But without some built-in protection, random
brainwashing causes you to select randomly among your strategies, bad
and good.

An intelligent brainwasher could probe your strategies by washing your
pspace with zero, one, two, etc., and see how you perform against 
his own programs.  Then select a washing number and appropriate strategy
for the rest of the battles.

A simpler model would make washing a natural part of all your strategies,
using zero as a wash number.  This would force your opponent to act like
a non-intelligent program with whatever strategy he has at zero.  He 
could respond by putting something safe like Die Hard at zero - even
better he could put Die Hard in nine positions and something else in
the other one, making it hard for you to force him out :-)  I feel
like I'm rambling...

> Why are you switching Die Hard and a stone?  Die Hard seams to beat all the same
> things that a stone beats.  That seems like a fooish choice...  Logically, you
> would want to run a scanner or scissors to kill things that Die Hard can't kill,
> right? I mean, that is what Pink was doing... And it was riding on top of the hill
> until you switched it up and pulled it off...

Think of pStone as a stone with a backup plan.  Simple fast p-code.

Die Hard gets about 20% wins against Agony and Taking Names.  The stone
gets about 50% wins.  And I'm working on trying other components that
could deal with replicators.  Actually Pink's performance was falling
off terribly as more pspacers appeared.  It was based on running trials
of a few battles and some of the pspacers are switching faster than that.

Paul Kline
pk6811s@acad.drake.edu



From: jwilkinson@mail.utexas.edu
Subject: Re: Warrior aging
Date: 1995/09/30
Message-ID: <44k2m6$7hv@geraldo.cc.utexas.edu>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it> <44ibs2$8n7@portal.dx.net> <44id2r$79f@geraldo.cc.utexas.edu> <44k08u$sn2@portal.dx.net>
newsgroups: rec.games.corewar

>No, that's sorta cheesy...  I mean, you could submit a scanner, and 
>when all the papers start falling off the hill, then just submit a stone with the
>same name.  I mean, that's an extreme example, but the whole point of 
>giving a score to a warrior, is that it is ONE warrior that got the score... 
>isn't it? 
>-John-

>I guess your newsreader ate the following post for lunch.  I mistyped 
>score when I meant age.  Say I have a program of age 279, and I make some 
>improvements.  I don't want to start over at age 0.

No, I understood.  If you submit a change to a warrior, that warrior should
lose it's score.  Because, if you change it, it is a new warrior, and thus
might not have been able to achieve the previous warrior's score.  I think
it's axiomatic...

Of course the idea of 'score' might be improved past simply the age of the
warrior.  Perhaps we could give a warrior one point if they are in slot
twenty, two if they are second from the bottom, ... , and 20 points if
they are in the first slot?

-John-




From: user@portal.dx.net (mneumenth)
Subject: Re: Warrior aging
Date: 1995/09/30
Message-ID: <44k08u$sn2@portal.dx.net>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it> <44ibs2$8n7@portal.dx.net> <44id2r$79f@geraldo.cc.utexas.edu>
newsgroups: rec.games.corewar

:In article <44id2r$79f@geraldo.cc.utexas.edu>,
: <jwilkinson@mail.utexas.edu> wrote:
:>user@portal.dx.net (mneumenth) wrote:
:>
:>>How about this, too.  If the name is the same, and there is a version 
:>>statement, give the new version the old one's score.  I've had a couple 
:>>of Beginner-hill warriors that I didn't upgrade, because I didn't want to 
:>>have an "unfair advantage."  I decided against this, even though the new 
:>>ones were VASTLY superior, because I figured that that's like submitting 
:>>a warrior multiple times.  Do it enough, and the competition gets knocked 
:>>off.
:>>
:>>mneumenth@kernvalley.com
:>
:>No, that's sorta cheesy...  I mean, you could submit a scanner, and 
:when all
:>the papers start falling off the hill, then just submit a stone with the
:>same name.  I mean, that's an extreme example, but the whole point of 
:giving
:>a score to a warrior, is that it is ONE warrior that got the score... 
:isn't it? >
:>-John-
:>I guess your newsreader ate the following post for lunch.  I mistyped 
score when I meant age.  Say I have a program of age 279, and I make some 
improvements.  I don't want to start over at age 0.

Mneumenth, cyborg dragon of Earth, helper of Dominia and Doom newbies, 
and collector of music CD's by unknown bands and of music soundtracks.
mneumenth@kernvalley.com






From: jwilkinson@mail.utexas.edu
Subject: Re: Warrior aging
Date: 1995/09/30
Message-ID: <44id2r$79f@geraldo.cc.utexas.edu>#1/1
references: <199509291026.LAA00553@iol-mail.iol.it> <44ibs2$8n7@portal.dx.net>
newsgroups: rec.games.corewar

user@portal.dx.net (mneumenth) wrote:

>How about this, too.  If the name is the same, and there is a version 
>statement, give the new version the old one's score.  I've had a couple 
>of Beginner-hill warriors that I didn't upgrade, because I didn't want to 
>have an "unfair advantage."  I decided against this, even though the new 
>ones were VASTLY superior, because I figured that that's like submitting 
>a warrior multiple times.  Do it enough, and the competition gets knocked 
>off.
>
>mneumenth@kernvalley.com

No, that's sorta cheesy...  I mean, you could submit a scanner, and when all
the papers start falling off the hill, then just submit a stone with the
same name.  I mean, that's an extreme example, but the whole point of giving
a score to a warrior, is that it is ONE warrior that got the score... isn't it?

-John-




From: ruhl@phoebe.cair.du.edu (Robert A. Uhl)
Subject: Re: P-space
Date: 1995/09/30
Message-ID: <44i33l$91h@hermes.cair.du.edu>#1/1
references: <199509290128.CAA01797@iol-mail.iol.it>
newsgroups: rec.games.corewar

In article <199509290128.CAA01797@iol-mail.iol.it>,
Beppe Bezzi  <bezzi@iol.it> wrote:
>
>>until then to reappear.  Sort of a crawdad lifestyle.
>                                    ^^^^^^^
>What's a crawdad ? Just hope nothing sheek ;-) (got crazy to figure out what
>you meant)

  A crawdad is an American creature also known as a crayfish or
crawfish.  It looks a lot like a 3 inch long lobster (no, I will not
convert that to metric).  They are delicious.  The expression comes
from the fact that crayfish are always hiding.

-- 
+------------------------------------------------------------------------+
| Bob Uhl | Spectre                  | `En touto nika' +                 |
| U of D  | PGG FR No. 42            | http://mercury.cair.du.edu/~ruhl/ |
+------------------------------------------------------------------------+



From: y91magpa@ida.liu.se (Magnus Paulsson)
Subject: How to cheat
Date: 1995/09/30
Message-ID: <y91magpa.812501524@news>#1/1
newsgroups: rec.games.corewar

Have a look at this warrior that I wrote when thinking of
the part of the score that you get from fighting yourself.

;How to score 150 against your own wariors!
pin		111
org		check
check		ldp.a	#0,ptr
		jmp	ptr
		dat	0,start
ptr		jmp	@0,loss		; make use of the asymetry of
		dat	0,win		; one warrior losing and one winning!
		dat	0,start

loss		stp.ab	#843,#952
		ldp.ab	#773,#0
		seq.ab	loss,-1
		jmp 	start
		dat		0
win		stp.ab	#843,#773
		ldp.ab	#952,#0
		seq.ab	loss,-1
		jmp 	start
		dat		0
start		djn.f   0,<-1

Put something like this in your warior and it won't tie as
much against it self. 
(I'm not sure that the score agains oneself counts on the hill)

This could give you a point or so on the hill.
(I used this in myVamp 3.8 but then itscored a lot less than 3.7, so ????)
(At first I had a bug that made it fall of the hill, leaving the 3.7
that I tried to kill of :-), that bug is still there)

I haven't tested the idea of putting another warior on the hill with
the same PIN, but if you have a couple of warriors on the hill!

Somehow this must be changed, at least on the one warior hills.

/Magnus Paulsson








From: max@alcyone.darkside.com (Erik Max Francis)
Subject: Re: RedGen Source - Evolving Corewar Warriors
Date: 1995/09/30
Message-ID: <0Tc0BD1w200w@alcyone.darkside.com>#1/1
references: <44c5n4$dhu@hermes.cair.du.edu>
newsgroups: comp.ai.alife,rec.games.corewar

ruhl@phoebe.cair.du.edu (Robert A. Uhl) writes:

>   But if the compiler company allows shareware, which costs, then what
> would it have against PD, which is free, or freeware?

Freeware and public domain are completely different things.  Freeware 
means that you are free to do with the materials what you like, but 
the original author still holds the copyright.  (This means that you 
can't modify them and rerelease them as your own.)  Public domain 
means that you waive your copyright on the material.  (This means that 
someone _can_ rerelease the materials as their own.)

Even if you use your definition where freeware is the same as public 
domain (which I reiterate that it is not), I don't see how this 
argument is supposed to hold water.  First, what compiler are you 
talking about -- the redcode compiler, or the compiler in which the 
MARS[-like] virtual machine was written in?  Second, how do you expect 
the license that comes with the compiler to affect what is written 
with it?  (There are always clauses which distinguish between the 
compiler itself and what you write with it.)


Erik Max Francis, &tSftDotIotE // uuwest!alcyone!max, max@alcyone.darkside.com
San Jose, CA, USA // 37 20 07 N  121 53 38 W // GIGO, Omega, Psi // the 4th R!
H.3`S,3,P,3$S,#$Q,C`Q,3,P,3$S,#$Q,3`Q,3,P,C$Q,#(Q.#`-"C`- // 1love // folasade
_Omnia quia sunt, lumina sunt._ // mc2? oo?  Nah. // http://www.spies.com/max/



From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: P-space
Date: 1995/09/30
Message-ID: <199509301210.NAA06682@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

>
>>An intelligent brainwasher could probe your strategies by washing your
>>pspace with zero, one, two, etc., and see how you perform against 
>>his own programs.  Then select a washing number and appropriate strategy
>>for the rest of the battles.
>
>Oooh... that'd be nifty.  Of course, would it be -successful-?
>
A third-generation brainwasher, just an idea don't know if one can implement
it and if it would ever work, can vamp you, force you to scan your p-space
till you find a nonzero value, read and store it and, from then on, he can
braiwash you with but a simple bomb, and use itself different strategies to
attack and brainwash. 
Hypnotism, more than Brainwashing :-)


>>A simpler model would make washing a natural part of all your strategies,
>>using zero as a wash number.  This would force your opponent to act like
>>a non-intelligent program with whatever strategy he has at zero.  He 
>>could respond by putting something safe like Die Hard at zero - even
>>better he could put Die Hard in nine positions and something else in
>>the other one, making it hard for you to force him out :-)  I feel
>>like I'm rambling...
>
>Exactly, if I had Die Hard in my corner, that's what I'd do. :)
>
Die Hard is getting a bit softer :-) Its losses are growing a bit. 

And what about others warriors? You can defend yourself from Brainwashing,
but doing so you pay to others; your stategic part will grow in size and/or
boot time for Thermite pleasure. Sounds good, the balance is kept, no single
strategy will dominate the hill.

>>Paul Kline
>-John-
>
-Beppe











From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: Warrior aging
Date: 1995/09/30
Message-ID: <199509301210.NAA06678@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

John Wilkinson wrote:

>>We can make a warrior age only if the outgoing warrior has a score >100,=
 in
>Sorry to burst your bubble there, Beppe.  But, according to your system,
>no warrior would ever make it past age 0. :)
>
Why ? Maybe I didn't explain well, look at this example:

5 slots hill filled of new warriors, none has aged till now, they are the
first 5 warriors submitted.

Warrior         score           age
Alfa            150             0
Bravo           145             0
Charlie         140             0
Delta           135             0
Echo            130             0

Unknown         0               0       This is the last 21=B0 at Pizza

Now I submit Foxtrot warrior and it scores more than Echo, the hill will
look something so:

Warrior         score           age
Alfa            150             1       warriors aged because Echo score is
> 100
Bravo           145             1
Charlie         140             1
Delta           135             1
Foxtrot         132             1

Echo            130             0

And so on ..

If now I kill Delta to submit Delta2 the old Delta will score very low, 5-10
points at Pizza, and  Delta2 will enter without changing others age, amd
starting at age 0, he still has no merit for being here.

Using warrior names is not fair because there is no control that a warrior
named Gorilla 2 is an evolution of Gorilla 1, and in any case it's another
warrior. Should Stefan submit Agony III he has to restart from zero not from=
 850

>However, I agree, if the warrior which fell of the Hill has been 'kill'ed,
>then no other warriors should age.
>
>-John-
>
-Beppe




From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: P-space
Date: 1995/09/30
Message-ID: <199510010052.BAA10724@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

Jonathan West wrote:

>
>What's brainwashing have to do with corewars? How does one 
>program "Brainwash" another?
>
Some programs use p-space, a protected memory area, to store results of
previous rounds and other informations, useful to change strategy according
to the outcome of the match.

A Brainwasher manage to force another program to execute instructions (STP)
that fill of garbish this area, so as the 'intelligent' program get confused
and uses wrong strategies against him. 

This can be accomplished in some different ways, there have been some
articles in the newsgroup, the common thing is that you have to use an
opponent's process because you cannot directly access enemy p-space.

Published brainwashing warriors are my Brain Wash and Steven Morrell's PGP


-Beppe Bezzi

<bezzi@iol.it>
Gaudeamus igitur ...




From: Beppe Bezzi <bezzi@iol.it>
Subject: Re: Newbie Q: what is ;assert?
Date: 1995/09/30
Message-ID: <199510010053.BAA10730@iol-mail.iol.it>#1/1
newsgroups: rec.games.corewar

>>Sorry for the newbie question, but I can't seem to figure out what
>>;assert is for:
>>
>>Warning:
>>        Missing ';assert'. Warrior may not work with the current
>>setting
>>Number of warnings: 1
>>
>I've been playing for a while now, and I still don't get the point of
>that thing myself. :)
>
>Most of my warriors have ";assert 1" as the line right under the ";strategy"
>lines.  That'll get rid of the error message, if you want to do it.
>
;assert is a condition of the environment that has to be met for the warrior
to function properly.

For example, if you have imps you must have a CORESIZE == 8000 assert line,
so that, if you submit your warrior to the wrong hill the compiler aborts
and warns you.

More examples are in p-mars.doc

-Beppe

>-John
>




From: Jonathan West <73742.1715@CompuServe.COM>
Subject: Re: P-space
Date: 1995/09/30
Message-ID: <44kccm$p1e$2@mhafm.production.compuserve.com>#1/1
references: <1995Sep28.104806@acad.drake.edu>
newsgroups: rec.games.corewar

What's brainwashing have to do with corewars? How does one 
program "Brainwash" another?

-- 
This is a .sig virus. Please copy me into your signature



From: Jonathan West <73742.1715@CompuServe.COM>
Subject: Re: What is a process?
Date: 1995/09/30
Message-ID: <44kc7h$p1e$1@mhafm.production.compuserve.com>#1/1
references: <447cd2$ks6@allnews.infi.net>
newsgroups: rec.games.corewar

Here is my Definition:

"A procces is a thing."

-- 
This is a .sig virus. Please copy me into your signature



From: jwilkinson@mail.utexas.edu
Subject: Re: Newbie Q: what is ;assert?
Date: 1995/09/30
Message-ID: <44k6cv$aa5@geraldo.cc.utexas.edu>#1/1
references: <44k567$7gc@netnews.upenn.edu>
newsgroups: rec.games.corewar

>Sorry for the newbie question, but I can't seem to figure out what
>;assert is for:
>
>Warning:
>        Missing ';assert'. Warrior may not work with the current
>setting
>Number of warnings: 1
>
I've been playing for a while now, and I still don't get the point of
that thing myself. :)

Most of my warriors have ";assert 1" as the line right under the ";strategy"
lines.  That'll get rid of the error message, if you want to do it.

-John




From: dherbst@gradient.cis.upenn.edu (Darrel Herbst)
Subject: Newbie Q: what is ;assert?
Date: 1995/09/30
Message-ID: <44k567$7gc@netnews.upenn.edu>#1/1
newsgroups: rec.games.corewar

Sorry for the newbie question, but I can't seem to figure out what
;assert is for:

Warning:
        Missing ';assert'. Warrior may not work with the current
setting
Number of warnings: 1





From: y91magpa@ida.liu.se (Magnus Paulsson)
Subject: Die Hard
Date: 1995/09/30
Message-ID: <y91magpa.812470282@news>#1/1
newsgroups: rec.games.corewar

Program "Die Hard" (length 90) by "P.Kline"
(contact address "PDK@ADMIN.DRAKE.EDU"):
;strategy minimal kill+, maximal survival
myVamp v3.7 wins: 49
Die Hard wins: 27
Ties: 24
 
Naah, not that hard ;-)
Sorry, just had to brag a bit.

/Paulsson