Article 1723 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!cs.utexas.edu!zaphod.mps.ohio-state.edu!howland.reston.ans.net!ux1.cso.uiuc.edu!news.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s
From: pk6811s@acad.drake.edu
Subject: _Push Off_
Message-ID: <1993Apr7.091844.1@acad.drake.edu>
Lines: 206
Sender: news@dunix.drake.edu (USENET News System)
Nntp-Posting-Host: acad.drake.edu
Organization: Drake University, Des Moines, Iowa, USA
Date: Wed, 7 Apr 1993 15:18:44 GMT

                                _PUSH OFF_
                        A midweek review of Corewar
                              April 7, 1993
-------------------------------------------------------------------------------
  I.  The Standings:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  50/ 41/  9                 Agony 5.1        Stefan Strack     158     116
 2  48/ 40/ 12              Dragon Spear             c w blue     155      64
 3  47/ 44/  9               Medusa's v7   Mintardjo & Strack     150     255
 4  46/ 43/ 11                Eclipse II              P.Kline     148     232
 5  45/ 43/ 12                 Iron Gate       Wayne Sheppard     148     918
 6  45/ 44/ 12                      Zipp             c w blue     145      18
 7  46/ 47/  7           Juggernaut v1.4         Anders Ivner     144      14
 8  36/ 28/ 36                Imprimis 6              P.Kline     144     362
 9  44/ 43/ 13           Paratroops v2.1         W. Mintardjo     144     323
10  38/ 36/ 27                    Nurgle             c w blue     139      54
11  39/ 41/ 21             Leprechaun 1b         Anders Ivner     137     874
12  32/ 28/ 40                  Midnight       Wayne Sheppard     137       6
13  32/ 26/ 42                        It        nandor sieben     137    1272
14  32/ 29/ 39           +0 Stormbringer       Dan Nabutovsky     136    1392
15  32/ 29/ 39               Sphinx v2.8         W. Mintardjo     136     960
16  40/ 45/ 15                 Emerald 4              P.Kline     134      21
17  39/ 46/ 15                  Sucker 5        Stefan Strack     133    1256
18  38/ 43/ 19               Moonstone 1       Dan Nabutovsky     133     186
19  38/ 44/ 17                  Herem II         Anders Ivner     132     736
20  30/ 32/ 38                      test              Unknown     129       1

21  21/ 44/ 36                  Demon2.2         Cormac Walsh      97       0

-------------------------------------------------------------------------------
 II.  The Basics:

       -Core War Archives are available via anonymous FTP at 
        soda.berkeley.edu in pub/corewar...

       -FAQ for this newsgroup is available via anonymous FTP at
        rtfm.mit.edu as pub/usenet/news.answers/games/corewar-faq.z

-------------------------------------------------------------------------------
III.  The Scoop:

Spring Break seems to mean something other than beaches and bars for 
KotH competitors.  The action has been furious!  In the section below,
The Quick Look, submissions are listed which fell short, or made it on
the Hill in 19/20th position (or so).  The number grows and grows, and
what is not shown is the number of _duplicate_ submissions - I attempt
to show only the highest score achieved.  What a great game!  A. K. 
Dewdney - you're keeping teenagers off the streets!  [course it has
been a while since SOME of us have seen teenage-hood :-)]

Agony and Dragon Spear have been taking turns in first place, both
scoring 50%+ wins at times.  Did I say something about not seeing 50% wins
for a while?  Check out this top five:
 1  52/ 40/  8                 Agony 5.1        Stefan Strack     165      69
 2  50/ 39/ 11              Dragon Spear             c w blue     162      17
 3  41/ 21/ 38                        It        nandor sieben     161    1225
 4  50/ 39/ 11                 Iron Gate       Wayne Sheppard     160     871
 5  50/ 42/  8               Medusa's v7   Mintardjo & Strack     157     208
Of course there were a couple of under-100 scores at the other end of 
the Hill :-)  [How did It get up there with all those scanners?]

Sometime it would be interesting to hear where these fighter-names come
from.  I finally figured out Agony and XTC - but it took me a while.
But 'Nurgle'?  'Herem'?

Not to be outdone in the Corewars emulator club, Albert Ma released a
new ms-dos version which, according to early reviews, is a screamer.
Thanks Ma - er, um, A.M.!

W. Sheppard didn't waste anytime getting re-connected.  He's been busy
hammering away from his new residence:
 21  32/ 57/ 11                  I'm Back       Wayne Sheppard     107 

Unfortunately Wayne got back just in time to see his favorite, long-running
stone/imp-spiral program get knocked off:
 21  29/ 29/ 42                     Night       Wayne Sheppard     128     989
Just short of a thousand too, what a shame :-)

D. Nabutovsky's been warming up some new fighters - maybe this had
something to do with it:
 19  32/ 28/ 40           +0 Stormbringer       Dan Nabutovsky     136    1388
 20  38/ 42/ 20               Moonstone 1       Dan Nabutovsky     134     182
which came back from E.J. Andrews' 'Paper-warrior' submission.  Yup,
paper warriors (replicators) definitely are hard on stones like Moonstone,
and imps can't beat them either - only tie.

And yours truly woke up and found a bug in Emerald 3, which made a _major_
difference in its performance.  Turns out that this is NOT a gate:
    spl 0,<-5
    dat <-7,#0
Now all you friendly Corewars-types had a chance to run Emerald 3, how come
you didn't point this out to me?  Along with this fix, I arranged things so
instead of bombing the anti-vamp component with a dat #0, it is bombed with
a 'jmp -1' which makes a faster core-clear.  Emerald 4 has stayed up a week
so far - hopefully some more bombers can make it up, then maybe we can get
back to my favorite warrior form - replicators [I've given up on self-
repairing fighters :-(]

-------------------------------------------------------------------------------
 IV.  The Outlook:

 2  39/ 20/ 41                G Paper /T         W. Mintardjo     159       1
 9  36/ 24/ 40                 test 8b@s              Unknown     148       1
 6  45/ 43/ 13                      Zipp             c w blue     147       1
10  31/ 26/ 43               G Paper /T2         W. Mintardjo     137       1
 6  43/ 45/ 12             Iron Gate 1.1       Wayne Sheppard     142       1

-------------------------------------------------------------------------------
  V.  The Quick Look:

21  27/ 49/ 24                         j            j.layland     104       0
21  20/ 71/  9                         r            j.layland      69       0
20  30/ 50/ 20                        v5            j.layland     109       1
21   5/ 82/ 13                       Get      Andre van Dalen      28       0
20  16/ 47/ 37                       TS3         W. Mintardjo      85       1
21   9/ 85/  6                       Try      Andre van Dalen      33       0
21  15/ 71/ 14                      Coin              Mestern      60       0
21  12/ 77/ 12                      Get4      Andre van Dalen      46       0
20  33/ 48/ 20                      Harm             c w blue     117       1
21  24/ 75/  1                      Orff              Fredrik      73       0
21   7/ 39/ 54                      Zipp             c w blue      75       0
21  20/ 77/  3                      bitz           Paul Bobby      62       0
19  26/ 52/ 22                      rock            j.layland     101       1
20  31/ 31/ 38                      test              Unknown     130       1
21   4/ 64/ 32                      test         Anders Ivner      44       0
21  32/ 59/  8                     seek1           Sasha Wait     105       0
21  26/ 42/ 32                     stone           Sasha Wait     111       0
20  16/ 56/ 28                     venio           Sasha Wait      75       1
21  11/ 77/ 12                    Dwarfs      Andre van Dalen      45       0
21  13/ 80/  7                    Test 4      Sami Tammilehto      46       0
19  36/ 46/ 18                    Zoiks!        C. Parrinello     125       1
21  25/ 59/ 15                    seek1b           Sasha Wait      92       0
21  12/ 86/  2                    seek1d           Sasha Wait      38       0
21   2/ 59/ 38                   BACKIMP           Paul Bobby      45       0
21  28/ 70/  2                   Blofeld              Fredrik      86       0
21  36/ 53/ 11                   Cleaver       Wayne Sheppard     120       0
21   7/ 53/ 40                   CraMPon             c w blue      61       0
21  18/ 62/ 20                   DemonII         Cormac Walsh      74       0
21   2/ 72/ 27                   GTZ 1.0       James Jesensky      31       0
20   2/ 35/ 62                   No Hope          Jeff Peters      69       1
21  11/ 81/  8                   Stapper      Andre van Dalen      41       0
21  17/ 66/ 17                   Sweeper        E. J. Andrews      67       0
21  17/ 69/ 14                   splits1           Sasha Wait      65       0
21  25/ 61/ 14                   veniont           Sasha Wait      88       0
21  28/ 50/ 23                  Buzzbomb           Paul Bobby     106       0
21  22/ 43/ 35                  Demon2.2         Cormac Walsh     102       0
20  28/ 49/ 23                  Orff 2.0              Fredrik     107       1
21  19/ 42/ 39                  Rabbit 1              Mestern      97       0
20  26/ 69/  5                  Roadkill          Jeff Peters      83       1
21   8/ 74/ 17                  fire 1.0       James Jesensky      42       0
21  13/ 60/ 27                  splits1b           Sasha Wait      66       0
21  30/ 49/ 21                  stone2dd           Sasha Wait     111       0
21  18/ 82/  0                 Blah v1.0           Paul Bobby      54       0
21  12/ 58/ 31                 Brimstone      Sami Tammilehto      66       0
21  33/ 49/ 18                 Discovery        David Johnson     116       0
21  27/ 52/ 21                 Simplex-9      Sami Tammilehto     102       0
21   1/ 47/ 52                 pepper1.1        Hank Turowski      54       0
21   1/ 44/ 55                No Hope v2          Jeff Peters      58       0
21  25/ 42/ 34                Oculomotor              Fredrik     108       0
20  31/ 62/  7                Orff 1.1.1              Fredrik     101       1
21  11/ 81/  8                Pepper 1.5        Hank Turowski      40       0
21  14/ 80/  6                sub-type-c             c w blue      47       0
11  32/ 25/ 43                test 87b@s              Unknown     140       1
21  38/ 53/  9               Light Speed       Wayne Sheppard     123       0
19  35/ 46/ 19               Meteor v4.0         W. Mintardjo     125       1
21   1/ 68/ 30              ImpBreed-1.1         Jonathan Roy      34       0
21  12/ 67/ 21              Parasite v 4      Roderick Easton      58       0
21   9/ 46/ 45              Pirates v1.1     Brant D. Thomsen      71       0
19  41/ 54/  5              Stoned Again             c w blue     128       1
21   6/ 71/ 23              SuperImp 1.0         Jonathan Roy      41       0
19  39/ 49/ 12              sub-type-cmp             c w blue     130       1
20   7/ 60/ 33             Brimstone 1.2      Sami Tammilehto      55       1
21   0/ 50/ 49             ImpLance2-1.0         Jonathan Roy      50       0
21   0/ 76/ 24             ImpLance3-1.0         Jonathan Roy      24       0
21   0/ 81/ 19             ImpLance4-1.0         Jonathan Roy      20       0
21   0/ 84/ 16             ImpLance5-1.0         Jonathan Roy      16       0
21  30/ 60/ 10             Simple-Bomber        E. J. Andrews     101       0
21  30/ 58/ 12            BscannerBomber       Wayne Sheppard     102       0
21  10/ 61/ 28            side-pipes 1.0       James Jesensky      59       0
21   4/ 54/ 43            superdwarf 1.0     Cliff Fitzmorris      54       0
21   1/ 69/ 29           DeathLance2-1.0         Jonathan Roy      33       0
21   1/ 92/  7           ImpLance2-1.0-t         Jonathan Roy       9       0
21  39/ 59/  2           No Ties Allowed       Wayne Sheppard     120       0
20   1/ 64/ 35        Self splitting imp         W. Mintardjo      38       1
20  29/ 68/  3       Thundering Buttucks       Jordan Horwich      91       1

-------------------------------------------------------------------------------
 VI.  The Hint:

The ADD instruction is used by many of the top programs to maintain their
slim figures (size).  This is because it adds both the A- and B-operands of the
two designated locations.  Thus with one ADD, a cmp-scanner can modify both
locations of its comparison, a bomber can set up one location for decrement
and one for bomb, and a vampire can create the appropriate jmp instruction for
its target location.  What other instructions act on both the A- and B-
operands?

Another slimming strategy is to use unrequired B-operands for storing
pointers or other data.  What instructions do not use their B-operand?

-------------------------------------------------------------------------------
VII.  The End:

Paul Kline
pk6811s@acad.drake.edu


Article 1732 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!cs.utexas.edu!wupost!uunet!vuse.vanderbilt.edu!stst
From: stst@vuse.vanderbilt.edu (Stefan Strack)
Subject: System validation
Message-ID: <C59Bt0.2Ko@vuse.vanderbilt.edu>
Sender: news@vuse.vanderbilt.edu
Nntp-Posting-Host: necs
Organization: Vanderbilt University School of Engineering, Nashville, TN, USA
Date: Sat, 10 Apr 1993 07:36:35 GMT
Lines: 129


The program below can be used to determine whether your corewar interpreter
complies with ICWS88 and evaluates operands "in-register" (i.e. is compatible
with KotH). It is based on Mark Durham's "Validation Suite" collection of
code sniplets and adds some timing and core initialization tests.

Results for the interpreters in soda's system directory that I could test:

c88v320		- passes
mars88	        - passes
mercury		- passes
KotH v3.1	- passes
kothpc		- fails (initializes with DAT #0,#0, mixed evaluation modes)
corwp302	- fls (initializes with DAT #0,#0)

Speaking of which, would somebody volunteer to put together a comparative
review of interpreters for one or more of the platforms (PC,Mac,Amiga,UNIX)?

This would become part of the FAQ or a separate regular posting. Contact me
for details. (You should not be a system author for obvious reasons :-)

Regards, Stefan (stst@vuse.vanderbilt.edu)

;redcode
;name Validate 1.0R
;author Stefan Strack
;strategy System validation program - based on Mark Durham's validation suite
;
;   This program tests your corewar system for compliance with the ICWS88-
;   standard and compatibility with KotH. It self-ties (i.e. loops forever)
;   if the running system is ICWS88-compliant and uses in-register evaluation;
;   suicides (terminates) if the interpreter is not ICWS compliant and/or uses
;   in-memory evaluation. A counter at label 'flag' can be used to determine
;   where the exception occured.
;
;   Tests:
;   -all opcodes and addressing modes
;   -ICWS88-style ADD/SUB
;   -ICWS88-style SPL
;   -correct timing
;   -in-memory vs. in-register evaluation
;   -core initialization

start   spl l1
count   djn count,#37      ;time cycles
t1      dat #0,#1
t2      dat #0,#3
l1      spl l2
        dat <t2,<t2
l2      cmp t1,t2
        jmp fail
        spl l4
        jmz l3,<0
t3      dat #0,#1
t4      dat #0,#2
l3      jmp @0,<0
l4      jmp <t5,#0
        jmp l5
t5      dat #0,#0
t6      dat #0,#-1
l5      cmp t3,t4
        jmp fail
        cmp t5,t6
        jmp fail
        jmp <t7,<t7
        jmp l6
t7      dat #0,#0
t8      dat #0,#-2
l6      cmp t7,t8
        jmp fail
        mov t9,<t9         ;test in-memory evaluation
t9      jmn l7,1
t10     jmn l7+1,1
l7      cmp t9,t10
        jmp fail
        mov @0,<t11
t11     jmn l8,1
t12     jmn l8+1,1
l8      cmp t11,t12
        jmp fail
        spl l9
        mov <t13,t14
t13     dat <0,#1
t14     dat <0,#1
t15     dat <0,#-1
l9      mov <t16,t16
t16     jmz l10,1
        jmp fail
l10     cmp t13,t15
        jmp fail
        add t17,<t17
t17     jmp 1,1
t18     jmp 2,1
        cmp t17,t18
        jmp fail
        add @0,<t19
t19     jmp 1,1
        jmp fail
        cmp t18,t19
        jmp fail
        spl l11            ;ICWS86 SPL will fail here
        cmp t20,t21
        jmp l12
        jmp fail
l11     sub <t20,t20
t20     dat #2,#1
t21     dat #0,#0
l12     cmp t20,t21
        jmp fail
t22     sub <t23,<t23
t23     jmp l13,1
t24     sub <-2,<1
t25     jmp l13+2,-1
l13     cmp t22,t24
        jmp fail
        cmp t23,t25
        jmp fail
        cmp start-1,t26    ;Core initialization dat 0,0
        jmp l14
        jmp fail
t26     dat #0,#0
l14     slt #0,count       ;check cycle timer
        jmp success
fail    mov count,flag     ;save counter for post-mortem debugging
        dat #0             ;and terminate
flag    dat #0
success jmp 0              ;loop forever

        end start


Article 1739 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!dog.ee.lbl.gov!network.ucsd.edu!swrinde!zaphod.mps.ohio-state.edu!howland.reston.ans.net!noc.near.net!news.centerline.com!uunet!vuse.vanderbilt.edu!stst
From: stst@vuse.vanderbilt.edu (Stefan Strack)
Subject: Summary of ICWS94 draft proposal
Message-ID: <C5EAz1.92H@vuse.vanderbilt.edu>
Sender: news@vuse.vanderbilt.edu
Nntp-Posting-Host: necs
Organization: Vanderbilt University School of Engineering, Nashville, TN, USA
Date: Tue, 13 Apr 1993 00:06:36 GMT
Lines: 45


Anders Ivner asked a while back - long enough for his post to expire here -
about what changes we might see in the next corewar standard, ICWS94. I am
giving a brief summary below. Keep in mind that the draft proposal (primary
author Mark Durham) is still very preliminary, so this is all subject to
change. We (i.e. the people in the committee) solicit your opinion on the
proposed changes; feel free to discuss them here.


                Summary of ICWS94 draft proposal

- redcode syntax is (for the first time) rigorously specified

- MARS operation, ditto; resolving ambiguities in the ICWS88 standard (see
in-memory/in-register evaluation) - a validation suite defines compatibility

- assembly files: comma between operands required, parenthesis allowed,
new pseudoop ORG (identifying first instruction to execute; same as END
start)

- a special load file format is introduced, allowing the separation of
assembler and interpreter (this may or may not make it into the final
version of the draft)

- read/write distance limitations are added as tournament variables.
(different from the current X-hill, out-of-range read/writes wrap around
rather than fail)

- a range of "tournament variable sets" (coresize, maxtasks,
cycles-until-tie, etc.) for KotH, ICWS-tournaments are listed

- more flexible opcode/addressing mode combinations (no illegal instructions)


_VERY_PRELIMINARY_:

- new addressing modes: postdecrement indirect, pre/postincrement indirect

- new opcodes MUL,DIV,MOD,XCH

- ADD/SUB revert back to ICWS86 behavior (no effect on A-operand)



-Stefan (stst@vuse.vanderbilt.edu)


Article 1740 of rec.games.corewar:
Path: hellgate.utah.edu!cs.utexas.edu!zaphod.mps.ohio-state.edu!uwm.edu!ogicse!flop.ENGR.ORST.EDU!prism.CS.ORST.EDU!wangsawm
From: wangsawm@prism.CS.ORST.EDU (W. Mintardjo)
Newsgroups: rec.games.corewar
Subject: Re: Summary of ICWS94 draft proposal
Message-ID: <1qdnegINNddc@flop.ENGR.ORST.EDU>
Date: 13 Apr 93 06:43:28 GMT
Article-I.D.: flop.1qdnegINNddc
References: <C5EAz1.92H@vuse.vanderbilt.edu>
Organization: CS Dept, Oregon State University
Lines: 48
NNTP-Posting-Host: prism.cs.orst.edu

stst@vuse.vanderbilt.edu (Stefan Strack) in his article writes:

 [Intro & some other items skipped]

>- read/write distance limitations are added as tournament variables.
>(different from the current X-hill, out-of-range read/writes wrap around
>rather than fail)
>
 Rather than using it as a limitation, what's about using it as a feature?
 (i.e add an opcode that allows the wrapping)

 [others skipped]

>- new addressing modes: postdecrement indirect, pre/postincrement indirect
>
 nod

>- new opcodes MUL,DIV,MOD,XCH
>
 what happens to DIV by zero or MOD by zero?

>- ADD/SUB revert back to ICWS86 behavior (no effect on A-operand)
>
 In my opinion, the current implementation is very intriguing and tricky. 
 Shouldn't miss it.


 One suggestion is to add an opcode which functioning as follows:
 Normally, after executing a non-jump instruction, a process would execute
 the instruction immediately succeding it. This opcode, say 'zzz 5, 0', would
 cause all the processes of that warrior to have 5 steps. Therefore, after
 executing a non-jump instruction, each of them would execute instruction
 which is 5 steps away and continued until reconfigured.

 Probably, this opcode sounds confusing. But imagine if one warrior was 
 programmed to move backward while the other was to move forward,
 or to disrupt paper programs by landing this opcode singly on it, or
 to write imp as follows:
  ??? 3369
  MOV 0, 3369

 Just a suggestion

>
>
>-Stefan (stst@vuse.vanderbilt.edu)

 Mintardjo W.


Article 1742 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!caen!zaphod.mps.ohio-state.edu!hobbes.physics.uiowa.edu!news.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s
From: pk6811s@acad.drake.edu
Subject: Eclipse II (with comments)
Message-ID: <1993Apr13.091531.1@acad.drake.edu>
Lines: 171
Sender: news@dunix.drake.edu (USENET News System)
Nntp-Posting-Host: acad.drake.edu
Organization: Drake University, Des Moines, Iowa, USA
Date: Tue, 13 Apr 1993 15:15:31 GMT

Since Eclipse II has been knocked off the Hill twice recently, I guess
it's time to publish. (I did describe parts of it while it was successful :)

Eclipse II went up and down KotH like a yo-yo, not so much depending
on _what_ was put up as _who_ was putting it up.  The theory behind it
is simple: if you could find a piece of the opponent, what would you
want to do to kill or stun the whole thing.  For imps of any size, that
would be taking the mov 0,xxxx instruction as a bombing increment and
bomb until zero is found.  For imps connected to their bombers, follow that
step with a right-to-left bombing run, dropping to the next trail as each
is finished.  During this step, bomb 'through' (like Ike) to kill anything
that was boot-strapped away.  Even better, use a spl-minusone bomb in
this step, if you can't kill all components, you might stun enough of 
them to make a difference.

After the bombing runs, launch a set of small bombers tailored to deal
with 2667-imps in case you missed them.  If your opponent is a stone,
he will be out-numbered by these bombers.

As Eclipse II evolved, I added bits to deal with difficult opponents.
Paratroops uses an initial scan phase also, and carpet-bombs right-to-left
on discovery.  So I added code to detect bombs in my next-phase code, and
go to mouse if necessary.  The spl-minusone bomb has a <57 b-operand
which prevents Imprimis 6 from launching its second set of imps.  I ran
dozens of runs against Leprechaun and Herem to get the placement of
the bomb-thru just right.  There are three partial reflections at 12
which once gave some protection (but not anymore).

Eclipse II beat most of the stone-imps, all of the vampires, some of
the stones, and depending on the author - anything else.  I have warned about
leaving pointers to your active code laying around before.  If you boot-strap
your program away from a decoy, but leave pointers in the decoy, :-)
The decoy is easy to find, so it doesn't take long to find your active
code.

Here is the (commented) source.  Try starting at 'next':

;redcode quiet
;name Eclipse II
;kill Eclipse
;author P.Kline
;strategy bscan, ringkiller, avamp, clear->gate's
;strategy including ideas from Plasma, Paratroops, and old Eclipse
;strategy added mouse for emergencies (Paratroops attack)
;strategy small change in anti-vamp
step     equ 3094
hold     equ scan-250
ptr      equ scan-120
ptr2     equ ptr+1
minone   equ scan-4000

splmin1  spl -1,<57          ; <57 is to keep Imprimis from launching 7-imp
start    sub #1,minone       ; create a dat 0,7999 for comparison
         jmp next            ; begin initial scan

ref3     add @7933,7934      ; partial reflection
         mov 7984,<7933
         mov 7983,@7931
         cmp 7950,<7930
         jmp 7995
         jmn 7994,<7928
         sub @52,@48
         djn 7992,#8
         spl 6
p2ck     spl 26
         jmp 16

clrback  mov ptr,ptr2        ; using a spl-minusone bomb
         add @ptr,ptr2       ; bomb right to left
         mov splmin1,<ptr2   ; also bomb 'thru' to wherever those locations point
         mov splmin1,@ptr    ; (kind of like 'ike')
         cmp -50,<ptr        ; watch for dat 0,0
         jmp clrback
         jmn clrback,<ptr    ; don't be fooled by a single 'dat 0,0' plant
clrback2 sub @save,@ptrptr   ; back up one '@save' increment
         djn clrback,#8      ; and continue bombing right to left
                             ; (some imp-trails are connected to their bombers)

phase2   spl g1copy          ; now launch small bombers
         
sta      spl avamp           ; also start anti-vamp
         jmp g2copy
         
         dat #0
ptr2bmb  dat #-100
g1a      dat #g1bomb+1
g1copy   mov <g1a,<g1new1    ; launch first set of bombers
         mov @g1a,<g1new2    ; two sets are launched to guarantee one successful set
         djn g1copy,#4
g1new1   spl @0,g1spl+5334+4
g1new2   spl @0,g1spl+2667+4
g1spl    spl 0,<-9
         mov g1bomb,<-9
         jmp -1,<-11
g1bomb   dat <-10,<-10
         
         dat #0
         dat #-100
g2a      dat #g2bomb+1
g2copy   mov <g2a,<g2new1    ; launch second set of bombers
         mov @g2a,<g2new2    ; this set will kill the first set if successful
         djn g2copy,#4
g2new1   spl @0,g2spl+5334+4
g2new2   spl @0,g2spl+2667+4
g2spl    spl 0,<-9
         mov g2bomb,<-9
         jmp -1,<-11
g2bomb   dat <-10,<-10

avamp    jmz 0,<2            ; little different anti-vamp
         sub @1,<1           ; first search for non-zero code
         mov 500,<700        ; then reverse the sign on it
         djn -1,#75          ; and carpet-bomb from where it points
bomb1    dat #0

ref1     add #step,@2        ; partial reflection
         jmz 7999,@7880
         cmp 3999,@7879
         slt #170,@7999
         jmp 7996
         slt #3900,@7876
         mov @7875,7745
         slt #20,@7999
         mov #5,@7998
         mov 3992,@7872
         add @7996,@7992
         jmn 7998,@7870
         
next     add #step,@ptrptr   ; initial scan
scan     jmz next,@ptr
ptrptr   cmp minone,@ptr     ; ignore djn-streams
         slt #170,@ptrptr    ; ignore self
         jmp next

         slt #3900,@ptr      ; create bombing increment in 'hold' (= @save)
save     mov @ptr,hold       ; if <3900, may be an imp
         slt #20,@save       ; if <20 use #5 for increment
         mov #5,@save
attack   mov minone,@ptr     ; bomb in forward direction
         add @save,@ptrptr
scanptr  jmn attack,@ptr     ; until zero b-field is seen
         jmz clrback2,p2ck   ; if p2ck is still zero (not bombed)
         jmp ms              ; then begin reverse-bomb attack, else go to mouse
                             ; (take that paratroops!)

ref2     cmp 3999,@7879      ; partial reflection
         slt #170,@7999
         jmp 7996
         slt #3900,@7876
         mov @7875,7745
         slt #20,@7999
         mov #5,@7998
         mov 3992,@7872
         add @7996,@7992
         jmn 7998,@7870
         jmz 7942,7933

ms       mov #8,0            ; fast mouse
         add #1431,mnew
         mov <ms,<mnew
         jmn -1,ms
mnew     spl @0,0
         jmz ms,ms
         mov 0,-3

         end start

Course this means I am working on version IV :-)

-Paul
pk6811s@acad.drake.edu


Article 1751 of rec.games.corewar:
Path: hellgate.utah.edu!cs.utexas.edu!uwm.edu!caen!kuhub.cc.ukans.edu!moe.ksu.ksu.edu!hobbes.physics.uiowa.edu!news.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s
Newsgroups: rec.games.corewar
Subject: _Push Off_
Message-ID: <1993Apr14.093317.1@acad.drake.edu>
From: pk6811s@acad.drake.edu
Date: Wed, 14 Apr 1993 15:33:17 GMT
Sender: news@dunix.drake.edu (USENET News System)
Organization: Drake University, Des Moines, Iowa, USA
Nntp-Posting-Host: acad.drake.edu
Lines: 170

                                _PUSH OFF_
                        A midweek review of Corewar
                              April 14, 1993
-------------------------------------------------------------------------------
  I.  The Standings:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  40/ 20/ 40             Night Crawler       Wayne Sheppard     159      44
 2  49/ 41/ 11              Dragon Spear             c w blue     157     146
 3  38/ 20/ 41                    NC III       Wayne Sheppard     156       1
 4  48/ 41/ 12                      Zipp             c w blue     155      15
 5  48/ 44/  8                 Agony 5.1        Stefan Strack     152     198
 6  37/ 23/ 40                Imprimis 6              P.Kline     150     444
 7  47/ 45/  8               Medusa's v7   Mintardjo & Strack     150     337
 8  34/ 24/ 42               Sphinx v2.8         W. Mintardjo     144    1042
 9  43/ 44/ 12                 Iron Gate       Wayne Sheppard     143    1000
10  32/ 22/ 46           +0 Stormbringer       Dan Nabutovsky     142    1474
11  30/ 20/ 50                        It        nandor sieben     141       6
12  38/ 37/ 24             Leprechaun 1b         Anders Ivner     139     956
13  42/ 47/ 11               Pippin v2.0   Arno M. Fuhlendorf     138       3
14  36/ 37/ 28                    Nurgle             c w blue     135      27
15  29/ 25/ 46              Silver Paper         W. Mintardjo     133      75
16  40/ 48/ 12                 Emerald 4              P.Kline     133      19
17  39/ 46/ 15                  Sucker 6        Stefan Strack     131      66
18  39/ 48/ 12               Charon v9.0   Cisek,Strack,Kline     131      31
19  36/ 46/ 18               Eclipse III              P.Kline     126      16
20  27/ 30/ 43          Consumption v2.0   Arno M. Fuhlendorf     124       8

21  12/ 78/ 10                    Invest      Andre van Dalen      47       0

-------------------------------------------------------------------------------
 II.  The Basics:

       -Core War Archives are available via anonymous FTP at 
        soda.berkeley.edu in pub/corewar...

       -FAQ for this newsgroup is available via anonymous FTP at
        rtfm.mit.edu as pub/usenet/news.answers/games/corewar-faq.z

-------------------------------------------------------------------------------
III.  The Scoop:

Last week this newsletter reported the premature demise of W. Sheppard's 
Night warrior, the first of several long-standing imp-stone programs to
fall of the Hill in months.  Well, Mr. Sheppard did not take that kindly,
no-siree-bob!  He has unleashed a series of new fighters in what can only
be called "The Week of Wayne's Weveange!".  Cover your eyes children :-)

First to fall victim to Wayne's attack was S. Strack's much-emulated
Sucker 5, at the ripe old age of 1261.  We all remember that Sucker 4
fell off the Hill just before making 1000 (when imps reappeared), so
we could almost consider this an effective age of over 2000 -- almost!
Course it didn't take long for Sucker 6 to appear and stake out a claim,
(though it's not riding as high as it used to).

Wayne's second victim was N. Sieben's mysterious 'It', another imp-stone
program of long-standing (1282).  Many times It seemed to be up when
other imps were down and down when others were up.  Just recently, It
was sharing the top 5 with four scanners.  But an unfortunate combination
of opponents, and a determined Sheppard-attack was all-she-wrote for It.
No telling yet how successful the new version of It will be :-).

Not content to just knock fighters off of KotH, Wayne released his
own stone-imp program, Night Crawler.  Studying all those opponents
must have given him an idea, 'cause NC is sitting pretty on top of the
Hill.  On top of that, his Iron Gate made 1000 just in time for this
edition (hey, even electronic newsletters have lead times :-)

D. Nabutovsky's +0 Stormbringer now has a 400+ lead in the
geriatrics competition and still looks tough.

Other programs of note to fall this week included Moonstone 1 by 
D. Nabutovsky (207), a surprising effective stone, and Paratroops v2.1 (393)
by W. Mintardjo, which used a scan->bomb->multi-bomber strategy to good effect.

Congrats to W. Mintardjo, whose Sphinx v2.8 has now passed 1000 challenges.

A. Ma was kept busy handling bug-reports for his new Corewars
emulator.  As the old saying goes, 'if you never write anything, you
will never get written up' or something like that.  Hope you get time
to work on fighters this week :-)

C.W. Blue is a quiet fellow -- quietly been taking over the Hill :-)
His Dragon Spear, Zipp, and Nurgle (and assorted sub-type-xxx's) are
beating up on a lot of good fighters.  Keep it up!

-------------------------------------------------------------------------------
 IV.  The Outlook:

 6  34/ 23/ 43              Silver Paper         W. Mintardjo     145       1
 7  43/ 43/ 14               Charon v9.0   Cisek,Strack,Kline     144       1
 7  44/ 40/ 15                  Sucker 6        Stefan Strack     149       1
10  39/ 36/ 25                    Nurgle             c w blue     142       1
 6  36/ 24/ 40              Bronze Paper         W. Mintardjo     149       1
 3  49/ 40/ 11                      Zipp             c w blue     159       1
 1  43/ 20/ 36             Night Crawler       Wayne Sheppard     166       4
 4  38/ 22/ 40                      Test       Wayne Sheppard     155       1
 2  40/ 20/ 40                     NC II       Wayne Sheppard     160       1
 2  36/ 20/ 44                    NC III       Wayne Sheppard     151       1
10  28/ 20/ 52                        It        nandor sieben     136       1
10  39/ 43/ 18               sub-type-ps             c w blue     136       1

-------------------------------------------------------------------------------
  V.  The Quick Look:

21  32/ 47/ 22                        v5            j.layland     116       0
21   9/ 82/  9                       Get      Andre van Dalen      35       0
19  31/ 44/ 25                      Rain       Wayne Sheppard     119       1
21  14/ 85/  1                      seer        Hank Turowski      43       0
21  30/ 32/ 37                      test              Unknown     129       2
21  31/ 66/  3                     Dwarf               Magnus      96       0
20  26/ 68/  5                    Hunter        Hank Turowski      84       1
20  23/ 59/ 18                    Invest      Andre van Dalen      88       1
20  27/ 30/ 43                    Jester             c w blue     123       1
21  28/ 50/ 22                    Plasma       Wayne Sheppard     106       0
21  28/ 36/ 37                    Scaper             c w blue     119       0
21  23/ 30/ 47                    p2roba        nandor sieben     115       0
21  11/ 84/  6                   Boinger         Jason Grundy      38       0
21  38/ 51/ 12                   CraMPon             c w blue     125       0
21  20/ 68/ 12                   DemonIV         Cormac Walsh      71       0
21  27/ 42/ 31                  DemonIII         Cormac Walsh     111       0
21  21/ 73/  6                  Diplomat        Rodney Kinney      68       0
20  28/ 35/ 37                 RotLD 2.2        nandor sieben     121       1
21   3/ 67/ 30                 Wimp v1.0     Brant D. Thomsen      40       0
19  33/ 48/ 20                Eclipse II              P.Kline     118       1
18  40/ 51/ 10                Iron Fence       Wayne Sheppard     129       1
21   1/ 90/  9                LERSSI III           Arto Vuori      12       0
18  39/ 46/ 15                Tomb Stone             c w blue     132       1
20  21/ 78/  1                sub-type-b             c w blue      64       1
20   4/ 64/ 32               Ground-Zero        E. J. Andrews      45       1
21  25/ 72/  2               dumb bomber        Hank Turowski      78       0
21  26/ 55/ 20              Ground-Zero+        E. J. Andrews      97       0
20   1/ 41/ 58              Hand of Doom      Jarkko Lindblad      61       1
20  38/ 57/  5              Stoned Again             c w blue     120       1
21  33/ 52/ 15              sub-type-cmp             c w blue     114       0
21  24/ 64/ 11             Diplomat 1.03        Rodney Kinney      84       0
21  27/ 55/ 18             Distance v1.1     Brant D. Thomsen      99       0
21  26/ 58/ 16             Distance v2.0     Brant D. Thomsen      95       0
21  32/ 39/ 29             Salamander II         Anders Ivner     125       0
21   0/ 88/ 12             Validate 1.0R        Stefan Strack      12       0
21   9/ 91/  0             cheap-scanner         A.T. Crowley      26       0
21  30/ 43/ 27            Am I Evil 10.0      Jarkko Lindblad     117       0
21  11/ 52/ 37            BscannerBomber       Wayne Sheppard      70       0
20  37/ 47/ 16            Moonstone v1.1              Unknown     127       1
18  28/ 29/ 43          Consumption v2.0   Arno M. Fuhlendorf     128       1
21  23/ 56/ 20       Thundering buttocks       Jordan Horwich      90       0
21  14/ 83/  2      JG-HI2-Hopping Imp 2         Jason Grundy      45       0

-------------------------------------------------------------------------------
 VI.  The Hint:

Newcomers to the game frequently confuse their recode assembler syntax
with actual runtime code.  Operators like * / + -, as well as EQU's
and parens are only 'assembler' operators, they are evaluated before
the program is loaded.  An example:

        mov 100,@2*10

is changed to:

        mov 100,@20

If you are suspicious that your emulator is not doing what you think
it should, take a look at the assembled instructions.

-------------------------------------------------------------------------------
VII.  The End:

Paul Kline
pk6811s@acad.drake.edu


Article 1754 of rec.games.corewar:
Path: hellgate.utah.edu!dog.ee.lbl.gov!network.ucsd.edu!usc!cs.utexas.edu!not-for-mail
From: wsheppar@st6000.sct.edu (Wayne Sheppard)
Newsgroups: rec.games.corewar
Subject: Subtraction coreclear
Date: 15 Apr 1993 01:43:03 -0500
Organization: UTexas Mail-to-News Gateway
Lines: 42
Sender: daemon@cs.utexas.edu
Message-ID: <9304150640.AA32984@st6000.sct.edu>
NNTP-Posting-Host: cs.utexas.edu


Subtraction coreclear, a new way to coreclear


Look at this:

;Name No ties allowed
mov <21,1+2234
sub 1,-1
jmp -2,-2234

After about 12,000 cycles, the move statement will decrement the
sub statement and it will look like this:

a mov <1,1212
b sub 1,-2
c jmp -2,-2234

As this runs, line c is subtracted from every location in the core.
Line b actually does the subtraction, while line a decrements
the pointer on line b.

This almost acts like a DAT coreclear.  Instead of MOVing a DAT at every
location, line c is SUBtracted from every location.
After the mod-2 bombing gets done first, the only things left
(besides Paper and Imp) will be SPL 0.  Sucker and other programs
use the SPL 0 in a self-splitting loop to keep you honest.
If the SPL 0 isn't on your bombing pattern, you need a coreclear
to eliminate him.

This is where the subtraction coreclear comes in. SPL 0
becomes SPL 2,2234.  The process that was trapped by the
SPL statement now scurries away a dies.

I discovered this coreclear by accident, but since then
I noticed it in Andy Pierce's Twill and Matt Hasting's SNM.

Wayne Sheppard
wsheppar@st6000.sct.edu

PS: Coming soon, the code for Night Crawler



Article 1760 of rec.games.corewar:
Path: hellgate.utah.edu!cs.utexas.edu!not-for-mail
From: wsheppar@st6000.sct.edu (Wayne Sheppard)
Newsgroups: rec.games.corewar
Subject: Night Crawler - the code
Date: 16 Apr 1993 14:19:24 -0500
Organization: UTexas Mail-to-News Gateway
Lines: 171
Sender: daemon@cs.utexas.edu
Message-ID: <9304161916.AA46448@st6000.sct.edu>
NNTP-Posting-Host: cs.utexas.edu



After watching many battles between Night and Iron Gate,
I recognized the weakness of Stone/Imp.  The imps rarely
get stunned.  Only a hit at the head would get it.  A hit
anywhere else would quickly be overwritten.  The scanners
usually stunned the stone, making the imps a sitting duck.

Also gone are the days when the imps would get the kill.
With every warrior having a gate or some other imp defence,
the stone was the place to get the win.

Like most of the older impspiral/dwarfs, Night had a very slow
startup.  I was using a Nimbus style startup.  16 process
impspiral took a long time to launch.  The first imp
did not execute until after 100 cycles had passed. Of course
this made the stone start slow, so I had to split a lot of processes
to it to speed it up.

To make improvements, I went to a smaller spiral with a quicker 
binary launch.  The first imp executes on the 35 cycle.  This
made the stone much quicker.  The seven process
imp spiral can run into a coreclear before it has set up a gate.
This turns many losses into ties or wins.

The stone part is a four line dwarf with a subtraction
coreclear.  Copied away from the decoy, it is a very
small target for scanners.
          
          # of DAT bombs for
Cycles   Night    Night Crawler 
 100       6          15
 200      17          37
 300      35          62
 500      81         115

As you can see, the stone puts out bombs quicker that before.

Score results for NC III against scanners
---------------------------  W   L   T
226  Charon v9.0            73  20   7
220  Agony 5.1              71  22   7
198  Iron Gate              62  26  12
192  Medusa's v7            61  30   9
173  Dragon Spear           55  37   8


And finally the code:

;redcode
;name Night Crawler III
;author Wayne Sheppard
;wsheppar@st6000.sct.edu
;strategy Forward decrement
;kill NC III
;
;Night Crawler is an improvement over the standard imp/stone.
;The features:
;  -quick startup, bombing starts immediately
;  -small ring lets stone run faster 
;  -small ring can gate crash slow coreclears
;  -small profile stone is smaller target for scanners
;  -large decoys to confuse scanners
;  -forward decrementing, might disable Paratroop attack

imp equ impcopy+1700  ;imp start
hide equ 1600         ;stone location 
stone             
spl 0,<-1001      ;Stone is only four lines long
mov <21,1+2234    ;Bombs at light speed
sub 1,-1          ;mod 2 pattern
djn -2,<-2234     ;converts to addition coreclear
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536  
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
dat <stone,<6536
start
mov stone+3,hide         ;Copy stone 
mov stone+2,<start       ;to a safe place
mov stone+1,<start
;spl start+hide-3,<4000   ;additional SPL caused imps to start too slow
mov stone,<start
spl start+hide-3,<3950  :Split to stone
ring
spl 9,<4050              ;binary ring startup is the fastest
spl 5,<4100              ;7 process 3-point spiral
spl 3,<4150
mov impcopy,imp          ;use one process to copy imp away from decoy
jmp start+hide-3,<4200   ;then jmp to stone
jmp imp,<4250
spl 2,<4300
jmp 2667+imp,<6350
jmp 5334+imp,<4400
spl 4,<4450
spl 2,<4500
jmp 1+imp,<4550
jmp 2668+imp,<4600
spl 2,<4650
jmp 5335+imp,<4700
jmp 2+imp,<4750
impcopy mov 0,2667
end start

Wayne Sheppard
wsheppar@st6000.sct.edu


Article 1766 of rec.games.corewar:
Path: hellgate.utah.edu!caen!uwm.edu!cs.utexas.edu!not-for-mail
From: wsheppar@st6000.sct.edu (Wayne Sheppard)
Newsgroups: rec.games.corewar
Subject: Iron Sword
Date: 19 Apr 1993 12:01:29 -0500
Organization: UTexas Mail-to-News Gateway
Lines: 31
Sender: daemon@cs.utexas.edu
Message-ID: <9304191658.AA08850@st6000.sct.edu>
NNTP-Posting-Host: cs.utexas.edu


I've tried to develop a scanner that wouldn't need a gate.  I
got a little success. Last night, using mercury, I ran 25,000 battles.
500 battles for each step size from 2 to 100.  The best step for
killing imps was 54 with 163 kills out of 100.

In desperation I tried a 3 point clear, which worked better,
but not good enough.

But it can't score enough, so I post it for the general public.



;redcode-quiet
;name Iron Sword
;author Wayne Sheppard
;strategy Cmp scanner/No gate/3 point coreclear
dist equ 34;94,98,42,54 <-these numbers worked good also

sub off,@x
loc cmp dist,0
slt #20,@x
djn -3,j
mov j,@loc
x mov s,<loc
sub new,loc
j jmn @-1,#-20
s spl 0
new mov @0-dist,<0-dist-1
sub #2666,@0-1              ;I tried to do it without this line
off dat <0-dist-dist,<0-dist-dist


Article 1768 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!dog.ee.lbl.gov!network.ucsd.edu!swrinde!zaphod.mps.ohio-state.edu!hobbes.physics.uiowa.edu!news.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s
From: pk6811s@acad.drake.edu
Subject: _Push Off_
Message-ID: <1993Apr21.100141.1@acad.drake.edu>
Lines: 190
Sender: news@dunix.drake.edu (USENET News System)
Nntp-Posting-Host: acad.drake.edu
Organization: Drake University, Des Moines, Iowa, USA
Date: Wed, 21 Apr 1993 16:01:41 GMT

                                _PUSH OFF_
                        A midweek review of Corewar
                              April 21, 1993
-------------------------------------------------------------------------------
  I.  The Standings:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  38/ 21/ 41             Night Crawler       Wayne Sheppard     154     112
 2  47/ 44/  9               Backstabber         Anders Ivner     150      32
 3  46/ 45/  8                 Agony 5.1        Stefan Strack     148     266
 4  35/ 25/ 40                Imprimis 6              P.Kline     145     512
 5  44/ 44/ 11              Dragon Spear             c w blue     145     214
 6  45/ 46/  9               Medusa's v7   Mintardjo & Strack     144     405
 7  33/ 23/ 44           +0 Stormbringer       Dan Nabutovsky     143    1542
 8  33/ 24/ 43               Sphinx v2.8         W. Mintardjo     143    1110
 9  32/ 22/ 46              Silver Paper         W. Mintardjo     143     143
10  43/ 46/ 11                      Zipp             c w blue     141      21
11  42/ 44/ 14                  Sucker 6        Stefan Strack     140     134
12  42/ 45/ 13                 Iron Gate       Wayne Sheppard     139    1068
13  39/ 40/ 21             Leprechaun 1b         Anders Ivner     138    1024
14  30/ 22/ 48                       Itt        nandor sieben     137      40
15  37/ 37/ 27                    Nurgle             c w blue     137      95
16  31/ 27/ 42                 Full moon       Dan Nabutovsky     135       7
17  41/ 48/ 12                    cproba        nandor sieben     133      39
18  40/ 47/ 13                 Emerald 4              P.Kline     132      87
19  32/ 53/ 16                    Zoiks!        C. Parrinello     110       8
20  24/ 50/ 26                  Passport              P.Kline      98       1

21   1/  2/  1              sub-type-aix             c w blue       5       2

-------------------------------------------------------------------------------
 II.  The Basics:

       -Core War Archives are available via anonymous FTP at 
        soda.berkeley.edu in pub/corewar...

       -FAQ for this newsgroup is available via anonymous FTP at
        rtfm.mit.edu as pub/usenet/news.answers/games/corewar-faq.z

-------------------------------------------------------------------------------
III.  The Scoop:

This week we have a little of the old, and a little of the new.

For the old, a big round of applause to +0 Stormbringer which passed 1500,
and also to A. Ivner whose Leprechaun 1b has joined the over-1000 ranks
(the over-the-hill gang? :-)

In the new category, S. Adkins released a new version of Core Wars Deluxe
(v2.0b) and deserves many thanks for supporting CoreWars.  Also Nandor
Sieben has created a program 'macroed' which takes limited-shorthand
input and creates standard redcode.  This may be the beginning of a whole
new generation of tools for program-writing.  Imagine a program which
could process instructions like:
    - copy the next M lines N times
    - insert the step-increment that optimally bombs M locations without
      bombing the previous N locations
    - make me a copy routine to bootstrap M locations starting at location
      X and write them at location Y
    - include xxx-program here
    - ??? (your imagination needed here)

Take note of S. Halvorsen's re-entry into the fray, ImpStop, bearing
these warnings:
;Strategy       This is my first attempt in a long time, so if you beat
;Strategy       me, I'll just have to come back with something worse !
Well, sticks and stones may break my bones, but let's see those scores! :-)

Did anyone else note the 'irony' in Wayne's comment announcing the
source-posting of Iron Sword:
  "But it can't score enough, so I post it for the general public."
Gee, thanks.

Lastly, if I had a program that scored like this, I would probably
call it 'pu' too :-)
   21  11/ 70/ 19               sub-type-pu             c w blue      51
-well, actually I have quite a few programs that score like that :-(

-------------------------------------------------------------------------------
 IV.  The Outlook:

 1  47/ 44/  9               Backstabber         Anders Ivner     151       1
10  30/ 23/ 47         Night Crawler III       Wayne Sheppard     137       1
 7  45/ 47/  8           Juggernaut v1.5         Anders Ivner     143       1
 2  35/ 18/ 47              Test Crawler       Wayne Sheppard     152       1
 2  40/ 21/ 40              sub-type-aix             c w blue     159       1
10  28/ 19/ 53                     proba        nandor sieben     138       1
10  43/ 46/ 11                   CraMPon             c w blue     140       1

-------------------------------------------------------------------------------
  V.  The Quick Look:

21   7/ 86/  7                        ik            j.layland      28       0
17  24/ 21/ 54                       Itt        nandor sieben     127       1
21  34/ 62/  4                       ugh            j.layland     106       0
21   1/  0/  4                      Zipp             c w blue       6      61
21  19/ 77/  4                      test            j.layland      61       0
20  35/ 51/ 14                     Irony       Wayne Sheppard     120       1
21  30/ 64/  6                     Snake       Wayne Sheppard      95       0
21  29/ 61/ 11                     Wedge       Rodney Schuler      97       0
21  32/ 60/  8                     test1            j.layland     103       0
21   9/ 82/  9                    Dwarfs      Andre van Dalen      35       0
21  23/ 69/  7                    Invest      Andre van Dalen      77       0
20  33/ 50/ 17                    Zoiks!        C. Parrinello     116       1
19  39/ 50/ 12                    cproba        nandor sieben     127       1
21  22/ 24/ 53                    p2roba        nandor sieben     120       0
21  24/ 70/  6                    spy2.0          Kirk Gorden      79       0
19  33/ 51/ 16                    test 2              P.Kline     116       1
21  20/ 74/  7                   Blofeld              Fredrik      65       0
18  41/ 48/ 12                   CraMPon             c w blue     134       1
21  18/ 74/  7                   DemonIV         Cormac Walsh      62       0
21  20/ 52/ 28                   ImpStop         S. Halvorsen      88       0
21   6/ 61/ 33                   cmp-add       Wayne Sheppard      50       0
20  27/ 68/  5                  Bscanner       Wayne Sheppard      86       1
21   4/ 31/ 65                  Buzzbomb           Paul Bobby      78       0
21   6/ 91/  3                  DNA ring         A.T. Crowley      22       0
21  24/ 43/ 33                  DemonIII         Cormac Walsh     105       0
21   0/ 93/  7                  Dust 1.1 Paul Svensson <paul@       7       0
20  30/ 29/ 42                  Ice Cube             c w blue     131       1
21  11/ 39/ 50                  MPCC 1.1         A.T. Crowley      82       0
20  33/ 58/  9                  Orff 3.2              Fredrik     107       1
19  28/ 39/ 33                  Passport              P.Kline     116       1
21  25/ 62/ 13                 Sleuth3.0       Kirk R. Gorden      89       0
20  13/ 57/ 30                Iron Fence       Wayne Sheppard      68       1
19  34/ 49/ 17                Iron Sword       Wayne Sheppard     118       1
20  33/ 47/ 19                Tomb Stone             c w blue     119       1
20  19/ 60/ 21                sub-type-b             c w blue      79       1
21  18/ 53/ 29                sub-type-i             c w blue      84       0
20  36/ 44/ 19                sub-type-n             c w blue     129       1
20  37/ 50/ 13               Charon v8.2   Cisek,Strack,Kline     125       1
18  37/ 44/ 19               Moonstone 2              Unknown     131       1
20  11/ 37/ 52               Pouser v2.1          Joshua Houk      86       1
21  30/ 53/ 18               sub-type-bs             c w blue     107       0
21  11/ 70/ 19               sub-type-pu             c w blue      51       0
21  25/ 66/ 10              ImpStop 1.01         S. Halvorsen      83       0
21   0/ 43/ 56              New Low v2.1          Joshua Houk      57       0
18  38/ 48/ 14              sub-type-cmp             c w blue     128       1
20  36/ 52/ 12             Distance v5.0     Brant D. Thomsen     119       1
21   3/ 41/ 56             sub-type-bs-2             c w blue      65       0
21   9/ 60/ 31            Be my Neighbor Michael Patterson (m      57       0
21   5/ 59/ 36            Metadwarf v1.0          Joshua Houk      51       0
18  29/ 30/ 41            Moonstone ring       Dan Nabutovsky     128       1
20  32/ 48/ 20            Moonstone v1.1              Unknown     117       1
21  18/ 70/ 11            Sleracil v0.5b           Arto Vuori      66       0
21  10/ 47/ 43         House Pouser v2.4          Joshua Houk      72       0

-------------------------------------------------------------------------------
 VI.  The Hint:

Several people have asked how I know what style warrior 'xxx' fighter
is.  Well, in some cases I don't have any idea.  Some others have
been posted, some have been described by their authors, and sometimes
authors have used ;strategy lines in earlier versions but removed
them in later versions.  I also make guesses based on how the fighter
performs against my own fighters.  For what it's worth, here is my best
guess for the current 20 fighters on KotH: (* indicates it has been posted):

Night Crawler *           imp-stone
Backstabber               brief bscan, core-clear, ???
Agony 5.1                 cmp-scanner, spl-0 carpet (Agony 2.4 *)
Imprimis 6 *              imp-stone, anti-vamp
Dragon Spear              cmp-scanner
Medusa's v7               cmp-scanner, spl-0 carpet (Medusa v4 *)
+0 Stormbringer           imp-stone
Sphinx v2.8 *             imp-stone
Silver Paper              paper-stone (Gamma Paper *)
Zipp                      ???
Sucker 6                  vampire (Sucker 5 *)
Iron Gate *               cmp-scanner
Leprechaun 1b *           bomber/bscanner
Itt                       imp-stone
Nurgle                    paper-stone
Full moon                 imp-stone
cproba                    scanner
Emerald 4                 stone, anti-vamp
Zoiks!                    ???
Passport                  paper-stone

I apologize for any mistakes.  Corrections and additions are appreciated.

The best policy is to ask the author - don't expect an immediate response,
some are busy people, but most are very helpful.  Some prefer to give 
hints, not code, and some of our best competitors are non-native English
speakers - so beware eh, WM?

-------------------------------------------------------------------------------
VII.  The End:

Paul Kline
pk6811s@acad.drake.edu


Article 1776 of rec.games.corewar:
Path: hellgate.utah.edu!dog.ee.lbl.gov!overload.lbl.gov!agate!howland.reston.ans.net!gatech!asuvax!asuacad!asmqk
Organization: Arizona State University
Date: Sunday, 25 Apr 1993 17:11:42 MST
From: <ASMQK@ASUACAD.BITNET>
Message-ID: <93115.171142ASMQK@ASUACAD.BITNET>
Newsgroups: rec.games.corewar
Subject: macrored.c
Lines: 105

This is the source of macrored. I'm not a real programmer ( I use Pascal :-)
so if it seems ugly let me know. ( It seems ugly enough to me like any C
source ) . This is not exactly the original, so don't use it with mars88 .
I used Turbo C 2.0 . I hope it will work on any platform.
Nandor.

#include "stdio.h"
#include "stdlib.h"
#define STRSIZ 200
#include "string.h"

FILE             *fin , *fout;
    char         dir[STRSIZ], name[STRSIZ], ext[STRSIZ], outstr[STRSIZ],
s[STRSIZ], ss[STRSIZ];
    char         *index;
    char         tomb[100][81];
    int          tpoint, i, j, k;
    char         number[4],number1[4];

main(int   argc,
     char  *argv[])
{
if ( ( argc < 3 ) || ( ! strcmp(argv[1],argv[2] ) )) {
   printf("\nRedcode macroprocessor by Nandor Sieben   4/25/1993\n");
   printf("Usage:\n");
   printf("macrored infile outfile \n");
   printf("\n");
   printf("You can use the following macro :\n");
   printf("         for 3         ;the compile variable \"for\" goes from 1 to
3\n");
   printf("LABELfor dat #for*2-1  ;\"for\" will be substituted by the actual
value\n");
   printf("         rof           ;end of for\n");
   exit(1);
}
if ( (fin = fopen(argv[1],"r") ) == NULL) {
   printf("******* Macrored error *******\n");
   printf("\n");
   printf("File does not exist\n");
   exit(2);
};
strcpy(outstr,argv[2]);
if ( (fout = fopen(argv[2],"w") ) == NULL) {
   printf("******* Macrored error *******\n");
   printf("\n");
   printf("Can't create file\n");
   exit(3);
};

while (!feof(fin)) {
      fgets(s,STRSIZ,fin);
      index = strstr(s,"for");
      if (index == NULL)
	 fprintf(fout,"%s",s);
      else {
	   index += 3;
	   sscanf(index,"%d",&k);
	   if (k <= 0) {
              printf("******* Macrored error *******\n");
              printf("\n");
              printf("Bad \"for\" parameter\n");
	      printf("%s",s);
              exit(3);
           }
           tpoint = 0;
           do {
		 fgets(s,STRSIZ,fin);
		 tpoint++;
		 strcpy(tomb[tpoint],s);
	   }  while (!((strstr(s,"rof") != NULL) || feof(fin)));
           for (j = 1; j <= k; j++) {
	       sprintf(number1,"%d",j);
	       switch (strlen(number1)) {

		    case 1:   strcpy(number,"00");
                    break;

		    case 2:   strcpy(number,"0");
		    break;
	       }
	       strcat(number,number1);
	       for (i = 1; i < tpoint ; i++) {
		   strcpy(s,tomb[i]);
                   do {
			 index = strstr(s,"for");
			 if (index != NULL) {
			    index[0] = number[0];
			    index[1] = number[1];
			    index[2] = number[2];
                         }
                   }  while (!(index == 0));
		   fprintf(fout,"%s",s);
               }
           }
      }

}
fclose(fin);
fclose(fout);
exit(0);
}






Article 1777 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!cs.utexas.edu!wupost!emory!athena!houk
From: houk@athena.cs.uga.edu (Socks)
Subject: A coupla warriors...
Message-ID: <C65tvG.DyH@athena.cs.uga.edu>
Organization: University of Georgia, Athens
Date: Tue, 27 Apr 1993 20:50:03 GMT
Lines: 154


Here's two pieces of code that I've been working on the past few days.
Unfortunately, they don't seem to be going anywhere...  Both are pretty
good against scanners - but everything else tends to wipe these guys
out...  Oh well - it was worth a try. :)

Bubble-scrape scored around 80-85 on the hill.  Mort Aux Vaches was around
75-80.  Both were around 20 wins, 60 losses, give or take a few.

-----

;redcode verbose
;name Bubble-scrape
;author Joshua Houk
;strategy  
;strategy  	Bombs entire core with SPL+DAT-bombs, then hits a spl 0
;strategy  	to tie...
;strategy  

datbomb	dat	#0		;regular ol' DAT bomb

splbomb	spl	@-1		;hopefully this will cause the other
				;program to split into a neverending
				;loop...

bomber	add	#3759,datbomb
	mov	datbomb,<datbomb	;lethal bomber

	add	#415,splbomb
	mov	splbomb,<splbomb	;stun bomber

	jmp	bomber
	end	bomber

-----

Note on Bubble-scrape:  I made a goof-up on the splbomb add instruction -
but for some reason it actually works (when the datbombs finish, the
program starts replicating like rabbits... :).

-----

;redcode verbose
;name Mort Aux Vaches
;author Joshua Houk
;strategy
;strategy	Multi-level bomber with core clear.  
;strategy

bomb	dat	#0
start	add	#10,bomb
	mov	bomb,@bomb
	add	#20,bomb
	mov	bomb,@bomb
	jmn 	start,bomb
	spl	run2
	jmp	start

	dat	#0
bomb2	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0

run2	add	#8,bomb2
	mov	bomb2,@bomb2
	jmn	run2,bomb2
	spl	clear
	jmp	run2

	dat	#0
	dat	#0
bomb3	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0
	dat	#0

clear	spl	imp

run3	add	#3039,bomb3
	mov	bomb3,@bomb3
	jmp	run3

imp	mov	0,1

	end 	start

-----

Comments, advice, suggestions, etc., are always welcomed.  Honest. :)

Joshua H
-- 
| houk@athena.cs.uga.edu | indie-list co-moderator | neurotic, but cute |
If a black cat's gonna cross my path it might as well be you.


Article 1779 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!dog.ee.lbl.gov!overload.lbl.gov!agate!howland.reston.ans.net!ux1.cso.uiuc.edu!newsrelay.iastate.edu!dunix.drake.edu!acad.drake.edu!pk6811s
From: pk6811s@acad.drake.edu
Subject: _Push Off_
Message-ID: <1993Apr29.105959.1@acad.drake.edu>
Lines: 160
Sender: news@dunix.drake.edu (USENET News System)
Nntp-Posting-Host: acad.drake.edu
Organization: Drake University, Des Moines, Iowa, USA
Date: Thu, 29 Apr 1993 16:59:59 GMT

                                _PUSH OFF_
                        A midweek review of Corewar
                              April 29, 1993
-------------------------------------------------------------------------------
  I.  The Standings:

 #  %W/ %L/ %T                      Name               Author   Score     Age
 1  38/ 23/ 38             Night Crawler       Wayne Sheppard     154     163
 2  48/ 42/ 10               Backstabber         Anders Ivner     154      83
 3  46/ 43/ 11              Dragon Spear             c w blue     148     265
 4  46/ 45/  9                 Agony 5.1        Stefan Strack     146     317
 5  44/ 43/ 13            Iron Gate 1.01       Wayne Sheppard     145      21
 6  36/ 27/ 38                Imprimis 6              P.Kline     145     563
 7  32/ 25/ 43                     ttest        nandor sieben     140      39
 8  39/ 39/ 22             Leprechaun 1b         Anders Ivner     140    1075
 9  43/ 47/ 10               Medusa's v7   Mintardjo & Strack     139     456
10  32/ 25/ 43              Silver Paper         W. Mintardjo     139     194
11  32/ 26/ 42           +0 Stormbringer       Dan Nabutovsky     138    1593
12  41/ 45/ 14                    a-test        nandor sieben     138      41
13  42/ 47/ 11                  Pysco IV        David Johnson     137      27
14  41/ 44/ 16                  Sucker 6        Stefan Strack     137     185
15  32/ 27/ 41               Sphinx v2.8         W. Mintardjo     137    1161
16  30/ 24/ 45                      ittt        nandor sieben     136      40
17  36/ 38/ 25                    Nurgle             c w blue     135     146
18  41/ 47/ 12                Eclipse II              P.Kline     134      18
19  39/ 47/ 14                 Emerald 4              P.Kline     130     138
20  34/ 48/ 18             Distance v6.0     Brant D. Thomsen     120       1

21  33/ 52/ 15                 Wimp v5.0     Brant D. Thomsen     114       0

-------------------------------------------------------------------------------
 II.  The Basics:

       -Core War Archives are available via anonymous FTP at 
        soda.berkeley.edu in pub/corewar...

       -FAQ for this newsgroup is available via anonymous FTP at
        rtfm.mit.edu as pub/usenet/news.answers/games/corewar-faq.z

-------------------------------------------------------------------------------
III.  The Scoop:

Another member of the OverTheHill Gang was pushed off the Hill this week -
W. Sheppard's Iron Gate program which had some success against the imps
was pushed off at the ripe old age of 1089.  That leaves just three members
of the OTH Gang: +0 Stormbringer, Leprechaun 1b, and Sphinx v2.8.  We have
to look way down the ranks to find the next candidate with a chance at
1000, Imprimis 6.  Well, here's one author crossing his fingers, (and
wishing he had maybe just left Imprimis 4 or 5 up, they would be there
by now :-/

Iron Gate's fall was due in part to a relentless attack by
C.W. Blue, who's sub-type-xxx fighters have been appearing all up and down
the list from day-to-day.  He keeps knocking them off, so it's unclear
what he's up to.  Is Blue preparing a swarm of tough fighters to release
all at once and take over?  Stay tuned...

The demise of Iron Gate gave rise to speculation that spl-jmp bombing
cmp-scanners had lost their competitiveness.  Carpet-bombing scanners
are still successful, but with Charon and Iron Gate being pushed off,
no other spl-jmp bombers were left (as far as I know :).  But Wayne
gave IG some new points, plugs, and paint - and now Iron Gate 1.01 is
ranked fifth.

In last week's _Push Off_ I speculated that Backstabber was a bscan,
core-clearing type of program.  Anders was kind enough to correct me
by revealing it is really a cmp-scanner (with a little different bomb).

-------------------------------------------------------------------------------
 IV.  The Outlook:

 8  43/ 47/ 11                  Pysco IV        David Johnson     139       1
 9  44/ 45/ 11              sub-type-cmp             c w blue     142       1
 7  44/ 45/ 11                sub-type-c             c w blue     144       1
 8  32/ 24/ 44             Snake-crawler       Wayne Sheppard     141       1
 5  46/ 46/  8                   Cleaver       Wayne Sheppard     145       1
 1  34/ 20/ 46                sub-type-i             c w blue     148       1

-------------------------------------------------------------------------------
  V.  The Quick Look:

21   4/ 56/ 40                       Foo     Brant D. Thomsen      52       0
21  15/ 54/ 31                      Ally              Mestern      75       0
20  35/ 52/ 13                      Test       Wayne Sheppard     117       1
21  29/ 63/  7                      ice2                 Ziad      95       0
21   3/ 67/ 29                      test              P.Kline      40       0
21  32/ 57/ 11                      test            j.layland     108       0
17  36/ 40/ 24                      test          Andy Pierce     133       1
21   5/ 58/ 37                     Tiger               Mester      51       0
21  18/ 51/ 31                    Invest      Andre van Dalen      84       0
21  10/ 80/ 10                    Lots20     Jeffrey Chrisope      40       0
20  24/ 36/ 40                    Scaper             c w blue     112       1
18  40/ 46/ 15                    a-test        nandor sieben     134       1
21  12/ 69/ 19                    pebble              Mestern      55       0
18  41/ 48/ 11                   CraMPon             c w blue     134       1
21  16/ 82/  2                   Rounder     Jussi Vehkalahti      50       0
21  28/ 59/ 13                   ccproba        nandor sieben      96       0
21  14/ 48/ 38                  MPCC 2.1         A.T. Crowley      81       0
21  15/ 30/ 55                  Passport              P.Kline     100       0
19  32/ 46/ 21                 Iron Gate       Wayne Sheppard     118       1
21  38/ 49/ 13                 Pysco III        David Johnson     126       0
20   7/ 73/ 20                 Replicant       Paul Whittaker      42       1
21   7/ 87/  6                 Vamp kill              Fredrik      28       0
19  32/ 44/ 24                 Wimp v3.0     Brant D. Thomsen     120       1
21  29/ 37/ 34                 Wimp v3.1     Brant D. Thomsen     121       0
21  34/ 54/ 12                 Wimp v4.0     Brant D. Thomsen     113       0
21  39/ 48/ 14                Eclipse II              P.Kline     130       0
21  13/ 43/ 43                ScareForce              P.Kline      84       0
21   0/ 44/ 55                Sludge 1.0   I'd_rather_not_say      57       0
21  21/ 36/ 42                multi mice        nandor sieben     106       0
20  38/ 53/ 10                sub-type-c             c w blue     123       1
21  26/ 65/  9               Double Orff              Fredrik      88       0
21  16/ 82/  2               Spreader v1     Jussi Vehkalahti      50       0
21  10/ 85/  5               White Kross          Joshua Houk      34       0
21   0/ 94/  5               Witchdoctor       Paul Whittaker       6       0
20  38/ 50/ 12              ExtraExtra 2              P.Kline     127       1
21  10/ 68/ 21              Parasite v 5      Roderick Easton      53       0
20  23/ 67/ 10              Vamkill Orff              Fredrik      78       1
21  24/ 60/ 16             Bubble-scrape          Joshua Houk      88       0
20  34/ 48/ 18             Distance v6.0     Brant D. Thomsen     119       1
21  26/ 69/  5            Improved Dwarf            Ben Burch      82       0
20  11/ 51/ 38           Bubble-scrape 2          Joshua Houk      71       1
21  22/ 68/ 10           Mort Aux Vaches          Joshua Houk      76       0
21  13/ 40/ 47          White Kross v1.2          Joshua Houk      87       0
21   1/ 64/ 35       Replicant (Nexus 2)       Paul Whittaker      39       0
21   2/ 55/ 43 Replicant (Nexus 2 Revisi       Paul Whittaker      49       0

-------------------------------------------------------------------------------
 VI.  The Hint:

How many ways are there to copy code from one place to another?  Lots!
Here are three ways to copy four lines of code (hide1-hide4):

hide1   xxx
hide2   yyy
hide3   zzz
hide4   aaa
old     dat #0
---------------
copy    mov <old,<new   ; short code 
        djn copy,#4     ; - you want a low profile, 
new     spl @0,old+1000 ;   or need to stay under 100-line limit
                        ; total time to copy and spl -> 9
---------------
        spl 1           ; multi-process 
        spl 1           ; - you are using multiprocess for other reasons
copy    mov <old,<new   ; total time to copy and spl -> 8
new     spl @0,old+1000
---------------       
copy    mov <old,<new   ; fastest
        mov <old,<new   ; - you don't care about length and
        mov <old,<new   ;   want the fastest start
        mov <old,<new   ; total time to copy and spl -> 5
new     spl @0,old+1000

-------------------------------------------------------------------------------
VII.  The End:

Paul Kline
pk6811s@acad.drake.edu


Article 1780 of rec.games.corewar:
Path: hellgate.utah.edu!caen!sdd.hp.com!elroy.jpl.nasa.gov!kilroy!jlayland
From: jlayland@kilroy.Jpl.Nasa.Gov (James Layland)
Newsgroups: rec.games.corewar
Subject: Macroized Paper
Date: 29 Apr 1993 17:10:27 GMT
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 65
Message-ID: <1rp263INNpp6@elroy.jpl.nasa.gov>
References: <1rl4n3$g6t@calvin.NYU.EDU>
NNTP-Posting-Host: kilroy.jpl.nasa.gov

Just thought I'd show that Nandor's redcode macro processor is 
good for more than abbreviating decoys.  Hmmm... now if we just
had a few more enhancements we could all quit this programming 
stuff and write programs like:

stone
binary launch 7-point imp
forward core-clear
end

Well... maybe not.

Anyway, here is a macrored-ized version of Flash Paper.  More or less.
I still had to write out two copies of the replicator.  Maybe 
multi-line macro subroutines??  This is solely for demonstration purposes;
I make no claims that the constants chosen here are anything resembling
"optimal".

;redcode 
;name Macro Paper
;author J.Layland
;strategy macroized version of Flash Paper 
;strategy some constants randomly changed for no good reason.
 
start	spl	p2	;hedge bets against lucky stone
	spl	1
	spl	1	;start 4 processes going for copy
	spl	p002
	spl	p003
	spl	p004

for 7

lfor	mov	#8,	8		;set source
pfor	mov	<lfor,	<tfor		;make copy
	mov	<lfor,	<tfor		 
tfor	spl	@0,	5001+200*for	;split to copy
	mov	<-1,	<-2000+200*for
 	jmz	lfor,	lfor		;loop
	dat	<2667, #1
	dat	<2667, #1 
 
rof

p2	spl	1
	spl	1
 
	spl	p005
	spl	p006
	spl	p007

l8	mov	#8,	8		;set source
p8	mov	<l8,	<t8		;make copy
	mov	<l8,	<t8		 
t8	spl	@0,	5001+200*8	;split to copy
 	mov	<-1,   <-2000+200*8	
 	jmz	l8,	l8		;loop
 	dat	<2667, #1 
 	dat	<2667, #1 
 
	end	start


James Layland
jlayland@grissom.jpl.nasa.gov


Article 1785 of rec.games.corewar:
Path: hellgate.utah.edu!utah-morgan!cs.utexas.edu!not-for-mail
From: wsheppar@st6000.sct.edu (Wayne Sheppard)
Newsgroups: rec.games.corewar
Subject: Reflections (was Re: _PushOff_)
Date: 30 Apr 1993 10:19:32 -0500
Organization: UTexas Mail-to-News Gateway
Lines: 35
Sender: daemon@cs.utexas.edu
Message-ID: <9304301516.AA57385@st6000.sct.edu>
NNTP-Posting-Host: cs.utexas.edu


<>cmp-scanners had lost their competitiveness.  Carpet-bombing scanners
<>are still successful, but with Charon and Iron Gate being pushed off,
<>no other spl-jmp bombers were left (as far as I know :).  But Wayne
<>gave IG some new points, plugs, and paint - and now Iron Gate 1.01 is
<>ranked fifth.
<>
<Since Iron Gate 1.01's strategy line just says he changed the scan 
<constant, perhaps its demise was because it was (I believe) the
<only scanner on the Hill with a published scan constant?  Any programs
<out there using reflections at 98?  Wayne could probably tell us...
<
<James
<jlayland@grissom.jpl.nasa.gov

James has guessed correctly.  I tried out various changes, but none
of them improved my score.  So I am still using the original code
with only a modification to the dist equ ***.

There was only one person using reflections at 98, but what hurt was
that he had three programs with reflections.  And you can be sure that
my new constant will not become public.

But I think that Paul might be stretching it a bit when he says that
spl/jmp scanners are dead.  Carpet bombers have a better chance at
catching an imp trail, but they will spend more time laying out
the carpet.  Plus, from what I know, the scan distance must be close.
Carpet bombers also have an advantage because they might pick up
some non essential data around your program, and then carpet the
entire thing.  spl/jmp scanners have an advantage of being faster,
they can scan more locations than carpet bombers.  But, because
you need two sets of increments, spl/jmps tend to be a line longer.

Wayne Sheppard



Article 1786 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!dog.ee.lbl.gov!overload.lbl.gov!agate!howland.reston.ans.net!zaphod.mps.ohio-state.edu!cs.utexas.edu!uunet!vuse.vanderbilt.edu!stst
From: stst@vuse.vanderbilt.edu (Stefan Strack)
Subject: EBS summer tournament
Message-ID: <C6B6uI.BAG@vuse.vanderbilt.edu>
Sender: news@vuse.vanderbilt.edu
Nntp-Posting-Host: necs
Organization: Vanderbilt University School of Engineering, Nashville, TN, USA
Date: Fri, 30 Apr 1993 18:18:18 GMT
Lines: 34



                _Announcing: The EBS summer tournament 1993_

This is the third attempt at getting an "interactive" corewar tournament
going; may this one will actually finish :-)

This is a double-elimination tournament with rounds every week.  Emphasis is
on pitting programmer vs programmer rather than program vs program.  You will
receive source for *all* warriors after each round (this is to save myself
some work). You can submit a new warrior before the next round (which of
course won't be publicized until after the next round), or continue with your
previous entrant.

Battles are KotH style:
        coresize: 8000
        maxprocs: 8000
        cycles until tie: 80,000
        maxlength: 100
        battles: 100 (win: 3 pts, tie: 1 pt)

The entry deadline will be the Friday before each round, 20:00 CST. Results
will be posted to r.g.cw and emailed on that same weekend. Source will be
emailed at the same time.

Send your entry to stst@vuse.vanderbilt.edu. One entry per person. Deadline
for 1st round submission is Friday, May 7th, 20:00 CST.

The tournament will take place only if there are at least 12 contestants, so
please send in your submissions. If there are more than 24 or so, I plan to
split the tournament into an A- and B-league. The B-league will be reserved
to programmers that do not currently have a warrior on the hill.

-Stefan (stst@vuse.vanderbilt.edu)


Article 1787 of rec.games.corewar:
Newsgroups: rec.games.corewar
Path: hellgate.utah.edu!dog.ee.lbl.gov!overload.lbl.gov!agate!howland.reston.ans.net!gatech!europa.eng.gtefsd.com!news.ans.net!cmcl2!acfcluster.nyu.edu!chrisopj
From: chrisopj@acfcluster.nyu.edu
Subject: Team Play
Message-ID: <1993Apr30.201828.1@acfcluster.nyu.edu>
Sender: notes@cmcl2.nyu.edu (Notes Person)
Nntp-Posting-Host: acf1.nyu.edu
Organization: New York University
Date: Sat, 1 May 1993 01:18:28 GMT
Lines: 20

Hoi!

I'm still a newbie, but learning here and there by combing the *huge*
files of archived warriors at soda.b.e.

Quick question: are there any facilities out there for having more than
two warriors in core at once...or more specifically, more than two
warriors, but stilll with *two* "sides", so that teammates would have to
work together somehow (IPC!)...

*shrug* Just asking. I could of course spend quite a long time just
figuring out what is here now, but there seem to be many posts in the
archives of people suggesting that stalemates of some kind have been
reached, or that the scope and range of innovation has been reduced to
finding better optimum numbers...

Just a suggestion, or not even that...;)
Jeff, whose 1 line test program did better on the hill than his 9 line
real one :(



