🚀 go-pugleaf

In article <23...@eerie.acsu.Buffalo.EDU> v132...@ubvmsd.cc.buffalo.edu writes:
>to me.  The nice thing is, core wars is available right now on the Mac if you
>want to play it.  So far, it's a little too complex for me to take a stab at
>it.  Has anyone else even heard of that game?

Sci Am had an article on it about 2 or 3 years ago, including directions
for entering the next Core Wars tournament.  Never followed it up,
though.
--
Hal Mueller            hmue...@cssun.tamu.edu          n270ca@tamunix (Bitnet)
Graduate Student, Department of Computer Science
Research Assistant, Department of Wildlife and Fisheries Sciences
Texas A&M University, College Station, TX 77843

Hey, if I wrote a Redcode interpreter (i.e. a Core Wars program) would
people be interested in a few games across the net?   (I see
comp.core-wars looming...)
-- 
\/ o\ Paul Crowley aipdc@uk.ac.ed.castle                       :-| Have a day
/\__/ "The whole is the sum of its parts, plus one or more bugs" aimsm@castle

In article <23...@eerie.acsu.Buffalo.EDU> v132...@ubvmsd.cc.buffalo.edu writes:
>
>	The post referring to the game where various pieces of code seeks to 
>hunt down and destroy various other pieces of code sounds a lot like 'core wars'
>to me.  The nice thing is, core wars is available right now on the Mac if you
>want to play it.  So far, it's a little too complex for me to take a stab at
>it.  Has anyone else even heard of that game?
>
>John Feinberg
>SUNY at Buffalo

I tried to respond to that, but something was down somewhere, and it came back
in my face.  Anyway, here's what I was going to say:

>I don't know if anyone else has mentioned this...
>
>During the the mid 1970s (around 1974/75), part of a class at Iowa State
>in the CS department had a program-organism  game.  There was an 'arena'
>of linear  memory and a  'machine  language'  that you  programmed  your
>...
>
>Tim Mikkelsen

	Core Wars.  In a recent Computer Virus article in Scientific American,
in the Computer Recreations section, some dope actually referred to Core Wars
as a training ground for writing viri.  It goes to show how little he knows
about Core Wars and viri in general, because Core Wars was really nothing like
virus writing, and the language wasn't even similar to anything on a chip.
It's always these people who get published, though.

	Darrin

mdchaney@iubacs
mdch...@bronze.ucs.indiana.edu
mdch...@rose.ucs.indiana.edu

YES!  I love corewars!  although, all i have been able to come across
is the IBM .EXE file.  does anyone have the C code for Corewars anywhere?
I would really really appreciate it if someone mailed a copy to me.,..
thanx in advance..  i'd also be willing to post code for others to battle..

ai...@castle.ed.ac.uk (Paul D. Crowley) writes:
>Hey, if I wrote a Redcode interpreter (i.e. a Core Wars program) would
>people be interested in a few games across the net?   (I see
>comp.core-wars looming...)

Sure, why not.  I always wondered why there wasn't already a newsgroup like
comp.core-wars on the net.  I would support the idea of such a newsgroup all
the way.
--
Sincerely,                         | be...@cip-s01.informatik.rwth-aachen.de
           Stephen R. van den Berg | ...!uunet!mcsun!unido!rwthinf!cip-s01!berg

In article <23...@eerie.acsu.Buffalo.EDU> v132...@ubvmsd.cc.buffalo.edu writes:
>to me.  The nice thing is, core wars is available right now on the Mac if you
>want to play it.  So far, it's a little too complex for me to take a stab at
>it.  Has anyone else even heard of that game?

Yeah - someone here at Oberlin has written a very spiffy version for
Sunview.  It displays the memory space and the little programs running in
different colors.  You can stick in programs here and there with the mouse.
Also comes with a fair number of classic sample programs like vampire, dwarf,
imp, worm, ...
The version of the language used in this application is called Mars; I believe
it's the same one talked about in Scientific American.
Nobody really plays with it around here, though.  I think the reason is that
the really interesting programs are too big and always lose.

Lars Huttar
hut...@occs.cs.oberlin.edu

In article <23...@eerie.acsu.Buffalo.EDU> v132...@ubvmsd.cc.buffalo.edu writes:
>
>	The post referring to the game where various pieces of code seeks to 
>hunt down and destroy various other pieces of code sounds a lot like 'core wars'
>to me.  The nice thing is, core wars is available right now on the Mac if you
>want to play it.  So far, it's a little too complex for me to take a stab at
>it.  Has anyone else even heard of that game?
>
>John Feinberg
>SUNY at Buffalo

Yes, I saw and ran the game a couple of years ago. At the time, it was a little
over my head, and the "HELP" with the game assumes you are a super-proficient
programmer. (Which I am not) If anyone has out there has experience with this
game, please post.  I might just dig out the game again.

`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'`'
David Ritchie     | "I teleported home one night with Ron and Sid and Meg. 
da...@wpi.wpi.edu | Ron stole Meggie's heart away and I got Sidney's leg"
WPI Box 1457, WPI, Worcester, MA 01609 |	-Douglas Adams
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

In article <27...@husc6.harvard.edu> hut...@occs.cs.oberlin.edu (Lars Huttar, NOT David Carlton) writes:
>In article <23...@eerie.acsu.Buffalo.EDU> v132...@ubvmsd.cc.buffalo.edu writes:
>>to me.  The nice thing is, core wars is available right now on the Mac if you
>>want to play it.  So far, it's a little too complex for me to take a stab at
>>it.  Has anyone else even heard of that game?
>
>Yeah - someone here at Oberlin has written a very spiffy version for
>Sunview.  It displays the memory space and the little programs running in
>different colors.  You can stick in programs here and there with the mouse.
>Also comes with a fair number of classic sample programs like vampire, dwarf,
>imp, worm, ...
>The version of the language used in this application is called Mars; I believe
>it's the same one talked about in Scientific American.
>Nobody really plays with it around here, though.  I think the reason is that
>the really interesting programs are too big and always lose.

Back in the heyday of 8-bit personal computers, a game company called MUSE
had a package called RobotWars that was along these lines. You had a chassis
with a radar, gun, and engine, and wrote the code to drive the robot. Then 
you put the robot thru its paces either on a testbench or on a battlefield
against other robots. The game came with a couple prewritten control programs
to spar against, and the company sponsored a tournament with a decent cash
prize to the best robot design/er. I used to play it a lot on an Apple II,
tho I think there were C64 and other versions available. MUSE went under early
on, tho. ("Play" isn't quite the right word; using the package consisted of
writing the program in their text editor, assembling it, and then loading up
a battlefield with your program and various others, then sitting back to watch
and see who survived...)

Nowadays there's a game called Omega that's pretty similar to RobotWars, it's
supposed to be available for PC, Atari ST, Amiga, etc... You can duel across
modems, and the company also is sponsoring a tournament. 
--
  -- Howard Chu @ University of Michigan
  ... the glass is always greener on the side ...

>Nowadays there's a game called Omega that's pretty similar to RobotWars, it's
>supposed to be available for PC, Atari ST, Amiga, etc... You can duel across
>modems, and the company also is sponsoring a tournament. 

Let's not forget C-Robots! The Omega for people who can't snarf a copy of Omega.
:-)
although, i never much liked C-Robots. it was too limited to be C. it's always 
annoyed me that my battleprograms don't have access to enviornment and such
things. my corewars program should at least be able to find the coresize! :->
but of course, that would be cheating...but its simple things like that that
can get annoying. i'd rather play "Super CoreWars", "Wheel Wars" whatever you
wanna call it. actually, i think it should be "Super CoreWars" since its more
fun taking full processor programs (C, Assembly, Fortran, Whatever) instead 
of Redcode.. you get the point.. whereas Wheel wars is just an effort to 
knock off another user... (a more appropriate name - "F*** the newbie")

ah well..

In article <36...@castle.ed.ac.uk> ai...@castle.ed.ac.uk (Paul D. Crowley) writes:
>Hey, if I wrote a Redcode interpreter (i.e. a Core Wars program) would
>people be interested in a few games across the net?   (I see
>comp.core-wars looming...)

I have an one for the PC.  I got it from grape.ecs.clarkson.edu.  If you 
cannot get it, I would be willing to mail it, but please FTP firtst.
The program is fairly fast, and includes numerous examples.  It does not,
however contain the technical details of the game, but I got that info
from sci am  (can't remember the date.)

------------------------------------------------------------------------
| Jochen Fritz            | For though we live in the world, we do not |
| joef...@pawl.rpi.edu   | wage war as the world does.-- 2 Cor. 10:3  |      
| usergk2s@rpitsmts.bitnet| You have heard it said, Love your neighbor |
| Noah [the peace monger] | and hate your enemy.  But I tell you: Love |
|                         | your enemies.  Matt. 5:43-44               |
------------------------------------------------------------------------

In article <27...@husc6.harvard.edu> hut...@occs.cs.oberlin.edu (Lars Huttar, NOT David Carlton) writes:
>The version of the language used in this application is called Mars; I believe
>it's the same one talked about in Scientific American.
>Nobody really plays with it around here, though.  I think the reason is that
>the really interesting programs are too big and always lose.
>
>Lars Huttar
>hut...@occs.cs.oberlin.edu

Actually, MARS was the simulated memory environment that a corewars battle takes
place in.  REDCODE is the language originally detailed in the the Computing 
Recreations column in S.A.

The point you made is true, however.  Corewars is a nice way to learn the ins
and outs of simple assembly programming, but has a fault in that it stifles 
development.  The most virulent corewars program is exactly 1 opcode long, and
it can usually mess up anything longer.

The reason for this is simple:  MARS schedules processes by simple alternation.
Because of this, the smaller a program is, the faster it will run.  Also, the
smaller a program is, the harder it is to hit.  Therefore, the degenerative case
is the IMP program, presented in its entirety:  MOV 0,1.   This goes smashing
through memory, ovewriting everything will MOV 0,1 instructions.

Now it can be argued that you can stop an imp.  However, any program designed
to stop an imp will be longer than 1 opcode.  The smallest one I know is 5 
opcodes.  Since only 1 opcode out of five is the effective "imp smasher", this
program has only 20% chance of success.  The other 80% of the time, the zippy
IMP will have begun corrupting the first parts of the IMPTRAP and have reduced
the two programs to two imps, chasing each other through MARS forever.  Since
the original program failed in its mission (to kill the IMP) while the IMP succ-
eeded it its mission (to corrupt the other program), you could make the case 
that the IMP won.

This can be remedied, however.  It should be stated that to win, all other 
processes must be dead.  If ties (i.e. a stable state where two or more processes
are still alive) are replayed continuously until a victor is found, the IMP
will always lose.

Now the problem is a little more like reality:  How to write a program that is
hardy enough but as small as possible?  DWARF might be the obvious answer, but
I am not certain.  If the size of MARS is increased (the original was only 8001
addresses, with circular addressing [i.e. address 8001 = address 0000]), larger
and more complex programs might stand a chance.  In this day and age, an 8K 
memory space is ridiculously small.  How about implementing 8M or 24M (w/swap
space)?  This would allow HUGE programs.... 


NOTE: The Mac version of CoreWars that I have seen implements a modified REDCODE
opset.  It allows for the optional SPL (i.e. fork) opcode as well as a strange
PRO (I think that was it) opcode that protects an address from ONE overwrite 
attempt.  This made for a simpler imp trap but in the end just slowed a program
down without giving it 100% protection from multiple imp attacks (aka the 
IMPGUN).   I never did  decide if I liked the extended instruction set or not.

So:  Who is going to run the First Net CoreWar?  

D.
dje...@sun.com