Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

1st International Corewar Programming Contest

8 views
Skip to first unread message

fizmo

unread,
Dec 9, 2010, 4:31:13 PM12/9/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

This is a single round Corewar programming contest
sponsored by COREWAR.INFO.

Corewar is a programming game in which two or more
battle programs (called "warriors") compete for the
control of the "Memory Array Redcode Simulator" virtual
computer ("MARS"). These warriors are written in an
abstract assembly language called Redcode. The object
of the game is to cause all processes of the opposing
warrior to terminate, leaving the victorious program
in sole possession of the machine.

--------------------------------------------------------
| Contest Rules |
--------------------------------------------------------

CORESIZE: 8000
MAXPROC: 200
MAXCYCLES: 80000
MAXSIZE: 20
MINDIST: 20
ROUNDS: 1000
INSTRUCTION SET: ICWS '94

- The opcodes stp and ldp are disallowed

- No self-fights

- All warriors submitted will be published at the end
of the contest

- Up to two warriors per author are allowed, but only
the best entry counts.

- Scoring is simple: 3 Pts for a win, 1 point for a tie

- Besides the ICWS '94 instruction set the entries must
run properly with CoreWin and pmars.

--------------------------------------------------------
| Contest Submission Guidelines |
--------------------------------------------------------

Submit your entries to: fizmo (at) corewar (.) info

Format your submission as follows:

;redcode
;name <- name of your entry
;author <- your name
;strategy <- if you like you can add some comments here
;assert 1
;kill <- assign which of your previous entries you
would like to replace

;evolver <- if you use an evolver program please state
here which one you was using

then add your program

Here's a sample entry:

;redcode
;name Dwarf
;author A. K. Dewdney
;assert 1
;strategy Throw DAT bombs around memory,
;strategy hitting every 4th memory cell.

bomb DAT #0
dwarf ADD #4, bomb
MOV bomb, @bomb
JMP dwarf
END dwarf

You can submit anytime until the deadline. You can
replace any time after you've submitted two possible
entries until the deadline.

I will every day (except December 24-26) publish the
updated ranking (without details like %W/%L/%T and
scoring) on the contest website at:

http://www.corewar.info/tournament/icpc1/

--------------------------------------------------------
| Top links for Corewar Newbies |
--------------------------------------------------------

The ICWS'94 draft (extended):
http://www.corewar.info/lexicon/94draft.htm

Clueless newbies should begin with Scott's Quick Core
War Guide for a brief overview:
http://corewar.co.uk/manley/guide.htm

A very nicely written German tutorial can be found here:
http://www.netcologne.de/~nc-zapfsa/files/Handbuch.pdf

If you're just starting to program in redcode you should
download MARS, the Redcode virtual machine. Recommended
are CoreWin for Windows:
http://corewar.co.uk/wendell/
or pMARS for virtually everything else.
http://corewar.co.uk/pmars/

Here is a Beginner's Guide to Redcode to learn the
essentials of '94 Redcode programming:
http://vyznev.net/corewar/guide.html

The Corewar Bibliography is an ideal starting point
for further research.
http://corewar.co.uk/biblio.htm

Also the Corewar Lexicon provides useful information:
http://www.corewar.info/lexicon/

If you like to dissect, study or play published warriors
then check out Koenigstuhl:
http://users.obs.carnegiescience.edu/birk/COREWAR/koenigstuhl.html

Further interesting website are:
http://corewar.co.uk/article.htm
http://corewar.co.uk/guides.htm

If you decided not to handcode your entry but use a
evolver program then have a look here:
http://www.infionline.net/~wtnewton/corewar/evol/
http://www.corewar.info/evolve.htm
http://www.corewar.co.uk/evolving.htm

--------------------------------------------------------
| Deadline |
--------------------------------------------------------

December 31, 2010


********************************************************
** COREWAR.INFO will be awarding $50 for the winner **
********************************************************

Happy coding

John Metcalf

unread,
Dec 10, 2010, 6:35:40 AM12/10/10
to
On Dec 9, 9:31 pm, fizmo <fizmo_mas...@yahoo.com> wrote:
> **   1st International Corewar Programming Contest    **

> **  COREWAR.INFO will be awarding $50 for the winner  **

Count me in! :-)

fizmo

unread,
Dec 17, 2010, 11:37:56 AM12/17/10
to
********************************************************
**      _     _       _____ _____ _____   _____       **
**     /_ |   | |     |_   _/ ____|  __ \ / ____|     **
**      | |___| |_      | || |    | |__) | |          **
**      | / __| __|     | || |    |  ___/| |          **
**      | \__ \ |_     _| || |____| |    | |____      **
**      |_|___/\__|   |_____\_____|_|     \_____|     **
**                                                    **
********************************************************

********************************************************
**   1st International Corewar Programming Contest    **
********************************************************

Here's the first update.

So far I received six entries from four authors. Two
warriors are handcoded, while the rest is evolved.

Ranking
# Name Author Type
-------------------------------------------------------
1 ClamTrucks WTN Maezumo
2 Decomposition WTN Maezumo
3 Cupid Attoparsec DataWraith YACE
4 ClpTiny P. Kline handcoded
5 Bellicose Paperweight DataWraith YACE
6 Dig'n'Die Matthias handcoded

The strategies of the entries are very diverse. There
is no clear leader yet, which keeps the contest very
thrilling.

Don't forget:

You can replace anytime until the deadline your
submitted entries. Just state with ;kill which of
your entry you want to replace

The deadline is December 31, 2010

fizmo

unread,
Dec 20, 2010, 1:02:55 PM12/20/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the second update.

So far I received 11 entries from 7 authors. Seven
warriors are handcoded, while four are evolved.
Handcoded warriors actually get a slight lead over
evolved warriors. Let's see if they can hold the
position in the next days.

Ranking
# Name Author Type
-------------------------------------------------------

1 Resurrection Anton Marsden handcoded
2 fairy lights John Metcalf handcoded
3 Fury Roy van Rijn handcoded
4 Unleashed Roy van Rijn handcoded
5 ClamTrucks WTN Maezumo
6 Decomposition WTN Maezumo
7 Shorty Anton Marsden handcoded
8 ClpTiny P. Kline handcoded
9 Iridium Cockroach DataWraith YACE
10 Cupid Attoparsec DataWraith YACE
11 Dig'n'Die Matthias handcoded


Anton Marsden is leading the ranking with John Metcalf
and Roy van Rijn on 2nd and 3rd place. It's still a
very, very close battle. So, don't hesitate and send
your entry to

fizmo (at) corewar (.) info

And don't forget:

fizmo

unread,
Dec 23, 2010, 1:01:54 PM12/23/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the third update.

So far I received 13 entries from 9 authors. Nine


warriors are handcoded, while four are evolved.

Handcoded warriors still get a slight lead over
evolved warriors.

Ranking


# Name Author Type
-------------------------------------------------------
1 Resurrection Anton Marsden handcoded
2 fairy lights John Metcalf handcoded

3 Break Stuff Anton Marsden handcoded
4 Fury 2 Roy van Rijn handcoded
5 Placeholder Michal Janeczek handcoded
6 Unleashed 2 Roy van Rijn handcoded
7 ClamTrucks WTN Maezumo
8 Decomposition WTN Maezumo
9 ClpTiny P. Kline handcoded
10 Iridium Cockroach DataWraith YACE
11 Cupid Attoparsec DataWraith YACE
12 Matt Daemon Samy Duc handcoded
13 Dig'n'Die Matthias handcoded

Anton Marsden is still leading with Resurrection the
ranking. Several authors are very close behind with
their entries. So, let's see what the next days will
bring and send your entries.

fizmo

unread,
Dec 26, 2010, 1:39:11 PM12/26/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the fourth update.

So far I received 16 entries from 11 authors. Twelve


warriors are handcoded, while four are evolved.

Handcoded warriors starts to get a lead over
evolved warriors.

Ranking
# Name Author Type
-------------------------------------------------------

1 Camper n00b inversed handcoded
2 Resurrection Anton Marsden handcoded


3 Break Stuff Anton Marsden handcoded

4 fairy lights John Metcalf handcoded
5 Bomber Supreme Plus... Geoff Cameron handcoded
6 Fury 2 Roy van Rijn handcoded
7 Unleashed 3 Roy van Rijn handcoded
8 Pioneer Michal Janeczek handcoded
9 Viking Michal Janeczek handcoded
10 ClamTrucks WTN Maezumo
11 ClpTiny P. Kline handcoded
12 Decomposition WTN Maezumo
13 Iridium Cockroach DataWraith YACE
14 Matt Daemon Samy Duc handcoded
15 Dig'n'Die Matthias handcoded
16 Cupid Attoparsec DataWraith YACE


inversed is now leading with Camper n00b the ranking
followed by Anton Marsden's two entries. The scores are
still pretty close, so every new entry could change the
ranking. Let's see what the next days will bring and

fizmo

unread,
Dec 27, 2010, 12:48:32 PM12/27/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the fourth update.

So far I received 19 entries from 12 authors. Twelve
warriors are handcoded, while seven are evolved.
Handcoded warriors still leading the ranking.

Ranking
# Name Author Type
-------------------------------------------------------

1 Mariner Michal Janeczek handcoded


2 Resurrection Anton Marsden handcoded

3 Rosetta Michal Janeczek handcoded
4 S1 Roy van Rijn handcoded
5 Camper n00b inversed handcoded
6 Break Stuff Anton Marsden handcoded
7 fairy lights John Metcalf handcoded
8 Bomber Supreme Plus... Geoff Cameron handcoded
9 P1 Roy van Rijn handcoded
10 Babylonian Roundoff Dave Hillis RedRace
11 ClamTrucks WTN Maezumo
12 Oceania Larvae Dave Hillis RedRace
13 Decomposition WTN Maezumo
14 ClpTiny P. Kline handcoded


15 Dig'n'Die Matthias handcoded

16 Iridium Cockroach DataWraith YACE
17 Cupid Attoparsec DataWraith YACE
18 Matt Daemon Samy Duc handcoded
19 RunningScared P. Kline home-grown

Michal Janeczek's latest entry is now leading the
ranking followed by Anton Marsden. Roy van Rijn jump up
the ranking with his new entry S1, while the recent
leader inversed's Camper n00b is now fifth. The scores


are still pretty close, so every new entry could change

the ranking. Let's see what the last days will bring. I
guess the traffic will start to increase now.
Send your entries.

fizmo

unread,
Dec 28, 2010, 1:20:40 PM12/28/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the fifth update.

So far I received 22 entries from 13 authors.
16 warriors are handcoded, while six are evolved.
Handcoded warriors leading the ranking.

Ranking
# Name Author Type
-------------------------------------------------------

1 Recognition 2010 Zul Nadzri handcoded
2 S1 Roy van Rijn handcoded
3 fairy lights John Metcalf handcoded
4 Voyager Michal Janeczek handcoded
5 Resurrection Anton Marsden handcoded
6 Camper n00b inversed handcoded
7 Break Stuff Anton Marsden handcoded


8 Bomber Supreme Plus... Geoff Cameron handcoded

9 Blottery inversed handcoded
10 Cassini Michal Janeczek handcoded
11 P1 Roy van Rijn handcoded
12 Snowdrift John Metcalf handcoded
13 ClamTrucks WTN Maezumo
14 Oceania Larvae Dave Hillis RedRace
15 Decomposition WTN Maezumo
16 NotScared P. Kline handcoded
17 Iridium Cockroach DataWraith YACE


18 Matt Daemon Samy Duc handcoded

19 Dig'n'Die Matthias handcoded
20 Cupid Attoparsec DataWraith YACE
21 Shrapnel Jam Dave Hillis RedRace
22 ClpTiny P. Kline handcoded

Zul Nadzri's entry is leading the ranking followed by
Roy van Rijn and John Metcalf. Huh, almost every entry
is shanking the ranking because the scores are still
pretty close. Let's see what the last days will bring.
I guess the traffic will increase even more.

Send your entries to

fizmo

unread,
Dec 29, 2010, 2:20:45 PM12/29/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the 6th update.

So far I received 23 entries from 14 authors.
17 warriors are handcoded, while six are evolved using
four different evolvers. Handcoded warriors still
leading the ranking.

Ranking
# Name Author Type
-------------------------------------------------------

1 Ultradian inversed handcoded
2 Y1 Roy van Rijn handcoded
3 Plan 9 inversed handcoded
4 Break Stuff Anton Marsden handcoded
5 Recognition 2010 Zul Nadzri handcoded
6 Resurrection Anton Marsden handcoded
7 G1 Roy van Rijn handcoded
8 VisionsOfThePast Nenad handcoded
9 Snowdrift John Metcalf handcoded
10 fairy lights John Metcalf handcoded
11 Bomber Supreme Plus... Geoff Cameron handcoded
12 ClamTrucks WTN Maezumo
13 Decomposition WTN Maezumo


14 Oceania Larvae Dave Hillis RedRace

15 Spirit Michal Janeczek handcoded
16 Opportunity Michal Janeczek handcoded
17 Adverbial Dilemma Dave Hillis RedRace
18 Iridium Cockroach DataWraith YACE
19 Matt Daemon Samy Duc handcoded
20 Dig'n'Die Matthias handcoded
21 NotScared P. Kline handcoded
22 Elbowing Earthmen DataWraith Galtonium
23 RunningScared P. Kline handcoded

inversed is back leading the ranking followed by
Roy van Rijn and Anton Marsden. Again, there is still
an up and down in the ranking. Let's see what the last
days will bring.

Send your entries to

fizmo

unread,
Dec 30, 2010, 4:05:01 PM12/30/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the 7th update.

So far I received 24 entries from 15 authors.
17 warriors are handcoded, while seven are evolved
using five different evolvers. Handcoded warriors still
leading the ranking.

Ranking
# Name Author Type
-------------------------------------------------------

1 Recognition 2010 Zul Nadzri handcoded
2 Circadian inversed handcoded
3 nameless two Michal Janeczek handcoded
4 Fury Forever Roy van Rijn handcoded


5 Resurrection Anton Marsden handcoded

6 Unleashed Unlimited Roy van Rijn handcoded
7 nameless one Michal Janeczek handcoded
8 Ultradian inversed handcoded
9 Break Stuff Anton Marsden handcoded


10 fairy lights John Metcalf handcoded
11 Bomber Supreme Plus... Geoff Cameron handcoded

12 VisionsOfThePast Nenad handcoded
13 Snowdrift John Metcalf handcoded
14 ClamTrucks WTN Maezumo
15 Decomposition WTN Maezumo
16 Oceania Larvae Dave Hillis RedRace
17 869f1fd7-fa0aa7fa-e3... bvowk Evolved
18 Adverbial Dilemma Dave Hillis RedRace
19 NotScared P. Kline handcoded
20 Iridium Cockroach DataWraith YACE
21 Elbowing Earthmen DataWraith Galtonium
22 Matt Daemon Samy Duc handcoded
23 Dig'n'Die Matthias handcoded
24 RunningScared P. Kline handcoded

Zul Nadzri again leading the ranking followed by
inversed and Michal Janeczek. Still a very close
contest and everything is still possible.

Tomorrow is the final day so don't forget to send your
final entries to

fizmo (at) corewar (.) info

And don't forget:

You can replace anytime until the deadline your
submitted entries. Just state with ;kill which of
your entry you want to replace

The deadline is December 31, 2010 Midnight Berlin time

I'll try to post the final ranking of the contest as
soon as possible sometimes during the coming night.

fizmo

unread,
Dec 30, 2010, 4:24:11 PM12/30/10
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

The deadline is December 31, 2010 Midnight Berlin time

Means in about 24 hrs from now ;-)

fizmo

unread,
Jan 1, 2011, 6:33:21 AM1/1/11
to
********************************************************
** _ _ _____ _____ _____ _____ **
** /_ | | | |_ _/ ____| __ \ / ____| **
** | |___| |_ | || | | |__) | | **
** | / __| __| | || | | ___/| | **
** | \__ \ |_ _| || |____| | | |____ **
** |_|___/\__| |_____\_____|_| \_____| **
** **
********************************************************

********************************************************
** 1st International Corewar Programming Contest **
********************************************************

Here's the final ranking.

17 authors sent 28 entries.
20 warriors were handcoded, while 8 were evolved
using four different evolvers, RedRace, Maezumo, ewon
and Galtonium.

Ranking
# Name Author Type
-------------------------------------------------------

1 Plan 9+ inversed handcoded
2 Circadian+ inversed handcoded
3 Unleashed Unlimited Roy van Rijn handcoded
4 Recognition 2010 Zul Nadzri handcoded
5 Happy New Year Michal Janeczek handcoded
6 Break Stuff Anton Marsden handcoded
7 fairy lights John Metcalf handcoded
8 Mendacious Dahl Dave Hillis RedRace
9 Resurrection Anton Marsden handcoded
10 Snowdrift John Metcalf handcoded
11 Pieface PhilB handcoded
12 VisionsOfThePast Nenad handcoded
13 Final Fury Roy van Rijn handcoded
14 Bomber Supreme Plus... Geoff Cameron handcoded
15 Trifle PhilB handcoded
16 Decomposition WTN Maezumo
17 ClamTrucks WTN Maezumo
18 869f1fd7-100fd975-fa... bvowk ewon
19 869f1fd7-fa0aa7fa-e3... bvowk ewon
20 Altruistic Whale DataWraith Galtonium


21 NotScared P. Kline handcoded

22 Elbowing Earthmen II DataWraith Galtonium
23 Matt Daemon Samy Duc handcoded
24 Mimetic Midwife Dave Hillis RedRace
25 Zwerg lain handcoded
26 Toffee P. Kline handcoded
27 Deathsaw Michal Janeczek handcoded
28 Dig'n'Die Matthias handcoded

And the winner of the 1st International Corewar
Programming Contest and the $50 price money is:

inversed

He's leading with both entries the ranking followed by
Roy van Rijn and Zul Nadzri.

I'll make all entries and the final results soon
available on the contest website:

http://www.corewar.info/tournament/icpc1/

I hope you all have enjoyed the contest.

HAPPY NEW YEAR

Neogryzor

unread,
Jan 1, 2011, 4:09:14 PM1/1/11
to
Oooops, missed it!

sayembara

unread,
Jan 3, 2011, 8:14:13 AM1/3/11
to
Gratz ! :)

inversed

unread,
Jan 7, 2011, 2:23:27 PM1/7/11
to
Typical paper seems overpowered under such rules: 200 processes is
enough to freely replicate, but it is hard to stun it reliably.

fizmo

unread,
Jan 7, 2011, 4:58:50 PM1/7/11
to
On 7 Jan., 20:23, inversed <reo...@mail.ru> wrote:
> Typical paper seems overpowered under such rules: 200 processes is
> enough to freely replicate, but it is hard to stun it reliably.

Hmmm, I thought they will become weaker. But I just used the pap and
pws from optimax benchmark and didn't took a close look to the
different paper styles.

Bjoern

unread,
Jan 7, 2011, 6:34:16 PM1/7/11
to
Sorry I couldn't make it. I kindled some ideas, but I am too out of
touch with the state of the art in Corewar. The competition really
made me want to play again, though.

The learning curve has become steeper, I suppose.

Is there some summary of the latest developments? Explanation of Stone/
Papers?

pauld...@aol.com

unread,
Jan 7, 2011, 8:53:00 PM1/7/11
to
On Jan 7, 1:23 pm, inversed <reo...@mail.ru> wrote:
> Typical paper seems overpowered under such rules: 200 processes is
> enough to freely replicate, but it is hard to stun it reliably.

I am totally surprised! I never thought paper could compete with the
200 process limit since a simple jmp 0 bomb would be an effective
stunner once the limit was reached. And since paper could not compete
there would be no scanners. So I optimized against stones :-(

P. Kline

Robert Macrae

unread,
Jan 8, 2011, 10:42:53 AM1/8/11
to
I would like to wish everyone a prosperous 2011.

And I have a theory.

One of the problems for Silks is the lumpy coverage of a replicator that
looks like
SPL
MOV
...
MOV
JMP
These grow in clusters nose to tail. 2nd generation has two copies back to
back; 3rd generation has 4 sets of processes in a cluster of only 3 copies
of the code; and so on.

Of course most Silks start with a launcher stage like SPL/MOV or
SPL/MOV/MOV but this just delays the problem by a generation or two because
in each generation there is only 1 replicator that still contains the
launcher and an exponentially growing number that don't.

I think this problem is quite severe, and that it is the reason that
replicators containing only SPL/MOV pairs are sometimes used instead of the
SM...MJ pattern. They solve the clumping problem, but at the huge cost that
most of the copies contain only a single SPL/MOV pair that can make only 1
copy -- on its own it can replicate but not multiply.

I think 200 processes may be good for Silks because it allows only 25
8-process replicators. That is high enough for the launcher and first
SPL/MOV pair to obtain good coverage, but low enough to keep coverage good
in later generations when clumping would otherwise become a problem. Once
the limit is reached the SPL becomes ineffective, so though the two MOVs
still write a nose-to-tail pair of copies only one of them gets any
processes.

If I'm right then 200 processes will favour SM...MJ replicators but have
a different (and I think generally negative) effect on SMSMSM...
replicators. These lose their advantage on coverage and are also hit by the
early failure of SPL replication. This results in a higher proportion of
processes in incomplete replicators, though also in the payload that might
compensate under some circumstances.

Regards,
Robert Macrae

0 new messages