WMCCC standings after round 10
Hornets
Just heard from Bruce on ICC:
The top standings after round 10 in the WMCCC.
1. Shredder 8
2. Ferret 7,5
3. crafty 6,5
VirtualChess
Fritz
Nimzo
7. Gunda 6
Dark Thought
Zeus
Francesca
With only one more round to go, it seems that only a NON-commercial
program will be the new World Champion.
It's gonna be either Shredder, or Ferret...
Hornets
Robert Hyatt
Hornets (fu...@iwn.nl) wrote:
: Just heard from Bruce on ICC:
I just heard from Tom. Crafty-Ferret was a draw. In round 10, Crafty
played Nimzo and lost after (according to Tom) playing a very bad Sicilian
opening variation. I had suspected that by the time the tournament ends,
the book would be responsible for 4 of the 11 games being won or lost.
The now-famous 4. Bc4 helped in a couple, then this Sicilian backfired.
If there's one thing that will be different next year, it is the book.
I'm going to start on a learning algorithm, but I'm going to do it with
a severe "twist." I'm going to make the "learned" stuff portable, so that
all the Crafty "clones" playing can supply information for the "original"
to use. :) Of course it will work both ways, and I'll make the format
public so that anyone that's interested can take advantage of the huge number
of games (and resultant "learning") from the many crafty clones. The original
has played nearly 50,000 games on ICC in the past two years, other clones are
also way up there. Might help in choosing better openings for sure...
Bob
Komputer Korner
Robert Hyatt wrote:
> snipped-
Who is the new amateur world micro computer chess champion programmer
for Shredder? And does he plan on taking his program commercial?
--
Komputer Korner
chessman
no so fast..we still need results from RD 11 ....which may have been played
already ..if not ..it would start 9:00 Am Jakarta time (9:00PM EDST)
Komputer Korner <kor...@netcom.ca> wrote in article
<326133...@netcom.ca>...
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: Robert Hyatt wrote:
: > snipped-
:
: Who is the new amateur world micro computer chess champion programmer
: for Shredder? And does he plan on taking his program commercial?
: --
: Komputer Korner
It's not over yet. Shredder is ahead of ferret, but there's another
game left...
Bob
Robert HYATT
On Sun, 13 Oct 1996, Chris Long wrote:
> Big suggestion:
>
> crafty should adjust the book on the fly, perhaps, but also you should
> have a standalone program that takes the base book and the games file and
> updates the book. Clearly, you'll be playing a lot with how exactly
> the book gets updated (based on previous results, rating, evals, etc.),
> and so you'll want to do complete rebuilds every so often. Thus, you'll
> want to keep around the complete games file. Of course, you'll probably
> be tweaking the base book from time to time also.
I have saved every game Crafty has played on ICC, Chess.net, etc. in a
huge PGN database, but have done nothing with it yet, although Mike
Byrne has dome some basic analysis to see what we win with, what we lose
with, etc...
>
> Ideally, the book could include info about general classes of players
> (e.g. play this is player weak, but not if player strong), or even
> *specific* players (don't play this against GM Dlugy).
>
I've thought about this, but the basic problem is that currently a
"position" is 16 bytes in the book. This would *greatly* expand that,
making the current 70 megabyte potentially enormous. The data is
important however. May be that only some moves need such a "comment"
so that the "position" can be dynamic in size perhaps...
> With the rise of internet servers and auto players I think this will
> turn out to be one of the neatest (but challening) aspects of computer
> chess. How to tweak a book when you play 20,000 games a year is a
> nifty question.
> --
As I've said before, it's far easier to produce a high rating if you operate
a program manually, because the number of games is going to be only a
fraction of the total that Crafty/Ferret/WchessX/etc play in automatic
mode. you can also prevent repeat games by forcing the program to vary,
even when it doesn't want to.
This is the most severe weakness in Crafty at present. It plays very well
against any program, so long as it gets a reasonable position out of the
book. However, the randomness I allow to avoid replication also lets it
walk into gross positions as well. In Cray Blitz, we "groomed" the book
over many years to control this. I've decided that I am *not* going to
hand-tune the book any more, it is too hard. Rather, I'm going to (over
time) make this sucker learn how to use the book intelligently. :) Harder
up front, but much better over the long-haul...
> Chris Long, Department of Statistics, Rutgers Univeristy
>
> Score: 0, Diff: 1, clong killed by a Harvard Math Team on 1
>
Komputer Korner
chessman wrote:
>
> no so fast..we still need results from RD 11 ....which may have been played
> already ..if not ..it would start 9:00 Am Jakarta time (9:00PM EDST)
>
> Komputer Korner <kor...@netcom.ca> wrote in article
> <326133...@netcom.ca>...
> > > snipped-
> >
> > Who is the new amateur world micro computer chess champion programmer
> > for Shredder? And does he plan on taking his program commercial?
> > --
> > Komputer Korner
> >
Shredder has at least a tie for 1st place.
--
Komputer Korner
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
Here's partial final standings after 11 rounds:
1. Shredder
2. Ferret
3. Nimzo-3
4. Crafty - Gunda1 (crafty) tie
the rest I don't know about yet... Crafty played Dark Thought in the
last round and (according to Tom) got black again, and played a bad
Sicilian line (don't know if it was a repeat of the game from the previous
round or not where crafty lost). The eval was -1.000 at about move 35,
but it managed to draw, against a program that probably should have beaten
it based on speed (I'd suspect Dark Thought was at least twice as fast
by running on a 400+ megahertz dec alpha workstation.) I won't know more
until I get the log files after Tom returns to the US. If I get some
official results via email with other program standings, I'll post them.
Basically it started with a bust (round 1 time problem), then played well,
then ended with draw-loss-draw. I hope it wasn't something I could have
prevented by being able to modify the book between rounds. Tom wasn't a unix
user, and I didn't try to have him modify the book stuff with VI, I felt that
would simply add too much to what he had to deal with. Whether this was a
mistake or not is unknown, but it was *my* mistake if so...
Bob
Chris Whittington
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>
> Komputer Korner (kor...@netcom.ca) wrote:
> : chessman wrote:
> : >
> : > no so fast..we still need results from RD 11 ....which may have been played
> : > already ..if not ..it would start 9:00 Am Jakarta time (9:00PM EDST)
> : >
> : > Komputer Korner <kor...@netcom.ca> wrote in article
> : > <326133...@netcom.ca>...
> : > > Robert Hyatt wrote:
> : > > > snipped-
> : > >
> : > > Who is the new amateur world micro computer chess champion programmer
> : > > for Shredder? And does he plan on taking his program commercial?
> : > > --
> : > > Komputer Korner
> : > >
> :
> : Shredder has at least a tie for 1st place.
> : --
>
> Here's partial final standings after 11 rounds:
>
> 1. Shredder
Congratulations !
> 2. Ferret
> 3. Nimzo-3
> 4. Crafty - Gunda1 (crafty) tie
fc crafty, gunda1
output nill ?
Chris Whittington
>
Ed Schröder
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>Here's partial final standings after 11 rounds:
>
>1. Shredder
>2. Ferret
>3. Nimzo-3
>4. Crafty - Gunda1 (crafty) tie
Congratualations to Shredder!!
- Ed Schroder -
Komputer Korner
Robert Hyatt wrote:
> 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
Of the top 4 programs, all are amateur programs. Is Crafty the only
one that is available to the public?
--
Komputer Korner
Tord Kallqvist Romstad
Komputer Korner (kor...@netcom.ca) wrote:
Nimzo 3 is a commercial program. It has a rating of 2380 on the SSDF list
(on a P5/90). The first two, I think, are not yet available to the public.
Tord
Alexander Fuchs
Komputer Korner <kor...@netcom.ca> schrieb im Beitrag
<32628D...@netcom.ca>...
> Robert Hyatt wrote:
> > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
> Of the top 4 programs, all are amateur programs. Is Crafty the only
> one that is available to the public?
> --
> Komputer Korner
Nimzo is a commercial program. Version 3.0 is available.
I hope Ferret will soon be available, too ...
Alexander Fuchs
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: Robert Hyatt wrote:
: > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
: Of the top 4 programs, all are amateur programs. Is Crafty the only
: one that is available to the public?
: --
: Komputer Korner
yes. Ferret will likely become a commercial program according to Bruce.
Personally, I think it is one of the best if not the best. Nimzo is a
commercial program already I think. I don't know anything about Shredder,
other than it plays very strongly of course. :)
Bob
Ernst A. Heinz
Bob wrote
> Crafty played Dark Thought in the last round and (according to Tom) got
> black again, and played a bad Sicilian line (don't know if it was a repeat
> of the game from the previous round or not where crafty lost). The eval
> was -1.000 at about move 35, but it managed to draw, against a program that
> probably should have beaten it based on speed (I'd suspect Dark Thought was
> at least twice as fast by running on a 400+ megahertz dec alpha workstation.)
According to Markus who operated DarkThought in Jakarta, the endgame tablebases
saved Crafty ... We intentionally let DarkThought play without Ken's endgame
databases this time. Maybe we change our mind again next year :-)
As for final standings, Markus mailed me the following final standings:
-1- 9.0 Shredder (hearty congratulations, Stefan!)
-2- 8.5 Ferret (sorry that DarkThought "cost" you 0.5 points, Bruce)
-3- 7.5 Nimzo-3
-4- 7.0 Crafty
-5- 7.0 Gunda
-6- 6.5 VirtualChess
-7- 6.5 DarkThought
-8- 6.5 Fritz
Furthermore, a 9 round blitz tournament was arranged on Tuesday, October 15,
resulting in the following five best contenders:
-1- 9.0 Ferret
-2- 7.5 Fritz
-3- 5.5 DarkThought
-4- 5.5 VirtualChess
-5- 5.5 Isichess
Cheers.
=Ernst=
P.S.
Thanks to all supporters of the DarkThought team -- we are still pondering
DarkThought's bad games in rounds 6 and 7 of WMCCC'96 which unfortunately
knocked us out quite early.
+----------------------------------------------------------------------------+
| Ernst A. Heinz, School of CS (IPD), Univ. of Karlsruhe, P.O. Box 6980, |
| D-76128 Karlsruhe, F.R. Germany. WWW: <http://wwwipd.ira.uka.de/~heinze> |
| Mail: <hei...@ira.uka.de> Tel: +49-(0)721-6084386 Fax: +49-(0)721-694092 |
+----------------------------------------------------------------------------+
"It has recently been found out that research causes cancer in rats!"
Komputer Korner
Tord Kallqvist Romstad wrote:
>
> Komputer Korner (kor...@netcom.ca) wrote:
> : Robert Hyatt wrote:
> : > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
> : Of the top 4 programs, all are amateur programs. Is Crafty the only
> : one that is available to the public?
> : --
> : Komputer Korner
>
> (on a P5/90). The first two, I think, are not yet available to the public.
>
> Tord
Then in the ICCA's eyes, there must be some difference between a
commercial program and a professional program. If the 25% rule is
still followed by the ICCA, then that means that if you don't sell
your program at all or if you make less than 25% of your total
income from your program sales then you are a non professional.
In addition, I would classify a program as an amateur program if
the programmer makes no money from his program. I think that it
is a dangerous course for the ICCA to have a dividing line of 25%
of income as the difference between amateurs and professionals. If I
have to pay to obtain Nimzo 3, then in my eyes it is a professional
program no matter what the ICCA rules are. So to sum up There are 4
categories:
1) Professionals who make more than 25% of their income from selling
their program or related chess software
2) Amateurs who make less than 25 % of their total income from
selling their program and related chess software
3) Amateurs who make nothing from their program
4) Amateurs who give their program away
--
Komputer Korner
Robert Hyatt
Komputer Korner (kor...@netcom.ca) wrote:
: Tord Kallqvist Romstad wrote:
: >
: > Komputer Korner (kor...@netcom.ca) wrote:
: > : Robert Hyatt wrote:
: > : > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
: > : Of the top 4 programs, all are amateur programs. Is Crafty the only
: > : one that is available to the public?
: > : --
: > : Komputer Korner
: >
: > Nimzo 3 is a commercial program. It has a rating of 2380 on the SSDF list
: > (on a P5/90). The first two, I think, are not yet available to the public.
: >
: > Tord
:
: Then in the ICCA's eyes, there must be some difference between a
: commercial program and a professional program. If the 25% rule is
: still followed by the ICCA, then that means that if you don't sell
: your program at all or if you make less than 25% of your total
: income from your program sales then you are a non professional.
: In addition, I would classify a program as an amateur program if
: the programmer makes no money from his program. I think that it
: is a dangerous course for the ICCA to have a dividing line of 25%
: of income as the difference between amateurs and professionals. If I
: have to pay to obtain Nimzo 3, then in my eyes it is a professional
: program no matter what the ICCA rules are. So to sum up There are 4
: categories:
I think it's a completely (a) stupid; (b) arbitrary; (c) nonsensical;
(d) ignorant; (d) <your adjective goes here> policy. Why 25% and not 19.5%?
The only logical course of action is to disband that particular topic and
throw it out. And just have a WMCCC event where any *program* can participate
so long as the author supports participation (I still think that (say) Richard
should have the right to say *no genius* should he feel strongly about an
event such as the Jakarta tournament, since it is his program.) The other
logical choice would be that if a program is "for sale" it's commercial.
*period*. The current "non-policy" is simply stupid...
:
: 1) Professionals who make more than 25% of their income from selling
: their program or related chess software
: 2) Amateurs who make less than 25 % of their total income from
: selling their program and related chess software
: 3) Amateurs who make nothing from their program
: 4) Amateurs who give their program away
5) Amateurs that think there should be one real winner, and not enough
different classes so that everyone gets a trophy. :)
Robert Hyatt
Ernst A. Heinz (hei...@ira.uka.de) wrote:
: Bob wrote
:
: > Crafty played Dark Thought in the last round and (according to Tom) got
: > black again, and played a bad Sicilian line (don't know if it was a repeat
: > of the game from the previous round or not where crafty lost). The eval
: > was -1.000 at about move 35, but it managed to draw, against a program that
: > probably should have beaten it based on speed (I'd suspect Dark Thought was
: > at least twice as fast by running on a 400+ megahertz dec alpha workstation.)
:
: According to Markus who operated DarkThought in Jakarta, the endgame tablebases
: saved Crafty ... We intentionally let DarkThought play without Ken's endgame
: databases this time. Maybe we change our mind again next year :-)
I have yet to see the moves posted, but this meshes with Tom's rather
cryptic "Crafty sac'ed a knight to salvage a draw." Sounds like an
EGTB hit could certainly account for this...
Alexander Fuchs
> 5) Amateurs that think there should be one real winner, and not enough
> different classes so that everyone gets a trophy. :)
:-). Good point!
How many trophies were awarded in Jarkata ? I hope every prog got one !!
Alexander Fuchs
Chris Whittington
Komputer Korner <kor...@netcom.ca> wrote:
>
> Tord Kallqvist Romstad wrote:
> >
> > Komputer Korner (kor...@netcom.ca) wrote:
> > : Robert Hyatt wrote:
> > : > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
> > : Of the top 4 programs, all are amateur programs. Is Crafty the only
> > : one that is available to the public?
> > : --
> > : Komputer Korner
> >
> > Nimzo 3 is a commercial program. It has a rating of 2380 on the SSDF list
> > (on a P5/90). The first two, I think, are not yet available to the public.
> >
> > Tord
>
> Then in the ICCA's eyes, there must be some difference between a
> commercial program and a professional program. If the 25% rule is
> still followed by the ICCA, then that means that if you don't sell
> your program at all or if you make less than 25% of your total
> income from your program sales then you are a non professional.
> In addition, I would classify a program as an amateur program if
> the programmer makes no money from his program. I think that it
> is a dangerous course for the ICCA to have a dividing line of 25%
> of income as the difference between amateurs and professionals. If I
> have to pay to obtain Nimzo 3, then in my eyes it is a professional
> program no matter what the ICCA rules are. So to sum up There are 4
> categories:
>
> their program or related chess software
> 2) Amateurs who make less than 25 % of their total income from
> selling their program and related chess software
> 3) Amateurs who make nothing from their program
> 4) Amateurs who give their program away
You forgot category 5. Chris Whittington who shall always have to
pay the ICCA $1000 entry fee whatever. (I have 13.1% of my income from
chess, and haven't published a chess program for more than three
years). I doubt they'll ever be able to swallow not taking the $1000
off me - or its either that or an outright ban, which is what I'm
expecting :)
Seriously the amateur / professional divide and the 'rules' which never
get consistently followed are just a nonsense.
I'ld be much happier if they treated all programs/programmers
equal. If there were limited free tickets or whatever then they
go to 'real' amateurs first; cut the divisive $1000 entry; and pay out
soem of the sponsorship as prize money.
Chris Whittington
> --
> Komputer Korner
Robert Hyatt
Alexander Fuchs (w3fu...@fsrz2.rz.uni-passau.de) wrote:
:
:
: > 5) Amateurs that think there should be one real winner, and not enough
Honestly I don't know. One would be enough, but there's been a "amateur
world champ title" for a few years at least.
Marc-François Baudot
> : >
> : > Komputer Korner (kor...@netcom.ca) wrote:
> : > : Robert Hyatt wrote:
> : > : > 1) Shredder 2) Ferret 3) Nimzo-3 4) Crafty
> : > : Of the top 4 programs, all are amateur programs. Is Crafty the only
> : > : one that is available to the public?
> : > : --
> : > : Komputer Korner
> : >
> : > Nimzo 3 is a commercial program. It has a rating of 2380 on the SSDF list
> : > (on a P5/90). The first two, I think, are not yet available to the public.
> : >
> : > Tord
> :
> : Then in the ICCA's eyes, there must be some difference between a
> : commercial program and a professional program. If the 25% rule is
> : still followed by the ICCA, then that means that if you don't sell
> : your program at all or if you make less than 25% of your total
> : income from your program sales then you are a non professional.
> : In addition, I would classify a program as an amateur program if
> : the programmer makes no money from his program. I think that it
> : is a dangerous course for the ICCA to have a dividing line of 25%
> : of income as the difference between amateurs and professionals. If I
> : have to pay to obtain Nimzo 3, then in my eyes it is a professional
> : program no matter what the ICCA rules are. So to sum up There are 4
> : categories:
>
> (d) ignorant; (d) <your adjective goes here> policy. Why 25% and not 19.5%?
The 25% rule is not even followed: in Paderborn, Virtual Chess was considered a
commercial program (I have no problem with this simple fact alone), but neither
Jean-Christophe, myself or our publisher were near those 25% at that time.
Jean-Christophe and myself have full time jobs which do not leave much time
for chess programming. I think some amateurs have much more time to spend
on their programs than we do, and are also commercially available. Strange.
Commercial availability is the only easy way to recognize amateurs from
non amateurs. But then this might not be a fair criterion for the ICCA to choose
whom to sponsor.
Marc-Francois
Robert Hyatt
Robert Hyatt (hy...@crafty.cis.uab.edu) wrote:
:
Actually I now have all 11 rounds for Crafty. Turns out this game wasn't
a tablebase game at all. here's crafty's search with tablebases turned
off:
White(51):
+---+---+---+---+---+---+---+---+
8 | | | | | | | | |
+---+---+---+---+---+---+---+---+
7 | | | *K| | | | | |
+---+---+---+---+---+---+---+---+
6 | | | | | | | | *P|
+---+---+---+---+---+---+---+---+
5 | | | | | | | | |
+---+---+---+---+---+---+---+---+
4 | *N| | | | B | | P | |
+---+---+---+---+---+---+---+---+
3 | | | K | | | | | |
+---+---+---+---+---+---+---+---+
2 | | *P| | | | | | P |
+---+---+---+---+---+---+---+---+
1 | | | | | | | | |
+---+---+---+---+---+---+---+---+
a b c d e f g h
clearing transposition table
clearing pawn hash tables
time limit 5:48
depth time score variation (1)
6-> 0.12 -0.720 Nb6 Kb3 Kd6 Kxb2 Ke5 Bc6
7 0.41 -0.376 Nb6 h3 Nd5+ Kb3 Nf4 Kxb2 Nxh3 Kc3
7-> 0.48 -0.376 Nb6 h3 Nd5+ Kb3 Nf4 Kxb2 Nxh3 Kc3
8 0.75 -0.230 Nb6 h3 Nd5+ Kb3 Nf4 Kxb2 Nxh3 Kc3
Kd6
8-> 1.14 -0.230 Nb6 h3 Nd5+ Kb3 Nf4 Kxb2 Nxh3 Kc3
Kd6
9 2.64 -0.307 Nb6 h3 Nd5+ Kb3 h5 g5 Nf4 h4 Ng2 Kxb2
Nxh4
9-> 3.00 -0.307 Nb6 h3 Nd5+ Kb3 h5 g5 Nf4 h4 Ng2 Kxb2
Nxh4
10 4.64 -0.127 Nb6 h3 Nd5+ Kb3 Nf4 h4 Nh5 Bf5 Nf6
h5 b1=Q+ Bxb1 Nxg4
10 7.79 ++ Kd6
10 8.54 0.000 Kd6 Kxa4 Ke5 Bb1 Kf4 h4 Kxg4 Bh7 Kxh4
Kb5
10-> 8.70 0.000 Kd6 Kxa4 Ke5 Bb1 Kf4 h4 Kxg4 Bh7 Kxh4
Kb5
time: 8.75 cpu:95% mat:-2 n:775531 nps:92656
ext-> checks:27653 recaps:1063 pawns:45459 1rep:132
endgame tablebase-> probes done: 923 successful: 0
Turns out that Crafty's draw detection code noted that this is a rook pawn
+ wrong-colored bishop and saw that it was drawn without the tablebases at
all... It probably did hit one, but it didn't really matter. In looking
through this, the game was interesting at times, but looked more and more
drawish at the end. In any case, Kd6 is certainly an odd-looking move that
gives up the knight, but which also forces a dead draw. In the above output,
it probed the tablebases 923 times, but none were present so there were no
hits. Normally these two numbers are identical because it doesn't probe
unless it knows there will be a hit...
Robert Hyatt
Marc-François Baudot (ka...@micronet.fr) wrote:
:
: The 25% rule is not even followed: in Paderborn, Virtual Chess was considered a
: commercial program (I have no problem with this simple fact alone), but neither
: Jean-Christophe, myself or our publisher were near those 25% at that time.
: Jean-Christophe and myself have full time jobs which do not leave much time
: for chess programming. I think some amateurs have much more time to spend
: on their programs than we do, and are also commercially available. Strange.
: Commercial availability is the only easy way to recognize amateurs from
: non amateurs. But then this might not be a fair criterion for the ICCA to choose
: whom to sponsor.
:
: Marc-Francois
I whole-heartedly agree. Time to drop that title, as there are lots of
very strong amateur programs and they don't need to be "shielded" from going
head to head with the commercial programs. I think that might have been the
original reason, not to mention the entry fee, but it's time to rethink and
eliminate that...
Tim Mirabile
hy...@crafty.cis.uab.edu (Robert Hyatt) wrote:
>I think it's a completely (a) stupid; (b) arbitrary; (c) nonsensical;
>(d) ignorant; (d) <your adjective goes here> policy. Why 25% and not 19.5%?
The whole idea of basing it on a percentage of income is silly. What if
Microsoft bought the rights to a "real" chess program, and sold a few
million copies. Their income from it would be _far_ less than 25%, but
it's clearly a commercial program.
>The only logical course of action is to disband that particular topic and
>throw it out. And just have a WMCCC event where any *program* can participate
>so long as the author supports participation (I still think that (say) Richard
>should have the right to say *no genius* should he feel strongly about an
>event such as the Jakarta tournament, since it is his program.)
Perhaps this should be added to the standard license agreement - that the
program may not be entered in any public tournament without permission.
>The other
>logical choice would be that if a program is "for sale" it's commercial.
>*period*. The current "non-policy" is simply stupid...
Agreed.
+----------------------------------------------------------------------+
| Tim Mirabile <t...@mail.htp.com> http://www.webcom.com/timm/ |
| TimM on FICS - telnet://fics.onenet.net:5000/ PGP Key ID: B7CE30D1 |
+----------------------------------------------------------------------+
brucemo
Chris Whittington wrote:
> fc crafty, gunda1
>
> output nill ?
>
> Chris Whittington
Nobody was able to figure out what they did to Crafty to make
Gunda. In my game it played a King's Gambit, which is weird
for Crafty. The operator of Gunda didn't speak English, so
questions hit a wall. In the heat of the moment I didn't
consider getting an interpreter.
bruce