9th Computer Olympiad 2004

Date: July 4-8, 2004

Place: Ramat-Gan, Israel

Operator: Jan Willemson

Version: GNU Go 3.5.8

19x19 Results:

Rank Name Go Int MFoG Indigo GNU Jimmy Wins
1 Go Intellect 1 1 0 0 1 1 1 1 6(*)
2 The Many Faces of Go 0 0 1 1 1 1 1 1 6
3 Indigo 1 1 0 0 1 0 1 1 5
4 GNU Go 0 0 0 0 0 1 1 1 3
5 Jimmy 0 0 0 0 0 0 0 0 0

(*) Go Intellect won the play-off.

9x9 Results:

Rank Name Int GNU Mag Ind MFoG Neu Ata Dum GoK Wins
1 Go Intellect 1 1 1 1 0 0 1 1 0 0 1 0 1 1 1 1 11(*)
2 GNU Go 0 0 0 1 1 1 1 0 1 1 1 1 0 1 1 1 11
3 Magog 0 0 1 0 0 1 1 1 1 1 1 0 1 0 1 1 10
4 Indigo 1 1 0 0 1 0 1 0 1 0 1 0 0 1 1 1 9
4 The Many Faces of Go 0 0 0 1 0 0 0 1 1 0 1 1 1 1 1 1 9
4 NeuroGo 1 1 0 0 0 0 0 1 0 1 1 1 1 1 0 1 9
7 Atarist 0 1 0 0 0 1 0 1 0 0 0 0 1 1 1 1 7
8 DumbGo 0 0 1 0 0 1 1 0 0 0 0 0 0 0 0 1 4
9 GoKing 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 2

(*) Go Intellect won the play-off

Game records can be found in regression/games/olympiad2004.

External link: http://www.cs.unimaas.nl/olympiad2004/results.html

Excerpts from the mailing list:

To: gnugo-devel@gnu.org
Date: Sat, 03 Jul 2004 00:19:33 +0200
From: Gunnar Farnebäck
Subject: [gnugo-devel] GNU Go in the Computer Olympiad 2004

GNU Go will participate in the Computer Olympiad 2004, July 4-8. Many
thanks to Jan Willemson who has volunteered to operate GNU Go at the
Olympiad and to David Doshay who generously paid the entrance fee.
Information about the tournament, including playing schedule, can be
found at http://www.cs.unimaas.nl/olympiad2004/. With some luck Jan
will be able to give progress reports from the tournament 

As most of you probably remember, GNU Go is reigning champion in the
19x19 tournament. This might be difficult, but not impossible, to
repeat this year with Many Faces of Go among the opponents. In total
there are five programs participating:

Indigo
Go Intellect
The Many Faces of Go
Jimmy
GNU Go

In recent tournaments we have had good results against Indigo, Go
Intellect, and Jimmy. Many Faces, on the other hand, still leads by
2-0 (*) in tournament games. Then again it was some time since they
last met. I would say that Many Faces has to be favorite, followed by
GNU Go and Go Intellect. But I'll be an optimist here and predict the
following results:

1. GNU Go
2. Many Faces
3. Go Intellect
4. Indigo
5. Jimmy

The 9x9 tournament is bigger and looks very open. The contestants are:

Indigo
Go Intellect
NeuroGo
The Many Faces of Go
DumbGo
Atarist
Magog
GoKing
GNU Go

The winner from last year, Aya, is missing and the remaining four
programs from last tournament, Neurogo, Go Intellect, Magog, and
Indigo placed 2-5 with very close results. GNU Go played evenly with
Neurogo on 9x9 in the Olympiad two years ago and Many Faces is almost
certainly in the contention as well. The remaining three programs,
DumbGo, Atarist, and GoKing are unknown entities to me. However, based
on a Computer Go List message, and the name, DumbGo can be supposed to
be weak. I'm not sure that the name Atarist inspires much confidence
either, but I have no real information about that program, nor about
GoKing.

I'll make a wild stab and predict the following results:

1. NeuroGo
2. Magog
3. GNU Go
4. Many Faces of Go
5. Go Intellect
6. Indigo
7. GoKing
8. Atarist
9. DumbGo

/Gunnar

(*) The only tournament games between Many Faces and GNU Go I can find
are from US Go Congress 1999 and 21st Century Cup 2002. They also both
played in 21st Century Cup 2001 and Gifu Challenge 2003, without
meeting each other.
From: Jan Willemson
To: GNU Go development <gnugo-devel@gnu.org>
Subject: Re: [gnugo-devel] Latest news from the Olympiad 
Date: Sun, 11 Jul 2004 13:42:23 +0300 (EEST)

> > Yesterday's unfinished game was decided by the Jury to the favour of
> > Indigo,
>
> Annoying, but mostly because GNU Go misread. It would still be
> interesting to know how the jury reasoned.

The game was sent for decision to Martin Müller (without any information
about the players, of course) and his decision was -- whoever moves, wins.
Since it was white's (i.e. Indigo's) move, Jonathan Schaeffer as the Jury
chairman decided to give the point to Indigo. Additional supporting
reasoning was that if all the stones in the unfinished corner are
considered alive, white wins as well.

[...]

Jan
From: Gunnar Farnebäck
To: GNU Go development <gnugo-devel@gnu.org>
Subject: Re: [gnugo-devel] Latest news from the Olympiad 
Date: Tue, 13 Jul 2004 02:54:25 +0200

I wrote:
> > Yesterday's unfinished game was decided by the Jury to the favour of
> > Indigo,
> 
> Annoying, but mostly because GNU Go misread.

See the end of regression/games/olympiad2004/19x19/gnu-ind.sgf.

The problem is in the lower left corner:

12 O O O O O X X . O O X
11 O X X X O O . X X X O
10 X X X X O . X X O O O
 9 . X O X O O O X O X X
 8 O X O O O X O O O X X
 7 . X X X X X O . O O X
 6 . . X . . X O O X X X
 5 . X O X O O X X X X X
 4 X O O + . . X O X O O
 3 X O . O O . X O X O O
 2 X O O . X . X O X X X
 1 . O . X . . . . . . X
   A B C D E F G H J K L

Black playing first can of course win the semeai trivially by making
two eyes with A7, A6, or A5. I think black E4 would also win the
semeai, although not quite so easily.

The problem is that white playing first is not entirely dead. If black
tries to keep white dead, the combination of white A1 and A7, followed
by D4, leads to a ko for the life and death of both. Black can
sacrifice the A4 string to gain certain life but in that case white
also lives, so it would be a failure.

Also interesting is the variation W A6, B A5, W A7, leaving black with
only one eye, but in this case black eventually wins the semeai. Still
we would want the owl code to see this attack, so that the semeai code
will get to analyze it too. The problem here, however, is that we have
a case of "eyeshape with cutting point",

12 O O O O O X X . O O X
11 O X X X O O . X X X O
10 X X X X O . X X O O O
 9 . X O X O O O X O X X
 8 O X O O O X O O O X X
 7 O X X X X X O . O O X
 6 O . X . . X O O X X X
 5 X X O X O O X X X X X
 4 X O O + . . X O X O O
 3 X O . O O . X O X O O
 2 X O O . X . X O X X X
 1 . O . X . . . . . . X
   A B C D E F G H J K L

which have troubled us a long time. GNU Go thinks black is locally
alive in seki here, which of course is wrong. Paul, should we try to
revive your old patch for this problem?

Another thing to investigate here is that there seems to be some kind
of persistent cache problem. When the position is loaded GNU Go wants
to play black A5 to eliminate a combination attack related to the
semeai, but that doesn't happen when replaying the full game.

/Gunnar