Davis/Sacramento Go Club

Volume 3 Issue 3 - October 1997

Next Tournament:

Membership Update:

As of 9/14 we have 32 paid members, up from 3 in June. The club welcomes new members: Kinuko Yamamoto, 5k, Sueng Ho Cho, 2d, and Nick Ballard, 5d, and renewals John Hughes and Gary Hynes.

Re: AGA Membership

The club is an American Go Club affiliate. If you are interested in AGA, write: American Go Association, P.O. Box 397, Old Chelsea Station, New York, NY 10113.

1997 Numbers:





Number of members:



Davis/Sacramento Go Club
c/o Frank Berkenkotter
Box 4, Guinda, CA 95637

The Club Committee:

Frank Berkenkotter


Will Haynes


Steve Burrall


Peter Kent


In Memory of Peter Redmond

We lament the passing of Peter Redmond, our friend, fellow club member, and worthy opponent for these many years. Our heartfelt condolences go to None Redmond and his family.


On September 13th, the D/SGC held its third tournament of 1997 at the CSUS in Sacramento. There were 16 players in 2 divisions. In Division I in some tough games with two new dan players, Steve Burrall, by virtue of his win over Peter Kent, won the division. In Division II, John Berdner, 8k (promoted to 7k), broke a tie with Frank Berkenkotter to win the division. A side note, young Matt Burrall scored two victories to join a four way tie for fourth through seventh.

Division I

  1. S. Burrall 4D


  • Kent 1k
  • 3-1

  • Ballard 5D
  • 2-2

  • Cho 2D
  • 2-2

  • Haynes 1k
  • 1-2

  • Shiah 2k
  • 0-3

    Division II

    1. Berdner 8k*


  • Berkenkotter 3k
  • 3-1

  • Newmiller 3k
  • 3-1

  • Hopkins 4k
  • 2-2

  • Murphy 6k
  • 2-2

  • Newbold 5k
  • 2-2

  • M. Burrall 19k**
  • 2-2

  • Kreiss 9k
  • 1-3

  • Charles So 3k
  • 1-3

  • Hughes 7k
  • 1-3

    *-- promoted to 7 kyu
    ** -- promoted to 17 kyu

    Club Standings

    The club gives a free membership and a cash prize with the most wins in 3 or more tournaments in 1997 in Div. I and II. Last year's winners were Will Haynes, 8 wins, and Louis Abronson, 13 wins.

    Division I

    1. Burrall 4D

    7 wins

  • P. Kent 1k
  • 6 wins

  • Haynes 1k
  • 5 wins

  • 4 players tied
  • 4 wins

    Division II

    1. Newbold 5k

    8 wins

  • Newmiller 3k
  • 7 wins

  • Sniegowski 6k
  • 7 wins

  • Berkenkotter 3k
  • 5 wins

  • 3 players tied
  • 4 wins

    U. S. Go Congress

    By Steve Burrall

    Kristen, Matt and I went to the '97 U. S. Go Congress and had an absolutely wonderful time. After a rough start in the non-air-conditioned dorms during a heat wave, things cooled off and went smoothly by the third day. There were so many planned events that we had trouble keeping track of where we were supposed to be and getting enough sleep at night! The children's program was outstanding and included lessons from various pros everyday. These included Jim Kerwin, Janice Kim, Jimmy Cha, Michael Redmond and others. The day-off tour of Gettysberg led by assistant tournament director and walking encyclopedia Keith Arnold was so good that I was inspired to read several history books afterward and I've got more on the shelf I can't wait to get into. We also enjoyed early morning runs with AGA president Phil Strauss and afternoon swims and soccer games for the kids. I didn't win any prizes, but Matthew won the children's handicap tournament and Kristen was on a winning team in the team Go tournament. Next year's Congress is in Santa Fe and we'll be sure to attend.

    U. S. Go Congress - Part 2

    by Willard Haynes

    The U. S. Go Congress was the last part of my trip to the east coast this summer. I had a very good time at the game analyses and lectures, particularly those by Jimmy Cha. I did moderately well over the board (I won about half of my games), but I hope I learned a lot.

    I went on the trip to Gettysberg, also. It was as good as Steve says, but he forgot to mention that it rained all day long. Despite this the trip was well worth it.

    Report From the Euro-Go Congress

    by Frank Berkenkotter

    This year's event was held in Marseilles, France, from July 19th to August 2nd, at Luminy Univ., an institution devoted to medical and physics research. The tournament winner was Lee Hyuk, 7d, of Korea and Moscow, with a 10-0 record. Second was Washari Wataru, 6d, of Japan and Paris, with a 9-1 record, and third was last year's winner, Guo Juan, 8-2, and a testament to the toughness of the competition.

    Attendance this year was 674 players and another 150 family members. The location was about 40 minutes walk from the Mediterranean Ocean through the "Calanques", a white chalky geologic layer seen throughout much of Western Europe including the well-known "White Cliffs of Dover". The location is also about 40 minutes form downtown Marseilles, dominated by the bay, water, seafood, a large North African population, and a more southern Provance oriented France.

    My record of 3-7 at the Congress was of course not without its "if onlys" and long explanations. As someone said of Go, progress in it presents very individual hurdles to individuals. Mine is quelling a nervous urgency and allowing a relaxed sense of contemplation to take control. I'm on lesson two of many chapters.

    On the social front, the Congress was a complete success. We played Go and backgammon every night, saw many old friends and on my last day in Paris, I played a game of Go in Club Lescot, the home of the Paris Go Club. In '98 the Congress will be in Romania and in '99 in Bratslava, Slovakia.

    Computer Go

    by Peter Kent

    I have been working for a little over a year on the problem of how to make a computer play go. I have a Macintosh Quadra 800, and I am using the Metrowerks Codewarrior programming environment for that purpose.

    The general programming strategy I am using is a balance between iteratively updating my program with new abilities as I think them up, and working out solutions in a broader sense, then programming it all in. I have written programs which do nothing but test ideas to make sure they can be programmed, and will work as predicted. So far, one or another of the programs I have written can: identify "groups", which for me are the sets of stones which have to live or die together; track which group is adjacent to which, and the number of connections between them, as well as the number of liberties of each; tell whether a move on the go board is legal or not, and allow the legal ones, updating the board position and remembering kos; play a game of go with itself, choosing legal moves randomly; label with a number series the pattern of stones in the vicinity of, and centered on, each empty point, the pattern identified being dependent on the pattern's shape but independent of the orientation of the pattern.

    I have considered four ways to approach the problem of "teaching the computer" all the knowledge it needs to play go well. They are: explicitly programming each concept into the computer; setting up a graphical interface tool for transferring knowledge from human to computer with clicks of the mouse; programming the computer to look through and learn from human game records; and setting up the computer to teach itself how to play by having it play games with itself and improve its set of strategies. I have not eliminated any of these as possible approaches for me since they all require the same underlying knowledge of the game in order to succeed. Their important differences are the differences in time and money needed for their implementation, and I don't know yet how different they are in those respects - I'm still working on the underlying knowledge! For now though, I am working mostly toward the first two, since they involve a tighter feedback loop between me and the program, i.e. they're not as potentially time consuming if they get out of control.

    [This article will be continued in the next issue of the newsletter.]


    by F. Berkenkotter

    Questions for the membership:

    Should we always have three divisions or should the number of divisions be based on the number of entrants? Are there any problems with proposed 1998 schedule? Is it okay to have 2 tournaments on Saturday and 2 on Sunday, or should they all be on either Saturday or Sunday? CSUS is only available for the length of time that we need on Saturdays. Please provide your suggestions by November 30.

    Proposed 1998 Schedule

    Modified last on November 16, 1997 by Jeff Newmiller.
    (Comments and suggestions welcome.)

    Davis/Sacramento Go Club Homepage