Thursday, May 31, 2012

Chess960: Chess education in the 21st century thanks to Mr Botvinnik

I grew up in an isolated part of the world that had no history or culture of playing chess. None of my family played chess or were even vaguely interested, none of my friends played it, nothing at school, no chess training colleges and the internet was only used for academic and military purposes. So I had to learn it from dusty old books, some of which I still cherish today.

Let me say something up front. I'm not a fan of the Botvinnik school! Personally I think the Botvinnik system of chess education is a curse upon chess and we will have to wait at least another couple of decades until the old fashion/imperialist/science oriented approach to chess education is finally out of the system. Unfortunately Kasparov and Kramnik are products of the Botvinnik school and as such are still supporters of it.

The Botvinnik system is responsible for the corrupt idea that it is good to study just a few Chess openings to very great depth. It has industrialised chess education. Yes it has been a great success, but it has been too successful because it has turned chess into a nerdy memory intensive male dominated game. Chess960 led by Bobby Fischer is actually a reaction against the Botvinnik approach I think.....

And here is the irony. I'm actually interested in computer chess just as Mr Botvinnik was! I started Chess too late when I was twenty years old after seeing Chessmaster 8000 in the shop window of the local game store. The love for Chess began.....better late than never. However reading theoretical books and combining it with a poor chess engine is not a good way to learn!

The problem was that the programmers did not properly dumb down Chessmaster and it would not emulate typical human tactical blunders. So I never really had it etched into my vision to see the typical patterns of tactical opportunity. Once I started playing real people, I wasn't seeing the typical tactical blunders that they and I make on a regular basis.

For me the journey has been to learn how to code in C++ and learn how to compile software, just so that I could materialize the engine I was always looking for as a young adult. Although these days with the internet there is not much use for a training chess engine, it's been a project.

This Stockfish960 training engine I've modified is almost the completion of everything I ever wanted in a training engine and it could not have been possible without the generosity of time of a community of people. I've had to hybridize the skill level concepts of Houdini and Stockfish to do it.
  1. It plays decent 960 openings for a couple of moves
  2. It makes decent tactical blunders to learn from
  3. It's struggles like a human in complex sharp tactical positions
  4. It improves like a human in quiet positions
  5. It absorbs time on the clock much like a human
  6. It allows active chess to be played against it
  7. It improves it's play just enough so that the endgame is a challenge for learning
  8. It's skill level advances so that you know where your limits are
  9. It will resign when the position is lost and it will accept a draw
The problem with the major engines out there today are these:
  1. Houdini patronizes the player by playing it's move instantly
  2. Houdini limits the depth but does not improve in the endgame much
  3. Houdini plays a lot of odd blunders that a human wouldn't make
  4. Stockfish weakens it's play positionally, but not so much tactically
  5. None of the engines play a decent Chess960 opening
The two issues that I still have to work on in Stockfish960 training engine are these:
  1. Very occasionally due to random chance, the Stockfish method for worsening the positional evaluation of the engine means that the engine might play a couple of non-human looking moves. It's a real challenge to remove these when they do happen because it is not a simple problem to perfectly simulate mediocrity with an engine.
  2. Houdini is absolutely brilliant at developing the queen in Chess960, but the Stockfish engine is wanting in that area. That will be a challenge to improve and hopefully the open source developers of Stockfish will address this issue one day. However for the purposes of chess960 training, it's not a big issue.
Anybody want to start an anti-Botvinnik revolution? 

:-) 

Enjoy 960

Saturday, May 26, 2012

Chess960: two move ideas no. 5

Here black has decided not to spend the time calculating the effects of symmetrical play after 1.f4 ...f5 and is not particularly inspired by 1.f4 ... Nb6. So instead of other possible replies to 1.f4 such as 1...b6 or 1...c5!?/Bb6 or 1...f6, black comes up with a nice harmonious gambit idea instead:

SP357 white to play: trace back the moves and try to see black's ideas

Enjoy 960
Hint: Bb6, Nc7, d5/d6, f5/f6 not in any particular order

Thursday, May 24, 2012

Knowledge is power or protection?

Over at Mark Week's blog I made an interesting but I think inaccurate comment that I thought Chess960 was struggling for these reasons:
  1. Ageing demographic in the western world nations and in China. The older community are reluctant to play Chess960 because it is new and different
  2. The ever decreasing young demographic is being drawn into the computer gaming mass market and therefore Chess960 is outcompeted
However I think that this is more accurate but not exactly correct either:
  1. The ageing demographic in the western world nations and in China is actually a good thing for Chess960. This is because the older demographic like Chess960 because it is less memory intensive and they can still play Chess focusing more on concepts and a life time of chess experience, rather than speed memory and prearrangement.
  2. The ever decreasing young demographic actually love traditional chess because they can utilise their young, sharp and speedy memories to great effect in the opening by using existing chess theory that is freely available especially in the internet age. Traditional chess is actually ideal to the young cohort (especially in youthful countries like India), because blitz chess is especially suited to memory prearrangement.
However I think there is even a better explanation, that is so simple, it is essentially Occam's Razor and I think is at the very heart of why Chess960 is slow on the uptake:

Knowledge is not power,
Knowledge is protection.



When we try to protect ourselves with knowledge, it appears like we are more "powerful" but actually what we do is to destroy diversity and simplify our lives in order to feel safe. But is that actually helpful to us?

Cheers

Chess960: Angelic start positions, where is the devil?

Here is an amazing SP. Almost nothing white does causes any stress for black at all. It feels like black is in heaven because no matter what white does, black has got a perfect reply:
SP459: How does white light the fire in this position?

The features that make this SP so "heavenly" to play for black:
  1. After 1.g4 black can play by symmetry with no apparent suffering. The rooks are exchanged off and symmetry continues. When it finally breaks, black seems completely fine.
  2. The queen can develop to f4/f5 and then to a4/a5 without every being attacked or threatened.
  3. White does not have a starting move that comes to anything! a4?! b4?!
  4. White 1. c4 ... g5!
  5. White 1. d4 ... g5!
  6. White 1. e4 ... g5!
  7. White 1. f4 ... g5!
  8. White 1. g5 ... g5!
  9. White 1. Nd3 ... g5! Nc5 (mate #1) Nd6!
  10. As soon as the military knights develop, they simply get in the way of the pawn front line. But if the pawn front line is developed first, black develops more quickly than white can.
  11. Incredibly, all the mating attacks are easily extinguished including Nc3/b5+Qf4 attacking c7 mate#
Heaven for black?

Enjoy 960

PS) Ok have checked it in a bit more detail. This is actually a really slow and creepy SP. The techniques seem to be to advance the kingside pawns threatening to break through on that side. Develop the military knights NOT to c3/d3 but b3/c3 and both sides castle queenside. This gives white the option of d3. Then white slowly starves black of good moves (kind of a very high level zugswang approach) and incrementally improves their own position until black is forced to play a truly weakening move. The slow strangling of black is accelerated if white can threaten to break through the pawns on the kingside as a long term threat as a function of having one extra tempo.

Sunday, May 20, 2012

Chess960: two move ideas no. 4

Here black has decided to play the opening actively with a beautiful pawn gambit idea that allows black a lot of freedom to develop quickly with the promise of a queen-side attack. You can notice the similarity of this idea with idea no.2 in this blog.

SP458 white to play: trace back the moves and try to see black's ideas

Enjoy 960
Hint: g6, h5, Bg7, Bg6, Nd6, O-O-O (not in any particular order)

The Art of Chess960

Hi there
No words required, just enjoy the symmetry that exists for a little while:

SP412 Move 5, black to play

Enjoy 960

Tuesday, May 15, 2012

Chess960: two move ideas no. 3

In the first two moves, black has thought of a nice idea to equalize since the c5 square is safe. You can notice a similarity with idea no.1 in that there is no serious downside in developing a major piece early.

SP928 white to play: trace back the moves and try to see black's idea

Enjoy 960
Hint: Rxc5, b6, f6, Ne6, Qf7, O-O

Chess960: two move ideas no. 2

In the first two moves, black has found a beautiful gambit idea, because developing the queen early has no serious downside:

SP522 black to play: trace back the moves and try to see black's ideas

Enjoy 960
Hint: d5, Qxf5

Chess960: two move ideas no. 1

Considering that many Chess960 players don't speak English, here is the beginning of a series of diagrams that contain ideas from the first two moves that can be seen without using many words to describe them.

SP785 black to play: trace back the moves and try to see white's ideas


Enjoy 960
Hint: e4, f5

Tuesday, May 8, 2012

Chess960: If you love genetics you will love 960

I do not understand why more geneticists and scientists are not playing Chess960. They would love it. Buried in the sequence of all Chess960 positions, are clusters of thematic positions. Tonight I played SP456 a very easy number to remember! This SP has to be one of the most hideously complex positions in the whole set of 960. Why is that? Here is the position:

SP456: Complexity!














The diagram on the right shows where you will find the KQR sequence on the king-side in the set of Chess960 positions. The diagram is rotated to fit on this page so SP001 is top left and reads down then across. In the diagram you can just see where the dark splotches represent the KQR positions. There are only 18 of them! If you take the mirror of them RQK then there are 36 positions, but Chess960 is not actually symmetrical about the half way because of the castling rule, and so I'm reluctant to include the mirror positions.

The amazing thing about the KQR sequence on the kingside edge are these features:
  1. There are probably at least 12 possible pawn moves to start the game that are all feasible, and must be ruled out by merit. Incredibly, every single pawn is defended which is pretty special.
  2. The a-pawn is particularly effective in this SP. This is because it develops a bishop on a2, it is backed up by a rook already, and if it is pushed up the file a4-a5 it potentially biffs a knight on b6.
  3. The green lines show that the bishops are poised to make some deadly attacks on the KQR sequence and the pawn move a3 or a4 followed by Ba2 attacking black's queen on g8, is a surprisingly effective development sequence
  4. It is difficult to know how to develop the queen (so what's new!)
  5. King-side castling is a high probability, and so the other side of the board and the center become major themes
  6. All the heavy weight pieces are concentrated on one side of the board, leaving all the minor pieces on the other side, and a lone undefended rook. Thus the queenside is a minefield of potentially undefended squares and holes in the position if it is not properly secured.
  7. There is a possibility to rapidly push out the kingside pawns f5/g5/h5 and not even castle at all.
The thing about SP456 is this, because all the minor pieces are concentrated on one side of the board, the chess960 players have to achieve three important high level tasks:
  1. They must structurally develop the queen-side to secure it from possible long term structural weaknesses
  2. They must structurally develop the queen-side to best develop the minor pieces
  3. They should be mindful of the powerful attacks that reflect back at the king-side.
Objectives 1 and 2 above are actually competing against each other. Yes it is possible to build a beautifully secure pawn structure on the queenside, but then the minor pieces won't be developed properly. It's a compromise and a potential minefield.

Enjoy 960.

Note: Queenside means the same as it does for traditional chess, except that the queen is not actually on the queenside!

Thursday, May 3, 2012

Chess960 & Double Chess960 for Winboard


Here is a modification of the wonderful Winboard software that I slightly modified to be more useful as a dedicated Chess960 application for single player verses engine training exercises. It includes a modified version of the latest Stockfish engine that plays better Chess960 openings.

A big thanks goes to the developers of Winboard and Stockfish, who have made their code open source for the rest of us to enjoy. I update the code from time to time with new ideas, so check back for the latest version.

Winboard-960 can be downloaded here:
For information on the Stockfish960Trainer engine, click *here*





Credits:

 * Winboard Copyright 1991 by Digital Equipment Corporation, Maynard,
 * Massachusetts.
 *
 * Enhancements Copyright 1992-2001, 2002, 2003, 2004, 2005, 2006,
 * 2007, 2008, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
 *
 * Enhancements Copyright 2005 Alessandro Scotti
 *
 * XBoard borrows its colors and the bitmaps.xchess bitmap set from XChess,
 * which was written and is copyrighted by Wayne Christopher.
 *
 * The following terms apply to Digital Equipment Corporation's copyright
 * interest in XBoard:
 * ------------------------------------------------------------------------
 * All Rights Reserved
 *
 * Permission to use, copy, modify, and distribute this software and its
 * documentation for any purpose and without fee is hereby granted,
 * provided that the above copyright notice appear in all copies and that
 * both that copyright notice and this permission notice appear in
 * supporting documentation, and that the name of Digital not be
 * used in advertising or publicity pertaining to distribution of the
 * software without specific, written prior permission.
 *
 * DIGITAL DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING
 * ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT SHALL
 * DIGITAL BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR
 * ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS,
 * WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
 * ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS
 * SOFTWARE.
 * ------------------------------------------------------------------------
 *
 * The following terms apply to the enhanced version of XBoard
 * distributed by the Free Software Foundation:
 * ------------------------------------------------------------------------
 *
 * GNU XBoard is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or (at
 * your option) any later version.
 *
 * GNU XBoard is distributed in the hope that it will be useful, but
 * WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
 * General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program. If not, see http://www.gnu.org/licenses/

Stockfish, a UCI chess playing engine derived from Glaurung 2.1
  Copyright (C) 2004-2008 Tord Romstad (Glaurung author)
  Copyright (C) 2008-2012 Marco Costalba, Joona Kiiski, Tord Romstad

GNU GENERAL PUBLIC LICENSE

Enjoy 960