Catalogue text:

Endgames without pawns occur frequently in practice and are extremely tactical in nature. John Nunn has extracted the most important information from endgame databases and presented it in the form of guidelines and specific key positions. With most competitive games nowadays being played to a finish in a single session, this knowledge may prove invaluable over the board. This new expanded edition features the most important information about six-man pawnless endings, for which the databases did not exist at the time of the first edition.

Katalogangabe zur Ausgabe, die bisher (1996-2001) bei Batsford verlegt wurde:
Following on from his successful book "Secrets of Rook Endings", John Nunn turns his attention towards endgames without pawns. These occur surprisingly often in practical play and are extremely tactical in nature. These endings have - until now - defied attempts at definitive analysis.
This book unites man and machine in the search for ultimate answers. The man, Dr. John Nunn, has been one of the world's leading chess players for more than a decade. The machine is the computer database created by Ken Thompson of Bell Laboratories, to check every detail of the analysis of these hitherto baffling endgames.

• Comprehensive coverage of all significant four- and five-man endings without pawns
• Full of instructive ideas and important new insights
• Every move checked by computer database
• Written by one of the world's foremost endgame theoreticians
• Essential knowledge for competitive players, with games increasingly being played to a finish

_________________________________________________

Following on from his successful book Secrets of Rook Endings, John Nunn turns his attention towards endgames without pawns. These occur surprisingly often in practice and are extremely tactical in nature.
This book unites man and machine in the search for ultimate answers. The computer databases created by Ken Thompson, formerly of Bell Laboratories, can state with certainty the correct result of any position with five pieces or fewer. John Nunn has extracted the most important information from these databases and presented it in the form of guidelines and specific key positions, which can be more readily digested by the human mind. With most competitive games nowadays being played to a finish in a single session, this knowledge may prove invaluable over the board.
Since the first edition of this book was published, the databases for many six-man endings have been created, resulting in some surprising and paradoxical discoveries. The coverage has therefore been expanded to include the most interesting features of these endings.

John Nunn is a grandmaster from England. He has won four individual gold medals and three team silver medals at Chess Olympiads. In the Chess World Cup of 1988/9, he finished sixth overall, ahead of several former World Champions. He is arguably the most highly acclaimed chess writer in the world, with two of his books receiving the prestigious British Chess Federation Book of the Year Award.


Introduction to the Second Edition
The first edition of this book, which was published in 1994, was the second volume in my trilogy exploring 5-man endgames (the first was Secrets of Rook Endings, now also available in a revised and expanded version from Gambit Publications; the third was Secrets of Minor-Piece Endgames). After some years Secrets of Pawnless Endings went out of print, but now I am delighted to be able to offer this new edition. The main changes are the 62 pages of extra material covering 6-man endgames and the lower price.
The original endgame trilogy was a huge step forward in endgame literature because it used the perfect knowledge of computer-generated databases to ensure the accuracy of the analysis. This was an innovative and, at the time, controversial step. However, over the intervening years the use of endgame databases has become an accepted feature of endgame writing and today it would be a brave (and probably foolhardy) author who would write about 5-man endgames without checking his analysis against the database.
Endgame databases are having a gradually increasing effect on endgame theory. Over the past few years two main developments have served to empha-size this point. The first is the integration of the databases into normal chess-playing programs. Access to the databases of 1994 was based on simply looking up the positions in the database. The result was that even if you had a position in which a simple piece exchange would lead to a winning 5-man endgame, the interface to the database would simply report 'position not found'. However, manufacturers of chess-playing software saw that the end-game play of their programs could be enhanced by a relatively simple step. When constructing the analysis tree for a given position, if a node in the tree was a five-man endgame, the node would be evaluated by looking up the re-sult in the database rather than analysing any deeper. Thus huge swaths of analysis could be eliminated, replaced by a simple look-up. The effect of this technique on endgame analysis is dramatic. If you have a position which is not in the database (because it contains 6 or 7 men, for example), but in which a winning database position can be forced within the normal search depth of the playing engine, then the program will announce a forced mate. The 'normal search depth' can be quite large; 30 ply (15 white and 15 black moves) is quite feasible. The only problem with this technique is that it only gives an upper bound for the distance to mate and the true figure is often substantially less.
The second main development is the construction of a large number of pawnless 6-man databases. Although other researchers have made such databases, Ken Thompson took the further step of making his available on the Internet, using an ingenious interface which allows anyone with a web browser to access them. The construction of the databases is only the first step since the task of extracting useful information is formidable, but I have made a first attempt in the lengthy chapter dealing with 6-man endgames (see p.323). In this chapter there are a number of surprises which upset existing endgame theory. In all cases where existing theory has been shown to be incorrect by the new databases, the error has been to underestimate the superior side's win-ning chances. Two particular cases are D+L v 2T and T+L v L+S with op-posite-coloured bishops. Although the superior side is only two 'points' of material up, both these endings are generally winning. There are some very long wins in this chapter, with the 242-move win on p.336 being the record.
As to the future, doubtless further efforts will go into understanding the da-tabases which already exist, and into the construction of further six-man data-bases. The next major step will probably be the construction of six-man databases with pawns, which will undoubtedly revolutionize many areas of endgame theory. There are many exciting developments to look forward to, but I hope that readers will enjoy the current 'start of the art' as described in this book.
Finally, I would like to thank Graham Burgess, Guy Haworth and Helen Milligan for their help with proof-reading this book. Of course, I accept responsibility for any errors which have crept through.

John Nunn