ChessBase Reviews

| index | part 1 | part 2 | part 3 | part 4 |

ChessBase 7 Part 2

My first article on ChessBase 7 seems to have been well received and I have decided to write a further article, which may develop into a series depending on our readers' wishes.

This time round I intend to take a closer look at Databases and their usefulness to correspondence players. I hope you all read John Mackie's letter in Magazine 64. John is perfectly correct when he points out that many databases contain inferior games by inferior players, and that many of these games contain blunders, as can games played by Grandmasters. This is particularly true with an ever-increasing number of cheap databases appearing on the market.

This is not to say that games by weaker players may not be useful, but it does mean that the correspondence player will have to pay far greater attention when evaluating the strength of a particular move if games by ungraded players are referenced. I wouldn't consider myself to be even a moderate player, yet some of my games have appeared in the analysis of books by top players. Indeed even weak players can on occasion find moves and lines worth considering.

One way to cut out the weaker games is only to buy high quality databases, such as ChessBase's Mega Database '99, but these are costly and retail in the UK at around £185. This sounds a lot but Mega Database '99 contains 1.1 million games and approximately 30,000 of these games are annotated. Looking only at the annotated games, if we guess that the average book contains 500 annotated games then that is equivalent to 60 books. If the average book costs £14.99 then that would work out at £899.40 worth of books. Perhaps Mega Database isn't quite so expensive when viewed in those terms.

Anyway, onto the main point of the article and a basic look at the use of databases. Compared to a book, which will follow an annotated game with the occasional game thrown in, a database is capable of finding perhaps several hundred games from a given position. This is particularly true at the start of the game. The book annotator from the same position may give several ines of analysis but these will generally only concentrate on the candidate moves which the author considers relevant. The database simply identifies all the games where that position has been reached.

The following moves were played in a game in the SCCA Championship a few years ago:
1.d4 Nf6 2.c4 g6 3.Nc3 d5 4.Nf3 Bg7 5.e3 0-0 6.cxd5 Nxd5 7.Bc4 Nxc3 8.bxc3 c5 9.0-0 Qc7 10.Qe2

A fairly standard position has been reached but I will use the Opening Report in ChessBase 7 with Mega Database '99 as the Reference Database to see what sort of information I can gain after White's 5th, 6th, 7th, 8th, 9th and 10th moves. Looking only at the recommended main replies for Black and the statistical information produced the following was found: [bold indicates move played in the game]

5...0-0	Black scores 50% (1564 games).
5...c6	Black scores 41% (37 games).
5...e6	Black scores 55% (10 games).
5...c5	Black scores 41% (23 games).
5...Bf5	Black scores 67% (3 games).
5...Nc6	Black scores 33% (3 games).
5...h6	Black scores 67% (3 games).
5...Bg4	Black scores 47% (7 games).
6...Nxd5	Black scores 52% (264 games).
7...Nxc3	Black scores 50% (134 games).
7...Nb6	Black scores 50% (122 games).
7...c6	Black scores 50% (3 games).
8...c5	Black scores 50% (126 games).
8...b6	Black scores 70% (5 games).
8...Nc6	Black scores 20% (5 games).
9...Qxc7	Black scores 55% (18 games).
9...Nc6	Black scores 47% (41 games).
9...cxd4	Black scores 37% (4 games).
10...Bg4	Black scores 50% (18 games).
10...Nc6	Black scores 47% (18 games).
10...b6	Black scores 68% (14 games).
10...Nd7	Black scores 62% (5 games).

Basically ChessBase has produced a number of candidate moves from each given position. In addition to the above, it also gave far greater detail of the moves and plans for most of the candidate moves identified, as well as information of which particular strong players have played that line in the past, with instant access to those games.

Clearly the main thing to note is that this is statistical information based on the games found in the database. The information produced can therefore only be as good as the information contained in the database. It may well be that all the games found on a particular line have produced good results for one side or the other, but that does not necessarily mean that any suggested move is best. Recent developments may have found a new move that totally overturns previous theory and has the other side winning easily. It is therefore very important for the correspondence player to keep their database up to date by adding new games on a regular basis in order to keep abreast of current theory.

It is the responsibility of players to check all the suggested moves themselves, and indeed too seriously look at other possibilities. This said, the information produced is a good guide to help the correspondence player navigate the opening and emerge with a position which is hopefully to their advantage and at the very least gives them an even chance of doing well.

The advantage of ChessBase 7 is that information can be found quickly and this allows players to devote more time to actually analyse the positions in their games. Clearly it is also an advantage to be able to instantly look at any of the games that have been found pertaining to a particular line which you are investigating, and with ChessBase it's a simple case of clicking on the game and you can instantly play through it.

Books may only give a few moves in a particular line and you have to constantly reset the board after playing through sub lines. Often a book will also not mention a weaker move which has been played by your opponent, so you may be left wondering what is the best way to take advantage of that move. Having a large database gives you a better chance of finding games where that move was played, and therefore getting an insight into possible ways of maximising any advantage that may come your way.

I don't know if either player used ChessBase in the above game, but it is interesting to note that at each of the junctions looked at, Black played what ChessBase reckoned was the best move from previous practice. Indeed Black did go on to win the game in question. Possibly both players were following book lines but with books you can never be sure whether or not your opponent has a better book than you do. Databases give you all the information and let you evaluate the usefulness of the moves yourself rather than rely on limited theory from perhaps a single reference manual.

New database formats are now appearing with ChessBase producing training CD's such as Amador Rodriguez's "Modern Ways of Playing the Sicilian". This particular CD has a database of nearly 6000 games as well as a tutorial format where the author guides you through the various ways of handling the variations under investigation. Useful test positions also help you evaluate how well you have understood the various lessons on the CD. It is possible that this format may become a serious rival for books as more and more opening CDs become available.