by Steve Lopez

The third version of the ChessBase Opening Encyclopedia has just been released. It's a substantially beefed-up version, with a database of over 1,100,000 games. In this issue of ETN, we'll have a look at this latest version of the Opening Encyclopedia (hereafter called CBECO).

What this disk isn't is a version of the Informator publishers' ECO, so let's get that oft-asked question out of the way right off the bat. The Informator ECO is composed almost completely of opening variations with no complete games included (except for the occasional footnoted miniature). The CBECO is primarily composed of complete games, while the real heart of the CD is the group of opening surveys which are included.

A little bit of history is in order here. Ten years ago, before Interrant access became the norm for computer users, it was difficult to find games and other chess information in a digital form. ChessBase made several product lines of floppy disks available, among them a set of opening floppies (originally called Opening Disks, later called Author Disks, and still later called Fritz Power Book Disks [no relation to the current PowerBook CD]). The typical disk in the series was written by a known authority on the opening in question; each author was a FIDE-titled player. There were three types of games on each floppy:

  1. Opening Surveys -- these were annotated game fragments, similar to what you see in any standard (print) opening encyclopedia. The fragment typically ended in the late opening or early middlegame and the final position received an Informant-style evaluation symbol. There were also numerous branch variations provided in each fragment; these also ended with an evaluation symbol. There wasn't a set number of surveys per disk; some authors preferred to provide a small handful of surveys, each of which had dozens of variations and subvariations, while other authors provided a large number of surveys containing a smaller number of variations (to make them easier to follow). The purpose of the surveys was to provide an overview of the variations of each opening, along with evaluations and analysis in symbolic form.
  2. Annotated games -- these were typically annotated using Informant-style symbols (for an international audience), although some authors used text in their annotations as well. These were complete games, mostly taken from master and grandmaster practice. As with the surveys, the number of these annotated games varied from disk to disk, but 100 to 250 was typical.
  3. Unannotated games -- these were provided for further study, once the reader had completed the surveys and annotated games. There were typically 2,000 or more unannotated games on each floppy disk, though the number could vary higher or lower depending on the opening and author.

Each floppy also contained an opening key (see ETN for March 11, 2001), which allowed you to go right to all the games of a specific variation just by clicking on that variation's listing in the key.

These floppies sold for about $29 each and there were dozens of them available. In 1997, ChessBase GmbH decided to combine all of these floppies into a single CD database, as well as add new material to cover all ECO codes that hadn't previously been covered in the series. The first edition of the CBECO had about a quarter-million games and sold for a fraction of the cost of the individual floppies (around $250 for the CD, compared to $2,000+ for all of the floppies). The CBECO was later updated in 1999 with the addition of 200,000 games and dozens of newer opening surveys.

The new ChessBase Opening Encyclopedia 2002 is still based on the original core of surveys and games from the 1997 edition, but much more material has been added. As stated previously, the database consists of over 1.1 million games. There are 3259 opening surveys on the CD (all 500 ECO codes are covered) and over 58,000 annotated games.

The one thing that dissatisfied me in the 1999 edition was the manner in which the new material had been added. In the first (1997) edition of CBECO, the opening surveys were the first games in the database (organized in order of ECO codes), followed by the annotated games, and finally the unannotated material. In the 1999 edition, the new material was tacked on to the end of the database -- the new surveys, plus annotated and unannotated games were all jumbled together at the end, making it difficult to scroll down the game list and pick out the newer survey material.

The 2002 edition of the CBECO has corrected this problem. All of the surveys appear in a "block" at the beginning of the database, once again organized by ECO codes. So if, for example, you want to scroll down to all of the Caro-Kann Advance surveys, you can easily scroll down to B12 and see all 56 surveys organized together in one place in the database. While it's a simple matter to do an ECO code search in ChessBase and in our playing programs, I sometimes just like to browse through the surveys -- the new organization makes this easy to do. And you can easily pick out the newer surveys by looking at the "Year" column.

The next "block" of games in the database consists of the annotated master and grandmaster games. Unlike the surveys, these are organized in chronological (rather than ECO) order. Following the annotated material comes the unannotated game block, again in chronological order. Organization isn't as critical with these games as it is with the surveys, since these games will likely be accessed by doing a search (as we'll see shortly).

So who has written these surveys and annotated the games? Taking a quick look at the annotator list we see (among many other names): Anand (248 games), Baburin (285 games), Blatny (1874 games), Kasparov (49 games), Khalifman (68 games), Kortchnoi (179 games), Yussupov (666 games), 51 games by the Polgar sisters. But the title of all-time king ChessBase annotator still belongs to Lubomir Ftacnik with a whopping 8267 games -- this guy has annotated more games than most people have played!

What level player would benefit most from the CBECO? Tough call, as I've known sharp 1200 players who have an easy time picking up concepts from games annotated with Informant symbols, while I've also known titled players who have a hard time in the absence of text annotations. Ultimately, it's an individual choice. The thing to keep in mind is that the CBECO is not a step-by-step, Reuben Fine-ish, text-based introduction to the openings. But if you're comfortable with standard print encyclopedias such as Modern Chess Openings, Nunn's Chess Openings, or ECO (which provide analysis using symbolic, rather than text, notation), then you'll have no problems with the CBECO. In fact, the advantage of CBECO is that it's an improvement over standard print opening references because of the wealth of complete games provided on the CD.

Let's see this in action. Let's say you've located an opening survey you're interested in studying (either by searching for a board position, doing an ECO code search, clicking on a variation in the opening key provided on the CD, or just browsing the list of opening surveys). As an example, we'll use a survey on the French Winawer written by Curt Hansen. He provides a lot of analysis in the form of additional variations and game references, but we'll skip those for right now. Going to the end of his suggested main line of the survey, we see this position:

which Hansen evaluates as being slightly better for Black. I'd like to investigate this position further. One thing I can do is have Fritz or another engine analyze the position. But I can also do a position search in the CBECO itself to see the games in which this position occurred. After a few minutes, I'm gratified to see that Jon Tisdall has annotated such a game between Hector and Nikolic (he uses lots of text annotations, too!).

So we can already see the main strategy you can adopt in using CBECO:

  1. Use the opening surveys to learn the basic variations and ideas of the opening;
  2. Play through related annotated games to get further opening information, as well as information and analysis of the middlegames and endgames;
  3. Further reinforce the ideas by playing through unannotated games to spot the same ideas and concepts in action.

The latter two steps put the CBECO head and shoulders above standard print opening encyclopedias. An encyclopedic opening reference in print form will typically give you just the opening moves and an evaluation, but after that you're on your own to figure out what kind of middlegame and endgame will occur. You'll need to figure out for yourself why that variation gets the evaluation shown for it. But with the CBECO, it's a snap to do a search for that evaluated position and get complete games which will illustrate the middlegames and endgames, as well as provide many answers as to why the line is evaluated as better for one side.

As is the case with many other ChessBase opening CDs, the CBECO comes with a statistical tree which you can use to browse through opening variations as well as use it as a huge opening book for Fritz and the other ChessBase playing programs (very handy for practicing the opening variations you encounter in CBECO). The database itself also comes with a complete variation key which can be used as a means of accessing games of a specific variation. However, please note that CBECO is not a standalone CD -- it requires ChessBase or one of the playing programs to access the database and opening tree.

Obviously, I'm a big fan of CBECO. It goes without saying that it's a valuable tool for over-the-board chessplayers. But as a correspondence player, I find CBECO absolutely indispensable for looking up opening variations as well as helping to chart my middlegame course in advance. When you take the plunge into the ocean of material in ChessBase Opening Encyclopedia 2002, I'm sure you'll also find it to be a vital addition to your chess arsenal.

Until next week, have fun!

You can e-mail me with your comments, suggestions, and analysis for Electronic T-Notes.