|
|||||||||
The ordinary player would still like to know how good these programs are from his point of view and whether they suit his specific individual needs, for the world of a chess programmer is at times far removed from that of a chess player. Sometimes the programmer creates a highly sophisticated system which can be seen and admired only from a distance! But its utility for the average player still remains a question. Programmers do offer elaborate tutorials to show how the system works. But the player some times wonders, “If I have to spend so much time mastering the program, when am I going to study chess?” This is a legitimate question. My own view is that a player should at least grasp the essentials of a program and it is not hard to learn. So where does one begin? In the present case a simple way of doing it is to get the demo version, Chess Assistant 7 Light from the program site itself. At least it would give us a “feel” of the system. Thereafter one can look at what experts have to say about the software. Currently there are at least two experts on the subject, Dr. Robert Pawlak and Dadi Jonsson. Pawlak’s articles may be found on his own website, Chess Reviews.com (http://hometown.aol.com/rjpawlak/ch_widow.htm) and also Chess Assistance.com. Check the FAQ column on both sites and they will offer quite a few insights into the business of program reviewing. The second expert in the field is Dadi Jonsson who writes the Chess OK column for Chesscafe website. The previous columns may be located in their archives. There must be readers impatient with this procedure and they may wish to learn everything through hands-on experience. That was the way I learned, and in this review I shall offer an opinion based my experience. My first acquaintance of a CA software was with its version 7.1, five years ago. This was a relatively simple system and I still retain fondness for the program. It had two databases, Hugebase with 2 million games and Guru database with 50,00 games. It included three analysis engines, Tiger 15, Tiger Gambit III and Ruffian, a very hot winboard/UCI engine in those days.
I recall my excitement when I discovered games of Rashid Nezhmetdinov, a player whose phenomenal combinational ability was admired by all his contemporaries, especially Tal. When he was asked for the happiest moment in his life, Tal replied, “ I rarely felt so happy as when I lost to Nezhmetdinov in our championship. What a beautiful game!” (For more on Nezhmetdinov, see the Wikipedia article about him, or visit http://nezhmetdinov.homestead.com/Nez2.html.)
The companion CD was Chess Openings 2003 and it covered the entire range of openings with assessment at the end of each line. It had its own database of games. If you looked up a position from an opening on the right side of the board you could see important lines of theory and at the bottom of the board a collection of games from the opening. Thus theory and practice were linked. Frankly speaking, I felt these programs were good enough and did not see the necessity for upgrade. It was only when I received the present product that I noticed the vast difference. CA 9.1 is a leap from CA 7.1 both in terms of size and strength. But first let me explain how to start. Along with the CD there comes a short manual with detailed explanation accompanied by screenshots and diagrams. This is good enough. But seeing is believing. The Help toolbar on the screen offers all the necessary information, beginning with first steps. The other toolbar, Tools, gives you the options, be it fonts and colours or the choice of engines. Once you open a database you will have to decide whether you want to see the games of a particular player or a particular position. These are OK. But in my view every CA program should also include the standard feature Troubleshooting with FAQ along with tips & tricks. This would take care of a host of problems like access violation, occasional crashes, and all kinds of error messages. Now let us consider the first steps. If you open the Search Bar you can check the Header that lists all players on the database in alphabetical order. It is easy to narrow your search and shortlist games between the same opponents. When you do, you learn a lot about the changing fortunes of both the players. A case in point is the catastrophic loss of Kasparov with the King’s Indian in the hands of Kramnik in Novgorod 1997.* I wanted to see if Kasparov played the King’s Indian in a serious tournament game. He never did. He had lost faith in his favourite defence. This proved to be fatal in the World championship Match with Kramnik three years later. But it is search by position that is of great value in understanding the game.
One can follow the rise and fall of a sharp variation in Sicilian or Semi-Slav through games starting from the same position and prepare accordingly. However, a typical CA database suffers from two kinds of drawbacks. It is not always accurate and comprehensive as expected of a good database. In recent years chess historians like Tony Gillam and Dr. Fiala (http://www.moravian-chess.cz/) have unearthed a number of games from various sources and published them. These games have yet to find their way into CA databases. The editors of CA programs should make it a point to consult the work of historians so that the databases become up-to-date. The present trend of adding only recent games to the database while allowing the older collection to stagnate is by no means desirable. Secondly, misspelt names are a problem. I could give quite a few examples. Here one
or two would suffice. It is Nezhmetdinov, Rashid and not Nezhmetdinov, Rashit. Similarly three names are listed as Anand: Anand/Fritz, Anand V2 and Anand,
Viswanathan. The first name is a consequence of Anand playing the match
against computers in Mainz 2002 event. The second again lists the two
computer games at Mainz and also adds for good measure the game Gunasekharan-Anand played in Chennai Open 2004(1-0 !) Now the second Anand
is quite different from his more famous namesake Viswanathan Anand. A major
consequence of these misspelt names is that there would be duplicates in the
database, needlessly inflating the number of games. Then the same error is
carried forward to each new version of the database. Hopefully, these
problems are attended to in the near future. This brings me to the contents of the DVDs under review. To begin with, they complement each other. CA 9.1 is the basic program and it is called the Starter Package. It has 3 databases. The Guru database has games of masters from 1807 to 2007. It has 100,000 games. Of these, about 33,209 games are annotated. The Elite database includes only games of prominent grandmasters played between 1995 and 1997. It has 5031 games. Of these, about 503 games are annotated. The last database is Comprehensive Openings 2007. This has far more advanced features than Chess Openings 2003 that I mentioned before. It also has 36,000 annotated games. Now the content of “annotations” in all these databases is variable. There may be no more than one or two lines of analysis in several games. All of them qualify as “commented’’ games only because the program sees any additional content by way of text, diagram and evaluation as comment or annotation. The second DVD Chess Databases 2007 offers three collections of games. The HugeBase includes games of both known and unknown players from the past centuries to the present, right till the 1st of April, 2007. There are 3,155,800 games of which 1,386,387 games are annotated. However, there is a more precise way of determining the range of annotations for each database. In the search function for commented games one could specify the number of variations (preferably 5) and it would give a fairer idea of annotated games. When I made A random search on the guru database (1,00000) it yielded a haul of more than 28,000 games with adequate notes. The Guru Database confines itself to games of masters and grandmasters from 1807 to 2007. There are 150,000 games. Of these, 36,028 games are annotated. The last database, Corr.2006, includes 376,450 correspondence games (1973-2005) of which 15,800games are “annotated.” But the degree of annotation is often minimal, limiting itself to a diagram or a single sign or word time at the end of the game. To say the least, it is of little use. In my opinion the best database of correspondence chess is UltraCorr** edited by Tim Harding, published by ChessMail. It has a wealth of deeply annotated games. I do not think Corr.2006 database could come anywhere near the rigorous standards of Ultracorr. In my view Convekta should confine itself to databases of over-the-board games. There is enough room for improvement in that area. To revert to our discussion of this DVD…each database is equipped with five (!) analytical engines, Shredder 9.11, Crafty, Dragon Delphi and Ruffian. In recent years Shredder 9.11 and Ruffian have built up a reputation. If you can’t afford Rybka 3, these would do with the present databases. To sum up, these two DVDs together form the Professional Package in CA parlance. Dear reader, now that you have a fair idea, feel free to check the features of CA 10, the latest version of the program, from the Convekta website, and if it suits your present needs, go ahead and try it yourself. Recommended.
*This game will be annotated in a subsequent review of a Kramnik CD.-Ed. **An updated version UltraCorr 2 would soon be released by Chess Mail.
|
The
Advertise to Single insert:
|
||||||||
|
|||||||||
Convekta Ltd., the well-known Russian
chess company, has just announced that it is going to release CA 10, a brand
new version of its program Chess Assistant, along with their engine Rybka 3 soon. In recent years Convekta products have carved a niche of their own in
the world of chess software. The flagship Chess Assistant in particular has
come up with new features and improved performance with each version.
The present-day
player takes chess software for granted. But for people like me brought up
on books and magazines it was an incredible experience to hold the entire
treasury of chess games in human history in the palm of one’s hand. What was
more, one could trace any memorable game mentioned in chess lore and enjoy
it. If a favourite player’s games had to be found, there was no problem at
all. In a fraction of a minute the whole set of games would be available on
the screen.
The analytical engines would always respond to my queries on
any position, with a ready variation, be it the middle game or the endgame. In the main they shed light on any possibility that did not occur over the
board. At times they would be exasperating too, giving differing lines on
the same position. But one chose the engine one wanted from the list.
