From: "World Chess Championship", INTERNET:newsletter@mark-weeks.com Date: 01/03/01, 11:38 Re: Chess History on the Web (2001 no.5) Site review - UPITT (VI; Karpov++) 'What? Not UPITT again!', I can hear you saying. No, this time UPITT is just our starting point. I'm going to devote the next few numbers of this newsletter to look at other Web sites which offer game collections that cover specific players. To compare these sites, I decided to tackle Anatoly Karpov's career. Karpov, born 23 May 1951, will celebrate his 50th birthday later this year. His best years may be behind him, but from 1975 to 1998 he was involved in 11 world championship title matches -- vs. Fischer, Korchnoi (x 2), Kasparov (x 5), Timman, Kamsky, and Anand. I hope that no one will deny that Karpov, the 12th world champion, was a worthy titleholder. He has always had his detractors. It is his unusual misfortune to have been born a half generation between two of the greatest and most charismatic chess geniuses of all time -- Fischer & Kasparov -- where comparisons with Karpov have not always been kind. It is also his unusual misfortune to have been involved in many controversial title matches -- the default win over Fischer, the marathon matches with Korchnoi and Kasparov, the FIDE facesaver with Timman, and the stacked match against Anand. It is his unusual fortune to have won twice a world champion title without defeating a reigning world champion -- against Fischer and against Timman. Many observers of the international chess scene tired of his legal wrangling with FIDE over his contractual right to remain world champion without playing a match, an issue recently settled through arbitration. If ever there was an ordinary world chess champion, it was not Karpov. What sites will we look at? I had help from 'gmgiffen', a member of the Yahoo Club for chess history, who identified nine sites with player collections... Chessaround http://members.aol.com/chessaround/chess/partiendown.html Chesscorner http://www.chesscorner.com/games/download/download.htm Fernschach http://www.fernschach-international.de/english/frame-download-engl.htm GMchess http://www.gmchess.com/digest/gamebase/ NBCI Chessdata http://members.nbci.com/chessdata/players.htm Ossimitz http://www.crosswinds.net/~ossimitz/player.htm Chesspawn http://www.chesspawn.de/weltmeister.htm ChessBase http://www.chessbase-online.com/playerdatabase.htm Homestead Observer http://www.homestead.com/Observer/titled.html ...I was pleased to find that each of these sites offers a Karpov collection, which is already an indication of his popularity. In this review, I'm going to compare the different Karpov collections with UPITT's. In the next review, I'll take a higher level look at all of the collections available at each site. --- To get started I followed the same procedure I used for Anand, Kasparov, Kramnik, and Staunton. The last effort was described in 'Chess History on the Web (2001 no.1)' where I built a playing record and game collection for Anand. I downloaded the Karpov file (KARP4-PG.ZIP) from the University of Pittsburgh (UPITT) archive at http://www.pitt.edu/~schach/. The DIZ file informs us that KARP4-PG covers 'GM Anatoli Karpov: 3079 Games'. The PGN file is dated 1998-07-14 and the last games in the file are from the Karpov - J.Polgar rapidplay (g/30) match, held June 1998 in Budapest. Polgar won +2-0=6, largely thanks to Karpov's problems with the clock. Since 3079 games is too many for the time I have available for this review, I first prepared the file by splitting it into two. The first part, which I'll discuss here, covered Karpov's career up to and including the first match (Moscow 1984-85; 48 games) with Kasparov; the second part covered the rest of his career. I chose KKI for two reasons. First, it represents a natural break in Karpov's career. For the years up to and including this match, he was the dominant player after Fischer's departure from active chess; after KKII, he became second fiddle to Kasparov. In my opinion, the crown was passed during the 16th game of KKII, although Karpov held the title of world champion until the end of that match. Second, my research technique is different for events before 1985. My collection of chess periodicals has many holes before 1985, and I rely on books for those events; after 1985, I can rely on magazines, which generally have more complete results and deeper background, especially for obscure events like simuls. The split yielded 1415 games for the period 1960 through 1984, and 1664 games for the rest of Karpov's career. After removing 262 duplicate game scores for 1960-84, I had a total of 1153 games in this collection. To create the index page on Karpov's tournament, match, and exhibition record, I consulted several books on Karpov:- - 'Karpov's Collected Games : 1961-1974' by David Levy. Covers Karpov's career through the 1974 match with Korchnoi -- all known game scores, many annotated, some annotated by Karpov. - 'How Karpov Wins' by Edmar Mednis. Covers Karpov's career from the 1971 Alekhine Memorial through the 1974 match with Korchnoi -- 93 annotated games, all won by Karpov. - 'Chess Is My Life' by Karpov & Aleksandr Roshal, translated by Kenneth P. Neat. Covers Karpov's career before the 1978 match with Korchnoi -- a mixture of biography, unannotated games, and games annotated by Karpov. While reading this book, I was struck by a recurring theme in comments on Karpov's style. * Roshal says (p.14), 'When observing Karpov's play or playing against him, one cannot help thinking that all his pieces are linked by invisible threads. This net moves forward unhurriedly, gradually covering the enemy squares, but, amazingly, not relinquishing its own'. * Alexander Bikhovsky, the permanent trainer of the USSR youth teams (p.70), 'After losing to [Karpov], his opponents [in the Student Olympiads] appeared somehow confused, it seemed that they could not quite understand the young grandmaster's ideas and moves'. * Ljubomir Ljubojevic (p.201), 'Karpov conducted this game [vs. Portisch, Portoroz 1975] splendidly. At first I found some of his moves not altogether understandable, and only after careful analysis did I discover their hidden strength'. * Mikhail Tal (p.295), 'Many of Karpov's intentions become understandable to his opponents only when salvation is no longer possible'. - 'Schach-WM 81 Karpow - Kortschnoi' by Helmut Pfleger & Otto Borik. Has a chapter on Karpov's career before the 1981 match with Korchnoi -- 44 page biography with many crosstables. - 'Chess at the Top' by Karpov, translated by Kenneth P. Neat. Covers Karpov's career from Montreal 1979 through Bath 1983 -- 49 games annotated by Karpov, tournament results before the 1984 match with Kasparov. Of these books, Amazon.com has only a paperback edition of 'How Karpov Wins' in stock. It can be found at Amazon at address... http://www.amazon.com/exec/obidos/ASIN/0486278816/worldchesscham.7 ...along with four short reviews. After researching the early years of Karpov's career and creating an overview, I uploaded the index page and the PGN file to... http://Mark_Weeks.tripod.com/chw01c01/karpov.htm ...from which the collection of PGN games can be downloaded. --- The 1153 games had the usual problems that I've seen with other collections from UPITT:- - Events missing a few games; - Games impossible to associate with any event; - Games from events not mentioned in any of my references; - Events with partial or missing round numbers. To resolve these problems, I downloaded the Karpov collection from each of the nine sites listed above. Here's a summary of my first impressions:- (A) Chessaround - German language site with collections in ChessBase format. Long after the page text has finished, something continues to load; it's a 200K background image! (C) Chesscorner; (F) Fernschach ; (G) GMchess; (N) NBCI Chessdata - All four sites have collections in PGN format; no problem to download. (O) Ossimitz - 'Annofritzed' collections in ChessBase format. The Karpov file is very big (1946K) and the download stopped after about 60% had been done, so I gave up. A day later I tried again; it stopped after 5%, after 20%, and after 50%, so I gave up again. Three days later I managed to download the file successfully. (P) Chesspawn - German language site with collections in ChessBase and PGN format. I downloaded the Karpov file in ChessBase CBV format, but when I tried to open it I received a 'CRC error' message. I downloaded it again, got the same message, and then downloaded the PGN format. I was unable to retrieve the Karpov collections from ChessBase and from Homestead Observer. For ChessBase, I used the 'Search the Player Database' on Karpov, got his page, and clicked 'Load games played with white'. This gave me a Java game viewer page titled 'Default MFC Web Server Extension' with 'Applet JChessApplet running'. Then my browser locked. The first time I had to close it with CTRL-ALT-DEL & the second time it locked my computer completely, so I gave up. I suppose the problem is due to the large amount of data involved, but I saw no option to reduce this. I tried ChessBase once again while finishing this article, but after clicking on 'Load games played with white', received a page with only the cryptic '-2147467259 (0x80004005)' and nothing else, which was obviously some kind of an error message. I went back to the home page at www.chessbase.com and discovered that the 'Player Database' link had a new address... http://www.chesslive.de/playerdatabase.htm ...Clicking on this link gave me a 'server not responding' message. It looked like something was changing at ChessBase. I hope the change is ready soon, because the Chessbase collection could easily be one of the best on the Web. When I clicked to download the Karpov file from Homestead Observer, I got a 'server not responding' message. A day later, I had the same problem. I tried on two other occasions, always with the same problem. --- In the following discussions, rather than the full name, I'm going to refer to the one letter code that I've assigned above to each site. When you see '(A)', this means 'Chessaround'; '(U)' means UPITT. I converted the ChessBase files to PGN and built a database from the PGN headers for each of the collections. The following table shows the number of games in each collection, the year in which the earliest game was played, and the year in which the last games were played:- Count First Last (A) 2694 1960 2000 (C) 1051 1961 1994 (F) 3381 1961 1999 (G) 2365 1966 1999 (N) 1588 1963 1995 (O) 3193 1960 2000 (P) 3381 1961 1999 (U) 3079 1960 1998 Closer inspection showed that collections (F) and (P) are identical, so I deleted (P) from the database. Now I needed to compare the seven remaining collections. It's a big job to compare two digital game collections move for move, so I looked at just the game headers for the year 1971. In November of that year, Karpov shared 1st & 2nd place in the Alekhine Memorial with Leonid Stein, in front of World Champion Spassky; in front of ex-World Champions Smyslov, Petrosian, & Tal; and in front of Bronstein & Korchnoi. A month later Karpov shared 1st & 2nd place with Korchnoi at Hastings. A new star had arrived on the chess scene. The following table lists the results that I derived for 1971 from the UPITT collection. The first column gives Karpov's score, the second shows the number of games on my file versus the number played by Karpov in the event, and the third identifies the event. +2-2=2 6/6 Training match vs. Korchnoi, Leningrad +9-0=8 15/17 Semifinal USSR Chp., Daugavpils +7-0=1 4/8 Student Olympiad, Mayaguez, Puerto Rico +4-0=1 5/5 Leningrad University Team Chp., Leningrad +2-1=4 7/7 Armed Forces Team Chp., Leningrad +6-0=1 6/7 USSR Team Chp., Rostov on Don +7-2=12 21/21 39th USSR Chp., Leningrad +5-0=12 17/17 Alekhine Memorial, Moscow +8-1=6 15/15 Hastings If we add up the second column we find that there are 96 games on my file out of 103 played by Karpov in the nine events. For each of the six games missing from the Daugavpils & Mayaguez events, Levy's book says, 'The score of this game is not available.' For Rostov on Don, Levy gives six game scores with Karpov scoring +5-0=1. The appendix to Levy's book gives +10-0=1, while Karpov & Roshal's book gives +6-0=1. I chose to use Karpov's book. The following table shows the number of games for 1971 in each collection, the number of games on file for Moscow (the Alekhine Memorial), and the number of games for Hastings. (A) 90 17 15 (C) 96 16 15 (F) 116 21 18 (G) 53 17 15 (N) 54 17 15 (O) 101 20 17 (U) 113 24 17 ('Moscow' includes 5 games from the USSR Chp.) Here we see a direct correlation between the size of a file and the number of duplicate games it contains. This is what I expected to find and I would have been surprised to see otherwise. For chess databases, quantity is no indication of quality. The (G) and (N) collections are smaller than the others because they have games for only three 1971 events -- Moscow, Hastings, and the USSR Championship. --- Where my own collection was missing round numbers for games, I added them by copying from whichever collection offered them. While working with the data and game scores, I was most impressed by the (N) collection. The games had round numbers where UPITT was missing them & the games were in the correct order on the file. This is a sign that the file has been prepared carefully and not simply copied from another source. Collection (A) often had good round numbers also. To complete partial events in my own collection, I added six games from these other sources. I then uploaded the changed PGN file, now having 1159 games, to my own site. I'll add the rest of Karpov's career in the near future. Bye for now, Mark Weeks