Moderator: Ras
Is it possible he doesn't know how to remove doubles? Or doesn't like to remove them??The first chess database had about twenty thousand games. Now my own database consists of twelve million games. Nowadays it's impossible to thoroughly study the game without computers," he said.
Very funny!James Constance wrote:Anand quote from http://www.chessbase.com/newsdetail.asp?newsid=5109Is it possible he doesn't know how to remove doubles? Or doesn't like to remove them??The first chess database had about twenty thousand games. Now my own database consists of twelve million games. Nowadays it's impossible to thoroughly study the game without computers," he said.
Merry Xmas!
James
Marcel van Kervinck collected games (using an automated script) from rated players only on FICS at 30 minutes per game or slower time control. Last I heard (several years ago) he had 13 million games.Stephen Ham wrote:Very funny!James Constance wrote:Anand quote from http://www.chessbase.com/newsdetail.asp?newsid=5109Is it possible he doesn't know how to remove doubles? Or doesn't like to remove them??The first chess database had about twenty thousand games. Now my own database consists of twelve million games. Nowadays it's impossible to thoroughly study the game without computers," he said.
Merry Xmas!
James
OK, so Vishy has quantity, but does he have equivalent quality. I don't think that's possible.
Amassing all the "quality" OTB, correspondence, engine, and centaur chess games will surely produce a figure of no more than three million, and probably quite a lot less.
So to reach that 12MM number, he's collected the games of weaker players and games played under less than ideal tournament conditions. So, sorting the "wheat from the chaff" with those is a waste of time.
All the best,
Steve
I don't know that this is entirely true. Of course, it is quite hard to eliminate all games considered weak or the like. If you reduce a database to only master games then yes, there is no question you will have a LOT less than 3 million. Still, is it that simple? If you want to prepare in a detailed way against an opponent, it can be interesting to see the player's games when they were young and how and what they played. For an instructor, those thousands of national youth championship games of sub-12 players is a real boon, whereas to a player like Gelfand, it would be useless waste. Who is to say what is valuable or not?Stephen Ham wrote:OK, so Vishy has quantity, but does he have equivalent quality. I don't think that's possible.
Amassing all the "quality" OTB, correspondence, engine, and centaur chess games will surely produce a figure of no more than three million, and probably quite a lot less.
Hello Albert,Albert Silver wrote:I don't know that this is entirely true. Of course, it is quite hard to eliminate all games considered weak or the like. If you reduce a database to only master games then yes, there is no question you will have a LOT less than 3 million. Still, is it that simple? If you want to prepare in a detailed way against an opponent, it can be interesting to see the player's games when they were young and how and what they played. For an instructor, those thousands of national youth championship games of sub-12 players is a real boon, whereas to a player like Gelfand, it would be useless waste. Who is to say what is valuable or not?Stephen Ham wrote:OK, so Vishy has quantity, but does he have equivalent quality. I don't think that's possible.
Amassing all the "quality" OTB, correspondence, engine, and centaur chess games will surely produce a figure of no more than three million, and probably quite a lot less.
Even pure correspondence game databases such as UltraCorr2 by Tim Harding are already approaching the magic million number, where you will find many rare resources including many postal games by the likes of Alekhine or Keres, which you may not find elsewhere.
Albert