Author: Paul Richards
Date: 12:21:34 06/11/99
Go up one level in this thread
On June 11, 1999 at 13:56:31, Dann Corbit wrote: >On June 11, 1999 at 13:50:28, Dann Corbit wrote: >>I can provide the epd in sequence if anyone wants it. >Find it here: >ftp://38.168.214.175/pub/EPD/famous-inorder.epd OK, I have reviewed the first position in context in the game "Anderssen - Dufresne (the evergreen)", and 19. Rad1 can be effectively refuted by 19...Bd4!. Instead black played 19...Qxf3 which immediately gives white a winning game. Then after 20. Rxe7+ black plays 20...Nxe7? which leads to mate in 4. as in the actual game. Instead 20...Kd8 holds on longer, though white still has a winning game. For curiousity's sake I will look at the other positions, but I have seen this sort of thing over and over in analyzing famous games. For example a particular Tal game will have sacrifice after sacrifice, ending in a beautiful mate. From the standpoint of a human player, psychologically the opponent plays into the game by accepting the sacrificed material. However a computer has no such compunctions, and this same brilliant game would have been crushed by even a weak program. The game is still *aesthetically* pleasing, but simply doesn't hold up under silicon-assisted scrutiny. This is why GMs tend to avoid playing this style of game against a machine. They can still beat the machine by playing against its weaknesses, but the style of game we humans really like to see is next to impossible to pull off against a machine. Certainly modern computer chess can often find holes in such games.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.