Author: Robert Hyatt
Date: 20:05:20 04/08/02
Go up one level in this thread
On April 08, 2002 at 22:32:42, Michael Vox wrote: >On April 08, 2002 at 12:58:25, Robert Hyatt wrote: > >>On April 07, 2002 at 16:18:27, Michael Vox wrote: >> >>> >>>I often hear programmers refer to 'anti-human' stuff programmed in the coding. >>>In layman's terms, what does this make the program do ? >> >> >>Several things. >> >>1. avoid blocked pawn structures so that the game stays active rather >>than becoming positional. >> >>2. avoid trades to keep things complicated. Rather than trade, retreat and >>do something else. The more pieces on the board, the easier for the human to >>mis-calculate.. >> >>3. play active rather than strategic openings. >> >>You get the general idea... > >Must be hard to program that many variables. > >What do you think about Kramnik's comment that the 8X CPU will be at 2800 ELO. > I think he doesn't know a thing about computer chess... And the pre-match hyperbole is out of control.. >What would you estimate a 2.5ghz X 8 Intel CPUs at... in ELO with Fritz 7b ?? I have no idea how well fritz's search works in parallel. If it is respectable, I would guess mid to upper 2500's... certainly not 2700+ as some would suggest.
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.