Author: Robert Hyatt
Date: 20:04:02 01/15/04
Go up one level in this thread
On January 15, 2004 at 16:09:14, Bob Durrett wrote: > >In another thread there was some discussion of what happens in a chess engine >when the sudden death deadline approaches and the engine must move very fast. > >The impression I got from that discussion is that the engine simply does not >work as intended because chess programmers assume, perhaps subconsciously, that >there will be enough time for the intended computations. When there is not >enough time the programmer's intentions are thwarted. > >I wonder whether or not any engines have been designed to perform differently in >such situations where time has almost run out. This really isn't a problem in that context. IE I play game/1sec with Crafty all the time, testing and stress-testing things. I pit crafty vs crafty and finish 30 games every minute. However, such fast games have lots of tactical mistakes, and I would not play two different engines against each other at that time control and then use the results to predict how they might do at game/2 hours... > >Perhaps of more interest is the question: "How would an engine designed to play >extremely fast chess differ from normal engines?" > >An artificial special case of practical interest would be where the engine's >opponent were given several minutes per move but the engine given only a couple >of seconds per move and pondering off. > >Bob D.
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.