Author: Bruce Moreland
Date: 12:02:47 10/03/99
Go up one level in this thread
On October 03, 1999 at 12:44:42, Steve Schooler wrote: >My understanding is that in game 5 of the last (6 game) Kasparov vs Deep Blue >match, at one point: Deep Blue was winning (or close to it); Deep Blue then >tactically blundered, which allowed Kasparov to achieve a perpetual; and >Kasparov overlooked it. > >Is my understanding accurate? If so, has anyone heard of an explanation for >this; I thought such a tactical blunder impossible? In absence of IBM >explanation, the only speculation I can conjure is: > >1. human intervention >2. flaw in search algorithm >3. horizon effect > >None of the above seem plausible, and no other explanation seems plausible. It isn't human intervention unless you are the type who thinks the CIA is poisoning your milk. I've looked at the end of that game a bit and I think that the computer simply messed up in computer fashion, meaning that it didn't see the perpetual because it was too deep. It's not like this is easy, extremely strong human players didn't see this during the game (including Kasparov), and a lot of analysis was invested in the game afterwards, which wouldn't have happened if the position was trivial. bruce
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.