Author: Robert Hyatt
Date: 10:07:14 01/12/00
Go up one level in this thread
On January 12, 2000 at 10:01:59, Chris Carson wrote: >On January 12, 2000 at 09:29:15, Robert Hyatt wrote: > >>Fide and USCF _BOTH_ would include this game. How do I know? Because it >>has happened to me in the past. If the computer goes down during the game, >>and the flag falls, the computer loses. No ifs, ands or buts. It has _always_ >>been like that... >> >Bob, > >Do you consider the 1995 Fritz 3 win on a Pentium 90 >over Deep Blue valid? Fritz 3 was the winner of the last >computer event that Deep Blue was in. Deep Blue finished >third. > >Source: http://www.gambitsoft.com/turnier/altwme.htm > >I do. IMHO. > >Strong performance of old software on slow machine IMHO. :) > >Best Regards, >Chris Carson No, because it didn't happen. In 1995 Fritz beat Deep Thought. Deep Blue didn't exist until 1996. The 1995 program was called "deep blue prototype" a few times... with the official description: "Deep Blue software, but running on the old deep thought hardware." But other than that, yes Fritz beat Deep Thought. Hsu/Campbell reported that in their lab, they had played a lot of games vs Fritz and were winning about 9 of every 10. But there is always that "1" game. And it could happen at the front of the 10-game series, or at the end. In Hong Kong, it happened at the front. At Hong Kong, DT was doing about 2M nodes per second, for reference. I can do that on an 8-way xeon. Or 4x that on a 16-way alpha. That was not seriously fast back then, being maybe 20x faster than the fastest micro programs. DB was way faster, 100x in fact. That is a huge difference.
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.