Author: Robert Hyatt
Date: 20:40:48 01/18/00
Go up one level in this thread
On January 18, 2000 at 18:03:10, Chris Carson wrote: >On January 18, 2000 at 17:00:41, Robert Hyatt wrote: > >>On January 18, 2000 at 16:43:39, Chris Carson wrote: >> >>>Source: http://www.research.ibm.com/deepblue/meet/html/d.3.1f.html >>>Source: http://www.sciam.com/explorations/042197chess/ >>> >>>Program Ver Date Rating NPS >>>Chip Test 1985 ?????? 50K >>>Deep Thought 0.01 May 1985 ?????? 700K >>>Deep Thought 0.02 Nov 1988 USCF 2551 720K >>>Deep Thought May 1989 ?????? 750K >>>Deep Thought II May 1991 USCF 2600+ 6.5M Kasp. says Fide 2450-2500 >>>Deep Blue Proto May 1993 ?????? 100M >>>Deep Blue Feb 1996 TPR 2650 100M >>>Deeper Blue May 1997 TPR 2840 200M >>> >>>Best Regards, >>>Chris Carson >> >> >>The "deep blue proto" is wrong. It used deep thought hardware. And hit a max >>of 6 M nodes per second, although Hsu reported that they hardly ever used this >>configuration in real games... DB itself didn't play any games at all until >>1996, as that was when the first batch of chips (new DB chips) became available > >Oops, Cut and paste error. :) > >Should add: > >Deep Blue Jr 1997 ?????? 10M > >Best Regards, >Chris Carson In that case 10M is probably wrong. DB Jr used 16 chess processors, which would give a peak of 32-36M. Aha... Hsu always factored that by 1/3, so 10M is in the right range...
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.