Author: Robert Hyatt
Date: 21:08:11 01/21/00
Go up one level in this thread
On January 21, 2000 at 23:27:52, John Hartmann wrote: >Just a brief thought on this topic. > >Would it really be safe to say that DB/DB2 would be that much more >highly rated -- a distinct concept from _stronger_-- given that Deep >Thought was something of an anomaly at the time, and that humans have >learned how to play against silicon in such a fashion as to offset >the gain in rating one would expect from the gain in strength? > >John Hartmann In 1985, I think so. But this was 1995, so it is at least likely that the GM players were aware of computers... But it is obvious that as they played more against it they would do better, IMHO, just as we see today. But then, don't forget, that this was deep thought. 1/100th the speed of DB, with < 1/10th of the chess evaluation... I think the speed and new stuff in DB would _more_ than offset this expected drop... And during the chase for the Fredkin, they played over 40 games vs GM players in various US tournaments... so it wasn't nearly as 'secret' as what happened in the Kasparov match... > >On January 21, 2000 at 15:34:30, Robert Hyatt wrote: > >>Here is the relevent quote from Hsu's book: >> >> After the Software Toolworks tournament... >> >> "The race for the Fredkin Intermediate Prize was officiall over. Deep >> Thought performad at 2776 for this (USCF scale) event, slightly behind >> the Hall of Fame result which was 2790 (hall of fame was another event it >> played in). By 1998 DT had played 42 rated games, including all the >> games played with serious bugs (US Open) was 2598. The best 25-game >> performance over the period was 2655 or 155 points higher than the >> requirements for the Fredkin Intermediate Prize" >> >>The opponents are listed in Monty's book... as are many of the game >>scores... >> >>Remember that DT's official USCF rating was 2551, which is lower than the >>performance rating for the same games, since the formula changes after >>the first 24 games are played. However multiple 2700+ performance ratings >>are remarkable. Particularly considering how much stronger DB/DB2 were than >>deep thought...
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.