Author: Mike S.
Date: 17:25:01 04/12/02
Go up one level in this thread
On April 12, 2002 at 17:30:33, Otello Gnaramori wrote: >On April 12, 2002 at 17:05:45, Mark Young wrote: > >>Todays top micros must be pure Genius next to Deep Blue as Deep Blue needed >100X to 1000X the NPS that the top micros needed to play about as well. > >Good point, Mark. I also wonder why everybody seems to talk NPS. I talk ply depth. Somebody told me, Deep Fritz reaches 15 plies in a typical middlegame position after 3 minutes, on an Dual Athlon XP 1900+ (=1600 MHz). I'm not sure what we can expect based on that, for 8 CPUs, which will be Pentiums I think. Clock rate is yet unknown AFAIK. I don't know much of that SMP stuff... maybe 16 plies average, if it achieves 3 times the speed of the Athlon dual. I think, typical for Deep Blue was 11(6) = 17 plies. I had that impression looking at some of the logs too. Ply depth is not a perfect comparison either. I can't prove it, but my opinion is that 16 Fritz7 plies would be - most probably - competitive compared to 17 plies of Deep Blue, 1997. Keep in mind, Fritz has one of the most advanced search algorithms, including things developed in the last 5 years. Knowledge has also grown constantly during Fritz' version history. But we have to wait and see, which depth it really can achieve. If it will be more like between 15...16 average (not 16 most often), I wouldn't be convinced (yet), that it is D.B. level already. Except Kramnik looses, then you can forget all the figures :o) Since there is no direct match Fritz - D.B. possible, that comparison will always be a matter of opinions, more than a matter of facts. Regards, M.Scheidl
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.