Author: Robert Hyatt
Date: 06:46:43 12/28/00
Go up one level in this thread
On December 28, 2000 at 00:24:40, Uri Blass wrote: >On December 27, 2000 at 22:38:23, Robert Hyatt wrote: > >>On December 27, 2000 at 04:58:02, Ed Schröder wrote: >> >>>On December 27, 2000 at 00:28:01, Robert Hyatt wrote: >>> >>>>On December 25, 2000 at 19:17:51, Christophe Theron wrote: >>>> >>>>>On December 25, 2000 at 02:31:32, jonas cohonas wrote: >>>>> >>>>>>Todays chessplaying progs, are really, really strong!! >>>>>>And maby Kramnnik is right when he says that progs are >>>>>>inferior in tactics and positional "understanding" how- >>>>>>ever it seems that even the super GM's (top 10) are see- >>>>>>king ways to play for draws and blocked pawnstructures >>>>>>they are ALL avoiding (maby except from Morozevich) att- >>>>>>acking, these "inferior" progs because of the simple fact >>>>>>that when it comes to calculation of: lines, exchanges, >>>>>>combinations, mate in 69# (lol), tempo's etc........... >>>>>>the prog's (top 10) are superior. And as the prog's get >>>>>>more and more knowledge and implementation of selflearning >>>>>>and arteficial intelligence AI. they WILL be superior to >>>>>>humans in all aspects of the game, that is just a question >>>>>>of time, just look at Deep junior and Deep Fritz/Fritz 6a-b's >>>>>>latest score against GM's. The question is not; if progs >>>>>>will beat human brains, but when they will reign supreme >>>>>>it might be in 5 years, my personal feeling is 5 years, but >>>>>>it might aswell be 50 years (i dont think it will be more >>>>>>than that though) many people might feel that the fun and >>>>>>intrique of the game will be taken out of chess, if the >>>>>>progs get unbeatable, but in that lies the real challenge >>>>>>for humans, to stretch there abillities and raise the level >>>>>>of their own game, try to come up with new ways to beat the >>>>>>progs, and thereby becoming better players, because no matter >>>>>>how good the progs get, there will always be the excitement >>>>>>of human v human games and matches. >>>>>>Personally i enjoy watching progs v humans and vice versa, >>>>>>I also enjoy a good celluloid battle in contrast to the >>>>>>sillicon binary shootouts between the top dog's of AI. >>>>>>By the way Deep blue lost to shredder 4 who played from a >>>>>>laptop with a pentiumll 450 mhz 128 ram so in conclusion >>>>>>you can have as many kn/s, if there is no sufficiant knowledge >>>>>>it is only calculations. Deep blue would probably rank as 9-10 >>>>>>on any rating list with todays PC progs(my opinion soley) >>>>> >>>>> >>>>>Deep Blue would be #1. >>>>> >>>>>Deep Blue Junior would probably not be above the best PC programs. >>>>> >>>>> >>>>> >>>>> Christophe >>>> >>>> >>>>Murray had DB junior online during a couple of the Kasparov/Kramnik games. >>>>With 20 chess processors, it was hitting over 40M nodes per second. I wouldn't >>>>want any part of that with today's PC hardware/programs... >>> >>>It would be nice to see the thing participate only once. The thing seem >>>to be alive so what is keeping them? >>> >>>Ed >> >> >>IBM is the problem. They have everything to lose, nothing to gain, >>unfortunately. > >I do not see what they can lose if it can get the best rating on ICC today. >They do not have to play in tournament but if they know that they are clearly >better than other programs they have nothing to lose because they can prove it >by getting the best rating on ICC today. They had the best-ever win/lose ratio on ICC years ago. There is nothing more to prove there. But if they lose _one_ game to a commercial chess engine, that is all that will be heard for the next few years. "This is the program that beat IBM's mighty deep blue." I can see why they would want to avoid that at all cost. Because if they played that program 100 games, and won 99, that one loss would still make the 'box front'. > > They are basking in the afterglow of DB2 vs Kasparov. >>They aren't going to do _anything_ to dampen that... > >Deep blue Junior is not DB2. > >Uri It actually is. Maybe 1/10th the speed or whatever, but still the _same_ engine / hardware combination. That is enough to make the risk outweigh the potential gain..
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.