Author: Shaun Graham
Date: 09:05:57 09/16/98
Go up one level in this thread
On September 16, 1998 at 06:41:00, blass uri wrote: > >On September 16, 1998 at 02:07:36, Shaun Graham wrote: > >>On September 16, 1998 at 00:49:49, blass uri wrote: >> >>> >>>On September 15, 1998 at 17:54:09, Shaun Graham wrote: >>> >>> >>>>5.Also i might point out humans make bad moves quite frequently even at the GM >>>>level, so if i saw a positional blunder, i wouldn't be thinking my opponent is >>>>playing like a computer, but rather "Yes!! I have a win!!" >>> >>>There are some positional blunders computers do that no GM is going to do >>>The probability for these blunders is small but sometimes computers do not >>>do right evaluation of a pawn ending or do not see an idea of a simple fortress >>>position. >> Well first of all you must realize that occasionally huge blunders do happpen but that in actual play those situations or relatively rare. Further as pointed out the only opportunity a GM would have to see it is either in one game(a tournament) or 6 games(a match) so the likelyhood of such a thing is even more greatly limited, and you would have no reason to ever think that it was anything more than an anomaly if such a move occured anyway. >>Well first of all you are arguing with a GM, second of all, may positional >>mistakes are made because a player thinks that a move is setting up a tactical >>manouevre that happens to turn out faultily. Further GM's do blunder and >>occasionaly blunder huge! in the last 5 years Nigel Short has hung his queen in >>a 40/2 game and also he has missed a mate in 1! > >The question is if he was in time trouble when he did this mistake. The answer is no, he wasn't in time trouble. > >>So to say that every GM would >>always avoid such and such a move isn't very reasonable IMHO. > >I can only say that I saw some mistakes of computers that I expect myself not to >do if I am not in time trouble and I am far from being a GM (I have 2020 fide >elo). > > Further as >>Morovic pointed out since if it was a tournament (meaning probably one game) you >>wouldn't think it was anything but an anomaly if some strange move did occur. >>If it was a match such strange moves certainly wouldn't happen in all of the >>games, again leaving little room for such an asumption. >>>> >>>> >>>>So ultimately the likelyhood that i would think another opponent especially a GM >>>>was using a computer against me in a tournament or a match is almost zero. >>> >>>There is another way to suspect that the opponent is a computer. >>>You can see that the opponent thinks about obvious moves. >>> >>Most computers don't take an overly amount of timeon obvious moves, and further, >>GM's often think on obvious moves because they frequently go ahead and calculate >>what they think their next move will be after the obvous move. > >It does not make sense to do it because you can do it in the next move when you >can see the position more clearly. > Well Uri what you need to realize is it doesn't make sense to YOU but in ACTUAL PRACTICE players take time on all sorts of moves that YOU might think are obvious for lots of different reasons. Heck i'm sure you have even heard of players taking 15 minutes on more to make the first move! Further for most grandmasters looking 1 move ahead mentally after an obvious move really isn't going to make the picture in their mind all that much clearer. Shirov has demonstrated that he can see some combinations that are 25 moves long without moving the pieces and rather quickly as well so 1 move isn't going to effect his vision very much at all.
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.