Author: Uri Blass
Date: 23:48:41 12/04/01
Go up one level in this thread
On December 04, 2001 at 12:38:36, Miguel A. Ballicora wrote: >On December 04, 2001 at 10:34:15, Uri Blass wrote: > >>On December 04, 2001 at 09:46:22, Christopher R. Dorr wrote: >> >>>Again, this brings us to the issue of 'true' computer strength. I Have the >>>next-to-latest versions of Tiger, Rebel, and Fritz. All of these could likely >>>'hang' with a 2550-2600 GM. But is it fair to say that is their 'true' strength? >>>I'm only a USCF 2150 or so, but I have been able to draw all of these programs >>>*much* more often that I 'should' be able to against a 2600 player. >> >>How many games did you play against 2600 players in order to know? >> >>> >>>I've gotten very proficient at playing for a draw against programs. I have blitz >>>draws against Crafty online. When you play for a draw at the outset, these >>>programs do *not* win nearly as often as would an equivalent human GM. They >>>often fall for things like pawn sacs to get to an opposite-color bishop ending, >>>or pawn sacs to reach dead drawn R+P endings, or even pawn sacs to get to drawn >>>K+P endings (he has 2 connected passers, but I have one protected one that he >>>can't move away from). >>> >>>In a thread on this earlier, a couple other expert/NM level players expressed >>>the same thing. I have a very good friend who is a USCF 2100 who was able to >>>draw Fritz 5 almost at will. >> >>Fritz5 is not the best software and he had a lot of time to learn it >>There is Fritz7 and it may be more interesting to know his results against >>Fritz7. >> >> Over several dozen games with muliple 'GM' >>>programs, I was able to get enough draws to extrapolate a USCF 2400-2450 >>>performance rating at G/15 and G/30 on a Celeron 800. >>> >>>So which is it? Are these programs the 2600s that they look like when they plan >>>Anand, or are they the USCF 2450s that they look like when they play me? One's a >>>strong GM, the other a weak IM. >>> >>>I am certainly *not* saying these programs are weak. But I am saying that they >>>do not conform to the typical performance pattern of a human, hence saying >>>*either* that they are strong GMs or weak IMs is misleading. Eventually, this >>>artifact will disappear as they become capable of playing GMs at the 3000 level, >>>but for now, it really isn't fair to say that Program X is a strong GM any more >>>than it is to say the same program is only a weak IM. >> >>It is not fair to say that they are weak IM's. >> >>Humans can choose the tournament that they play when they get their rating. >> >>Choosing for programs tournaments when their performance is bad is not fair. >>I see no problem with choosing tournaments when they can perform better. >> >>I am also interested to know the performance of humans who get the draws against >>programs when they play against GM's. >> >>I suspect that the same style is going to help them to earn rating points also >>against GM's. > >No way. GMs can make a position that is quiet to be a double edged one on >purpose, either positionally or tactically. Even at the risk of being slightly >inferior. Computers do not do that. If they go to slightly inferior positions then I expect their results to slightly suffer from it relative to the case when someone does not play in a style that forces them to go to inferior positions. >GMs try EVERYTHING to test that you know how to draw. They play with the clock, >they change the tempo of the game etc. etc. Besides, going to an endgame >praying to resist is the worst strategy you can do against a GM. I almost have no experience against GM's but the only games that I could avoid losing against masters with Fide rating above 2300 were games when I went into an endgame(there was one game when I beated a master who did not want a draw but he had rating of less than 2300 at that time). In one game I had slightly inferior endgame but the advantage of my opponent was not enough to win and in another game I did not play the opening well and lost a pawn but my opponent blundered and I got the pawn back and drew the game(an analysis proved that I even missed a win in that game) The first game was a game when I decided to go to the inferior position in order to draw inspite of the fact that I believed that there is objectively a better option(I prefered less pieces in the board) I remember that analysis of the game later with Genius3 found no tactical mistakes. I was a pawn down in the endgame when both sides had knights but both sides had passed pawns and the position was similiar to comp-comp game that I read between mepistho and another program when inspite of the pawn advantage it was a draw. The second game was not a game that I played well and my opponent simply did not play well and I can estimate his rating based on this game as 1800-1900 but rating is not based on one game. My rating is close to 2000. >If you play for a draw against a GM, you are fried. >Not to mention that if you are out of book early, you can kiss your game >goodbye. I do not know. It may be interesting to see some database of games of GM's against weak players who can draw often against computers to see if you are right. >>The last tournament when tiger14(anti human mode) played against humans proved >>that it is not so easy to draw against programs(tiger beated most of the players >>and they were clearly better than 2150 USCF rating). > >From the games I saw, most of the players played normal chess Nobody forced them to play normal chess and I guess that if they thought that it is easy to change their style and draw they could do it. Uri
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.