Can A GM And Rybka Beat Stockfish?

[rcmaddox]

The evidence shows that a very strong human chess player like Valeri Lilov, Gary Kasparov, Magnus Carlsen or even someone like Boris Gelfand, Vadim Milov or Ruslan Ponomariov can beat Stockfish even without the help of Rybka.

Chess engines have limits in their chess understanding and how far they can see into the game while a strong human GM can see these limits that chess engines obviously have and exploit them.

What evidence? In a long match they might draw a few and possibly win 1 game but they would lose by a big margin.

I guess his including Valeri Lilov in the list of players should have been my first clue.

[APassionForCriminalJustic]

The evidence shows that a very strong human chess player like Valeri Lilov, Gary Kasparov, Magnus Carlsen or even someone like Boris Gelfand, Vadim Milov or Ruslan Ponomariov can beat Stockfish even without the help of Rybka.

Chess engines have limits in their chess understanding and how far they can see into the game while a strong human GM can see these limits that chess engines obviously have and exploit them.

What evidence? In a long match they might draw a few and possibly win 1 game but they would lose by a big margin.

I guess his including Valeri Lilov in the list of players should have been my first clue.

None of these guys would even have a chance. It is one thing to have Rybka at least assist you in match play (and even Rybka does not do much), but to actually contend against SF is about as futile as trying to punch your way through a steel door. Human beings rely heavily on intuition and judgment with some tactical play of course - they would not be able to make sense of quite a few of the moves that would be played in front of them. This whole evidence shows crap is one very stupid, and dumb comment to say the least.

Human beings cannot exploit anything as they simply are way too out searched, and they will make many inferior moves in each game. They are no match at all. Humans belong with playing other human beings. What they should not be doing is trying to play a machine which turns chess into a game of mathematical impossibility, and tactical precision that humans cannot play.

Even with all of this said, engines understand more than enough. Engines today are not the engines of 10 years ago when they really were stupid.

[Henk]

The evidence shows that a very strong human chess player like Valeri Lilov, Gary Kasparov, Magnus Carlsen or even someone like Boris Gelfand, Vadim Milov or Ruslan Ponomariov can beat Stockfish even without the help of Rybka.

Chess engines have limits in their chess understanding and how far they can see into the game while a strong human GM can see these limits that chess engines obviously have and exploit them.

What evidence? In a long match they might draw a few and possibly win 1 game but they would lose by a big margin.

I guess his including Valeri Lilov in the list of players should have been my first clue.

...
Even with all of this said, engines understand more than enough. Engines today are not the engines of 10 years ago when they really were stupid.

There are exceptions like Skipper. You are talking about good engines.

[Uri Blass]

The evidence shows that a very strong human chess player like Valeri Lilov, Gary Kasparov, Magnus Carlsen or even someone like Boris Gelfand, Vadim Milov or Ruslan Ponomariov can beat Stockfish even without the help of Rybka.

Chess engines have limits in their chess understanding and how far they can see into the game while a strong human GM can see these limits that chess engines obviously have and exploit them.

What evidence? In a long match they might draw a few and possibly win 1 game but they would lose by a big margin.

I guess his including Valeri Lilov in the list of players should have been my first clue.

None of these guys would even have a chance. It is one thing to have Rybka at least assist you in match play (and even Rybka does not do much), but to actually contend against SF is about as futile as trying to punch your way through a steel door. Human beings rely heavily on intuition and judgment with some tactical play of course - they would not be able to make sense of quite a few of the moves that would be played in front of them. This whole evidence shows crap is one very stupid, and dumb comment to say the least.

Human beings cannot exploit anything as they simply are way too out searched, and they will make many inferior moves in each game. They are no match at all. Humans belong with playing other human beings. What they should not be doing is trying to play a machine which turns chess into a game of mathematical impossibility, and tactical precision that humans cannot play.

Even with all of this said, engines understand more than enough. Engines today are not the engines of 10 years ago when they really were stupid.

It is obvious that GM's cannot win.
It is not obvious for me that they cannot draw most games.

comp-comp games prove nothing because computers usually do not play for a draw.

A strong GM who know that the computer is stronger may play for a draw and trade pieces even at the price of being a pawn down in some endgame that he knows that he can defend.

It is not the way that GM's usually play because they usually play against weaker players so they have no reason to use it but maybe they can change their style.

The question if some GM can draw most games against Stockfish or Komodo in tournament time control is an open question and nobody offered a big prize money in order to give GM's motivation to try it.

[Milos]

A strong GM who know that the computer is stronger may play for a draw and trade pieces even at the price of being a pawn down in some endgame that he knows that he can defend.

It is not the way that GM's usually play because they usually play against weaker players so they have no reason to use it but maybe they can change their style.

The question if some GM can draw most games against Stockfish or Komodo in tournament time control is an open question and nobody offered a big prize money in order to give GM's motivation to try it.

What is most of the games 50-60%, so GM would have around 30% against SF or Komodo in FIDE TC???? This is total BS.
With good opening book, and proper contempt factor any GM (Carlsen included) would not have more than 10% score. Put 100 million prize like in box match it would still be miserable 10% or less.

[Ozymandias]

The question if some GM can draw most games against Stockfish or Komodo in tournament time control is an open question and nobody offered a big prize money in order to give GM's motivation to try it.

At least your confidence, in the GMs ability to draw games against engines, is consistent with past statements from you. Some nice betting could ensue, if some stronger GM where to take white, in a reenactment of the "Draw Odds Match with Benjamin", against (say) SF running on an iPad Air 2.