Unbeatable

[jsgroby]

I read this statement over the weekend. Not sure if it is true. What are your thoughts?

No human has beaten a sufficiently strong chess engine under official tournament conditions since 2005.

Really?

Jeff

[hgm]

This is a so-called meaningless statement.

Any engine that was beaten was obviously not 'sufficiently strong'. So there is no way it could ever be false. It is a tautology.

[Thomas Lagershausen]

This statement is totally bullshit.

Between humans and computers didn´t happened any matches since a lot of years.

For a human it is easy to play against a computer. The human plays a closed opening and the computer is helpless.

A so prepared grandmaster can draw almost every game against a computer.

[hgm]

'Draw' is of course not the same as 'beat'...

If no games were played, the statement is also automatically true. You cannot be beaten without playing.

[Graham Banks]

This statement is totally bullshit.

Between humans and computers didn´t happened any matches since a lot of years.

For a human it is easy to play against a computer. The human plays a closed opening and the computer is helpless.

A so prepared grandmaster can draw almost every game against a computer.

I disagree. Even most of the GMs seem to disagree too from what you read and hear.
Not unbeatable, but certainly minimal chances of winning many games.

[rcmaddox]

This statement is totally bullshit.

Between humans and computers didn´t happened any matches since a lot of years.

For a human it is easy to play against a computer. The human plays a closed opening and the computer is helpless.

A so prepared grandmaster can draw almost every game against a computer.

The OP's statement is, as pointed out, tautological, and it's also true that GM's aren't playing matches against computers these days.

But I disagree that "anti-computer" closed position strategies will enable GM's to draw almost every game. The matches that have been played since 2005, including the odds matches some years ago, suggest otherwise. And there's this:

"Computer programs these days are actually stronger than human players, significantly stronger. If I played a computer in a match I would get destroyed, I would be slaughtered, and that's saying something considering where I'm at in the world of chess." GM Hikaru Nakamura - Q&A at Wahington University, St Louis, MO 2/26/11

[jsgroby]

This statement is totally bullshit.

Between humans and computers didn´t happened any matches since a lot of years.

For a human it is easy to play against a computer. The human plays a closed opening and the computer is helpless.

A so prepared grandmaster can draw almost every game against a computer.

I disagree. Even most of the GMs seem to disagree too from what you read and hear.
Not unbeatable, but certainly minimal chances of winning many games.

Here is the article I came across with that statement in it. Part of it was an April Fool's joke, but the part of the article with the statement in it is before the link to the joke.

http://scienceblogs.com/startswithabang ... st-solved/

[kranium]

This statement is totally bullshit.

Between humans and computers didn´t happened any matches since a lot of years.

For a human it is easy to play against a computer. The human plays a closed opening and the computer is helpless.

A so prepared grandmaster can draw almost every game against a computer.

http://en.wikipedia.org/wiki/Human%E2%8 ... ss_matches

In November 2005, 3 former FIDE world chess champions, Alexander Khalifman, Ruslan Ponomariov and Rustam Kasimdzhanov played against computers Hydra, Junior and Fritz. The computers won 8 to 4.[18][19]

Ponomariov – Junior 0–1
Hydra – Kasimdzhanov 1–0
Fritz – Khalifman 1–0
Ponomariov – Fritz 1–0
Kasimdzhanov – Junior 1/2–1/2
Khalifman – Hydra 1/2–1/2
Hydra – Ponomariov 1–0
Fritz – Kasimdzhanov 1/2–1/2
Junior – Khalifman 1–0
Ponomariov – Junior 1/2–1/2
Kasimdzhanov – Hydra 1/2–1/2
Khalifman – Fritz 1/2–1/2

I totally disagree Thomas..

now, 10 years later, chess engines are vastly improved, and much much stronger (400 ELO or more?)
top GMs (and even world champions) have no chance whatsoever, IMHO

...that's why no GM today will play engine matches, even for big money...they don't want to (and for the sake of their career, can't afford to) risk such humiliation

[lucasart]

I read this statement over the weekend. Not sure if it is true. What are your thoughts?

No human has beaten a sufficiently strong chess engine under official tournament conditions since 2005.

Really?

Jeff

Yes. In fact, I am surprised about the 2005. I would have guessed 2000 or so. In 2005, who won? in what conditions? against which program? Fritz and Junior were not the strongest, even in 2005. At the time it was Shredder and Fruit.

Note that we're talking about beating a computer in tournament conditions. It does not mean winning a game every now and then, and losing 90%+ games. It does not mean winning on your own comp (no referee watching), with access to the computer's score and PV. Or by repeating the same opening again and again until you start to memorize the game until you reach a favorable position, by trial and error. Or by taking back moves. People can win games like that, but not tournaments, not in tournament conditions.

[reflectionofpower]

This statement is totally bullshit.

"Computer programs these days are actually stronger than human players, significantly stronger. If I played a computer in a match I would get destroyed, I would be slaughtered, and that's saying something considering where I'm at in the world of chess." GM Hikaru Nakamura - Q&A at Wahington University, St Louis, MO 2/26/11

I remember posting this segment of the video. Here is where he states this on You Tube: