Re: Letsplaychess.com presents: Learning from a chess engine - Alekhine Defence with 2.Nc3

[MrTrigov]

Rg8 doesn't help after deep analyze with rybka black still can be mated

[Fairfield1466]

Fritz 12 found Ng6+ in 51 seconds on my slow computer. This move was best after 2 minutes, 3 minutes and 5 minutes as well, so it didn't "accidentally" get it right.

[MrZingnigga]

why can an engine not find the knight sac at depth 10? it pays off only 3-4 moves later.

[madrashot]

Good video, but quoting Rybka at "depth 8" (round about 7 mins in) doesn't really show very much in my opinion... you need to leave it for a good few mins before you get anything sensible/ worth looking into (other than obvious tactics).. you can bet the players thought for more than a few seconds!

[jbahadur10]

If you want to expose the weakness of chess engines, I suggest posting a very deep positional game by someone like Fischer or Karpov and show how the computer doesn't even consider a move that humans recognize as brilliant.

Even a game like that great positional queen sacrifice you posted is more suitable... even though my engine considered the queen sac, it was not its preferred move, and I don't think the computer really understood how good it was.

[jbahadur10]

It's a nice game, but I'm not sure it's a great advertisement for the limitations of engines. Computers can analyze open tactical positions like this pretty quickly. Like majnu, I analyzed with Fritz 10 and it spotted Ng4 almost immediately, but rates white at only about equal after 34...Rg8. If Rh3+, white can no longer capture on g6 after Kg7. Alekhine simply played inaccurate defense with Rf7, a fact that the computer is very good at pointing out.

[Razieluk]

I think beginners will get a lot from studying chess engines. The game shown here is from two top players so it's not surprising that the computer didn't find the right move.

[pussuw]

Actually, rybka found Ng6+ on depth 11.

But after h7xg6 Qg4 finds the defensive resource Rg8 as mentioned before and evaluates the position as completely equal 0.00. At about depth 20 it evaluates it as a small advantage to white but still drawing.

[sullathegenius]

I wish chess engines didnt even exist. I think they are killing the game of chess.

[GilesFarnabyChess]

@sullathegenius Yeah, as printing killed manuscripts and computers typewriters, we should rather all go leave in a cave

[xXxWilliExXx]

I can thoroughly recommend My System to any keen player (a basic understanding of the game is needed). By getting a feel for some of Nimzowitsch's ideas (many of which are still relevant today) and playing through his annotated games, I feel like I have improved a lot.

Anyway, I just felt like sharing that for anyone who is considering getting the book. Nimzowitsch was quite a witty writer as well, so the book is enjoyable in more ways than one!

Nice commentary.

[gologram]

Nimzovich,as well as Tal,is my countyman.Yesterday I sat the same place where Tal used to sit thinking over the board.

[Majnu2006]

Hi kingscrusher,

Fritz 10 does find 33.Ng6+ after 2 minutes, but rejects it because black has a better defence than in the game was played.

Variation:

33.Ng6+, hxg6

34.Qg4, Rg8! (not 34...Rf7?)

and it seems that there is no winning continuation for white!

I suppose that Nimzowitsch and Alekhine both missed 34...Rg8.

[secret103]

What is a game engine? Especially for video games like the ones on the xbox360.

[brrrdman85]

Great example of computer deficiencies. The best way to beat an engine is to make subtle positional moves that might help further down the line. Getting into strict tactical lines is no good, because the computer will out-calculate a human opponent every time. Great video.

[FarbrorBergman]

Acually my compoter finds the move 33.Ng6+ but it doesnt follow up with 34.Qg4 but 34. fxg6 and later goes on to win after ...Qc7 35.Qh3+ ..Qg8

36.Rxc3 the compoter says that black goes from +2.55 to -3.58