Please turn on javascript in your browser to play chess.
Only Chess Forum

Only Chess Forum

  1. Subscriber Marinkatomb
    wotagr8game
    29 Oct '12 18:05 / 5 edits
    I really don't pay attention to Engine v Engine matches as they're generally quite dull and don't really serve too well as study material. However, as anyone who has checked chessbase today will see, there is a minor classic of a game which highlights just how far computer chess has come. What do people think of this? If Paul Morphy was playing black here then this game would probably have been printed and reprinted hundreds of times. Black is actually down 4 pawns at one stage, but the piece activity is just mind blowing!



    A link to the article..
    http://www.chessbase.com/newsdetail.asp?newsid=8591

    I recommend watching the video that is posted also... 🙂
  2. 29 Oct '12 18:42
    I was expecting a game between two players here at redhotpawn. 😳
  3. Subscriber Marinkatomb
    wotagr8game
    29 Oct '12 18:51
    Originally posted by paulbuchmanfromfics
    I was expecting a game between two players here at redhotpawn. 😳
    Did you play through the game? It is a CRACKING game!

    The analyst talking in the video (http://www.youtube.com/watch?feature=player_embedded&v=xWdMqvGMxF4) raises an interesting question.

    Given these new developments in computer evaluation of positions, are we seeig the beginning of a 'romantic' era in computer chess? Houdini doesn't just look at the material on the board, it is judging the piece activity and material position in tandem. It actually passes up a number of moves that would restore material equality in this game and in fact sacrifices further material! It's a very instructive game if you look it through a few times.. 🙂
  4. 29 Oct '12 19:33
    Originally posted by Marinkatomb
    Did you play through the game? It is a CRACKING game!

    The analyst talking in the video (http://www.youtube.com/watch?feature=player_embedded&v=xWdMqvGMxF4) raises an interesting question.

    Given these new developments in computer evaluation of positions, are we seeig the beginning of a 'romantic' era in computer chess? Houdini doesn't just look at t ...[text shortened]... fices further material! It's a very instructive game if you look it through a few times.. 🙂
    Yes, I played through it (and gave it the thumbs up). I enjoyed it. Nxh7 is a computer move that starts the ball rolling. Most players generally don't grab those rook pawns like that. It's unbeliavable that there is enough time/comp for moves like Bd7 and a5. Being able to see that white is too tied down in every possible variation is another computer trait. It's shocking that black has enough time to pull this off. There's not really much else to say. I did notice that you ask which side is preferred on move 28. Black is actually winning a piece there by force. White has the compensation in pawn numbers, but his devlopment is nearly nil. Black is in the driver's seat all the way. If Rybka gets beat down like this, what chance does a human have?

    P.S. In an unrelated question, is this the same Rybka as years ago. I remember a friend of mine used to boast about Rybka being the strongest and all that. Then, Rybka's creator got sued or something for ripping off another program (if I rememeber correctly). Did they have to go back and modify Rybka to stay out of trouble, or is this the same super strong Rybka from years ago?

    Anway, that's a nice game, and I enjoyed it very much.
  5. Subscriber Marinkatomb
    wotagr8game
    29 Oct '12 19:48
    Originally posted by paulbuchmanfromfics
    Yes, I played through it (and gave it the thumbs up). I enjoyed it. Nxh7 is a computer move that starts the ball rolling. Most players generally don't grab those rook pawns like that. It's unbeliavable that there is enough time/comp for moves like Bd7 and a5. Being able to see that white is too tied down in every possible variation is anoth ...[text shortened]... strong Rybka from years ago?

    Anway, that's a nice game, and I enjoyed it very much.
    Yes i only commented because white is just lost (amazingly quickly as well).

    Like i said, i don't really follow computer chess but the article says that this is the same version of Rybka that won the computer WC 4 years in a row! Apparently Houdini does something called 'stacking' (don't ask me what that is) which is something to do with disregarding certain lines after a shallow evaluation in order to follow more promising lines, sometimes up to 40 ply deep. It's no wonder it won this game if you consider this, it could have conceivably have worked out the entire game by move 15! 🙄
  6. Donation ketchuplover
    G.O.A.T.
    29 Oct '12 20:43
    not too shabby