Alert icon
We're changing our privacy policy. This stuff matters.  Learn more  Dismiss

Chess Endgame: King, Knight and Rook Pawn vs. King

Loading...

Sign in or sign up now!
Alert icon
Upgrade to the latest Flash Player for improved playback performance. Upgrade now or more info.
2,252
Loading...
Alert icon
Sign in or sign up now!
Alert icon

Uploaded by on Jun 10, 2010

An endgame where a knight and pawn advantage proves insufficient.

Also I rant about artificial intelligence in chess.

Category:

Sports

Tags:

License:

Standard YouTube License

  • likes, 3 dislikes

Link to this comment:

Share to:

Uploader Comments (GreenCastleBlock)

  • I just use a table base. Perfect endgame play :) I just put the starting position into it and you are right, Kc2 is a draw, and Kc1 is a mate in 11.

  • @Involuntarity Okay, what if the board were 26x26 and the Knight was on W23? Pretty sure that's not in your tablebase :)

  • In this particular position, all the AI has to do is to compute the min distance(moves) from the current knight position to the target squre( D4, E3, etc, since they are equivalent) . So even if there is large number of n (squares), it's still easy to use this method to determine the best move. Rybka fails to do so maybe because the author didn't implement this particular position(cornered pawn/king with enemy king one step away to control the escaping square and with a knight) into it.

  • @arthas702 This assumes the computer can recognize the concept of a "target square" applies

  • @GreenCastleBlock I think the target square is those that controls C2, and of course, we have to specify that for the AI in the database.

  • @arthas702 But the concept of a "target square" that you must get a piece to and win is a very specific strategy that only applies to a few kinds of position. How do you intend to have your computer know when to start looking for such a square?

see all

All Comments (45)

Sign In or Sign Up now to post a comment!
  • I found this video by looking for king rook and knight endgame, cause I had this game where my opponent has a pawn on h7 my king was on h7 and his king was on h6 and I had a rook on g1 it was his turn (white) and I was thinking of Rh1+ and pick up the queen but he choosed a knight and I never trained myself to mate with king and rook against king and knight I had to draw what a shame.Nice video btw eventho it's not what I was looking for I enjoyed it, your explenation was crystal clear

  • @GreenCastleBlock thank you verry much, this is no kidding im telling you this is the absolute best educative chess video i ever watched in my life... its so simple but i think it will enrich my chess understanding so verry much.. its really awesome, thank you for sharing that

  • Why are you harping on the computer again? I highly doubt that a human would be able to get this far if the game were 64,000 squares big.

  • So in this type of position, just move the king on the same color square as the square the opponent's knight is on.

  • Actually, I just wanted to say that Rybka 4 will be able to tell that the position at 2:00 will be drawn.

  • Hi Matt, very good video and thanks for sharing. I now wonder how many times thru the years I have been both black and white in this very postion and either won or lost when it should have been a draw because there was no easy or quick way to determine what the correct move was for either side, that is until now. :)

  • @GreenCastleBlock But that doesn't matter ?

  • @Involuntarity

    Computers are not necessarily better at calculating, only faster. That in turn allows them to consider many more variations than a human player and (usually) come to a better battleplan.

    But they are not better at calculating, since they make the same calculations as a human player.

    Humans could never let a computer do things they have not come up with themselves in the first place. The speed at which a computer can do the calculations is what makes them so good at chess.

  • @Involuntarity

    Computers are not necessarily better at calculating, only faster. That in turn allows them to consider many more variations than a human player and (usually) come to a better battleplan.

    But they are not better at calculating, since they make the same calculations as a human player.

    Humans could never let a computer do things they have not come up with themselves in the first place. The speed at which a computer can do the calculations is what makes them so good at chess.

  • Well made vid, Matt. That's some nice sound logic you just shared with us. When all the chess engine programmers start incorporating your givens and if then statements into their product, they will hopefully kick back some sort of royalties!

Loading...

Alert icon
0 / 00Unsaved Playlist Return to active list
    1. Your queue is empty. Add videos to your queue using this button:
      or sign in to load a different list.
    Loading...Loading...Saving...
    • Clear all videos from this list
    • Learn more