K+3N v K and K+4N v K

  • #1

    Does anyone know where I can find the method for mating with a king and three knights against a lone king?

     

    Also, is there any reason why in the 6-man endgame tablebases (ie. 4 pieces plus the kings) there is no K+4N v K? I know it's a crazy situaiton, but it is certainly possible, and *might* occur. It would be easy to just say "sac one of the knights and play from there" but that does not really give the solution. (Solved in post #2)

  • #2

    On Nalimov, which is here:

    http://www.k4it.de/index.php?topic=egtb&lang=en

    there is this information: *** NEW! All 6-men tables online (except 5 vs. 1). ***

    Since K+4N v K is 5 vs 1 it won't work. You would probably need to ask the author(s) as to why 5 vs 1 is not included. Maybe it is hard to program, who knows?

  • #3
    BlueKnightShade wrote:

    On Nalimov, which is here:

    http://www.k4it.de/index.php?topic=egtb&lang=en

    there is this information: *** NEW! All 6-men tables online (except 5 vs. 1). ***

    Since K+4N v K is 5 vs 1 it won't work. You would probably need to ask the author(s) as to why 5 vs 1 is not included. Maybe it is hard to program, who knows?


    Of course! I read that, (I use the same site for my EGTB searches), I clearly just forgot how to count to 5. Are you aware if the 5v1 EGTB has been generated, and if so is there somewhere to view it on the web?

  • #4

    5 vs. 1 is not included in the tablebase because any decent engine can calculate a win with its usual algorithm for any of those position. (Fritz 8 had no trouble with K + 4N vs. K).

  • #5
    Loomis wrote:

    5 vs. 1 is not included in the tablebase because any decent engine can calculate a win with its usual algorithm for any of those position. (Fritz 8 had no trouble with K + 4N vs. K).


    Good to know. Thanks.

  • #6

    No I don't know if the 5v1 EGTB has been generated. I looked a bit around on the Internet and just got the impression that it is something that takes a huge amount of calculation. It is probably a big task, but I guess it will be accomplished sooner or later.

  • #7

    The trouble is not with how long it takes to calculate. The calculation of the 5 vs. 1 positions would be considerably shorter than 4 vs. 2 and 3 vs. 3 because one side is going to mate fairly quickly in 5 vs. 1. 

    The point is that as it is the tablebase is already large, 1.2 terabytes. So it is worthwhile to cut out positions that can be won some other way.

  • #8
    Loomis wrote:

    The trouble is not with how long it takes to calculate. The calculation of the 5 vs. 1 positions would be considerably shorter than 4 vs. 2 and 3 vs. 3 because one side is going to mate fairly quickly in 5 vs. 1. 

    The point is that as it is the tablebase is already large, 1.2 terabytes. So it is worthwhile to cut out positions that can be won some other way.


    But surely the aim of EGTB generation is that eventually you will encapsulate all possible board positions, thus solving the game?

  • #9

    No, I don't think that's the aim at all. 

Top
or Join

Online Now