Melik & Me with GM Khachiyan and BrotherJosh is LIVE on Chess.com/TVClick here to watch!
Upgrade to Chess.com Premium!

Can Houdini 3 be beaten?

  • 3 months ago · Quote · #21

    C-nack

    This is so full of crap, and anyway 9.Ng4 is an invalid move, I think you meant Ng3.

  • 3 months ago · Quote · #22

    EscherehcsE

    Cnacnel wrote:

    This is so full of crap, and anyway 9.Ng4 is an invalid move, I think you meant Ng3.

    Yeah, 9.Ng3 is what he meant. The rest of the moves are correct. To get it into PGN shape, you'd have to clean it up a bit (e.g. change several occurences of ch to +).

  • 3 months ago · Quote · #23

    Rasparovov

    First example of a game is not played by Houdini 1.5 on a strong computer that's for sure. 
    Nothing more to do here. 

  • 3 months ago · Quote · #24

    C-nack

     

     Sorry, but this is just stupid, Houdini on full strength doesn't make mistakes like 24... Ba4 or 27... Rh5. You must've done something wrong in preferences. Just a couple of things, can't bother to show all variations, mistakes and inaccuracies.



  • 3 months ago · Quote · #25

    sloughterchess

    nameno1had wrote:

    Maybe we could set up a match between you and Houdini 3 through the community...this will lay this to rest...

    If a willing participant would come forth that owns Houdini 3...the two of you could continue to exchange moves in the forum and we could follow along with updated game sequence boards.... ?

    Any takers, have this already been proven this is a sham ? Let's find out...

    With random opening choices, Houdini will win easily; however, if this is a test of the soundness of the Two Knights' Defense then:

    I believe that the Berliner Gambit and the alternative to the Berliner given in MCO 15 are both +- i.e. 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nd4 6.c3 b5 7.Bf1 Nxd5 8.Ne4 Qh4? (Ne6 9.Bxb5ch Bd7 10.Bc4 +/-) 9.Ng3 Bg4 10.f3+- or 9.Ng3 Bb7 +-. The Ulvestad where Black plays 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 b5 6.Bf1 h6 (book according to MCO 15) is +-. The Wilkes Barre/Traxler 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 is +/- The main line 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Qf3 is only +/=,  and a tougher test. Houdini should win in the main line. It is also likely that Houdini will win with Black in the Fried Liver (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.Nxf7) because the Fried Liver is close to equal; the Lolli looks equal, but Houdini may not know correct book (Heisman Variation---1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.d4 Nxd4!=); if it plays known book it might well lose. I'd be happy to take on Houdini 3 to test any of these openings with the understanding I'll be lucky to draw in the main line or against 5...Nxd5. Since I can't play endgames well then an evaluation of +- in the late  middlegame should be an adequate test.

  • 3 months ago · Quote · #26

    nameno1had

    sloughterchess wrote:
    nameno1had wrote:

    Maybe we could set up a match between you and Houdini 3 through the community...this will lay this to rest...

    If a willing participant would come forth that owns Houdini 3...the two of you could continue to exchange moves in the forum and we could follow along with updated game sequence boards.... ?

    Any takers, have this already been proven this is a sham ? Let's find out...

    With random opening choices, Houdini will win easily; however, if this is a test of the soundness of the Two Knights' Defense then:

    I believe that the Berliner Gambit and the alternative to the Berliner given in MCO 15 are both +- i.e. 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nd4 6.c3 b5 7.Bf1 Nxd5 8.Ne4 Qh4? (Ne6 9.Bxb5ch Bd7 10.Bc4 +/-) 9.Ng3 Bg4 10.f3+- or 9.Ng3 Bb7 +-. The Ulvestad where Black plays 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 b5 6.Bf1 h6 (book according to MCO 15) is +-. The Wilkes Barre/Traxler 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 is +/- The main line 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Qf3 is only +/=,  and a tougher test. Houdini should win in the main line. It is also likely that Houdini will win with Black in the Fried Liver (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.Nxf7) because the Fried Liver is close to equal; the Lolli looks equal, but Houdini may not know correct book (Heisman Variation---1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.d4 Nxd4!=); if it plays known book it might well lose. I'd be happy to take on Houdini 3 to test any of these openings with the understanding I'll be lucky to draw in the main line or against 5...Nxd5. Since I can't play endgames well then an evaluation of +- in the late  middlegame should be an adequate test.

    Yeah, drawing probably is attainable if you have Houdini 3 and so does the person you'd be playing...it wouldn't be rocket science...

  • 3 months ago · Quote · #27

    sloughterchess

    Shivsky wrote:

    what was the time control setting for Houdini 3?

     

     

    The problem is not time control; the problem is I bought a POS from a mass outlet that rhymes with smart. It has 4 GB memory and 325 GB hard drive, but a celeron!! processor. My lap top is running through molasses between Windows 8 and the Houdini program; my computer guy is building me a computer that can run the optimum systems requirement of the program.

  • 3 months ago · Quote · #28

    C-nack

    I would also win against 4-ply Houdini :D

  • 3 months ago · Quote · #29

    sloughterchess

    nameno1had wrote:
    sloughterchess wrote:
    nameno1had wrote:

    Maybe we could set up a match between you and Houdini 3 through the community...this will lay this to rest...

    If a willing participant would come forth that owns Houdini 3...the two of you could continue to exchange moves in the forum and we could follow along with updated game sequence boards.... ?

    Any takers, have this already been proven this is a sham ? Let's find out...

    With random opening choices, Houdini will win easily; however, if this is a test of the soundness of the Two Knights' Defense then:

    I believe that the Berliner Gambit and the alternative to the Berliner given in MCO 15 are both +- i.e. 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nd4 6.c3 b5 7.Bf1 Nxd5 8.Ne4 Qh4? (Ne6 9.Bxb5ch Bd7 10.Bc4 +/-) 9.Ng3 Bg4 10.f3+- or 9.Ng3 Bb7 +-. The Ulvestad where Black plays 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 b5 6.Bf1 h6 (book according to MCO 15) is +-. The Wilkes Barre/Traxler 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 is +/- The main line 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Na5 6.Bb5ch c6 7.dxc6 bxc6 8.Qf3 is only +/=,  and a tougher test. Houdini should win in the main line. It is also likely that Houdini will win with Black in the Fried Liver (1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.Nxf7) because the Fried Liver is close to equal; the Lolli looks equal, but Houdini may not know correct book (Heisman Variation---1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 d5 5.exd5 Nxd5 6.d4 Nxd4!=); if it plays known book it might well lose. I'd be happy to take on Houdini 3 to test any of these openings with the understanding I'll be lucky to draw in the main line or against 5...Nxd5. Since I can't play endgames well then an evaluation of +- in the late  middlegame should be an adequate test.

    Yeah, drawing probably is attainable if you have Houdini 3 and so does the person you'd be playing...it wouldn't be rocket science...

    The only "proof" that I am playing the game not just relying on Houdini 3 (which I will only use to blunder check) is that my choices will, occasionally, differ from Houdini's choices.

  • 3 months ago · Quote · #30

    nameno1had

    If you are blunder checking, you might as well just ask for it to give you the best moves....besides, if are doing analysis in multi pv mode and you are picking alternatives that are 0.03 pawns less, but still adding to or maintaining you positions value....that isn't you beating houdini....I fail to see how using houdini will gain you a victory over the same program. The only way in my opinion to beat it would be to have the 5 best players at longer time controls, use the next three best engines and try to take advantage af any flaw in houdini 's program eveluation, that is if one can be found that is great enough to exploit.....
  • 3 months ago · Quote · #31

    sloughterchess

    nameno1had wrote:
    If you are blunder checking, you might as well just ask for it to give you the best moves....besides, if are doing analysis in multi pv mode and you are picking alternatives that are 0.03 pawns less, but still adding to or maintaining you positions value....that isn't you beating houdini....I fail to see how using houdini will gain you a victory over the same program. The only way in my opinion to beat it would be to have the 5 best players at longer time controls, use the next three best engines and try to take advantage af any flaw in houdini 's program eveluation, that is if one can be found that is great enough to exploit.....
     
    From move 1 you are absolutely right. What I propose to do is "handicap" Houdini 3 by requiring it to play "book" as defined by MCO 15 which is not an ancient text. It came out in 2008, yet I predict that even Houdini cannot survive in book openings because all theory pre-Houdini, must be put through its brain, and, I predict, major innovations will be found.
     
    Maybe post members could carve up MCO 15 or BCO or ECO or New In Chess articles and just publish cooks of the main lines in all openings. This would be a real benefit to post members who then can organize these cooks and use them in their own games.
Back to Top

Post your reply:

Log In
Join
Post New Forum Topic
^ Ads keep Chess.com free. Upgrade to remove ads! ^
Chess.com TV - Live Broadcast!

"Melik and Me with GM Melik Khachiyan and BrotherJosh!"

^ Ads keep Chess.com free. Upgrade to remove ads! ^