Can Houdini+Critter+ Stockfish+Komodo be Combined?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
pichy
Posts: 2564
Joined: Thu Mar 09, 2006 2:04 am

Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by pichy » Wed May 09, 2012 5:56 pm

Can Houdini+Critter+ Stockfish and Komodo be combined by the same way the Bonanza+Gekisashi+ GPS and YSS were combined to create AKARA?

It is amazing to see how AKARA select the best move based on vote as it is explained here by analazing its log file.
http://www.youtube.com/watch?v=lUnbzhnDIvA

User avatar
JuLieN
Posts: 2948
Joined: Mon May 05, 2008 10:16 am
Location: Nantes (France)
Contact:

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by JuLieN » Wed May 09, 2012 6:02 pm

Jorge, read your PMs, please ;)
"The only good bug is a dead bug." (Don Dailey)
Image [Blog: http://tinyurl.com/predateur ] [Facebook: http://tinyurl.com/fbpredateur ] [MacEngines: http://tinyurl.com/macengines ]

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by fern » Wed May 09, 2012 9:14 pm

Maybe it can be done if we could also combine Shakespeare, Joyce, Anatole France and Hemingway in just one novel.

Fern

syzygy
Posts: 4809
Joined: Tue Feb 28, 2012 10:56 pm

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by syzygy » Wed May 09, 2012 9:38 pm

It is of course possible to feed all engines the same position and then pick the move that has a plurality of engine votes. How strong the resulting system would be compared to the individual engines I do not know.

The result would probably be prettier, but less interesting, than a novel written by letting Shakespeare, Joyce, Anatole France and Hemingway vote on each next word ;-)

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by fern » Wed May 09, 2012 9:57 pm

To put any number of programs giving votes for a move and then having a simple device to play the most voted is not problem at at all, but there is in that nothing like a fusion of ideas or whatever, just a democratic process. And certainly what we could get is not the "best" move as a supposedly result of a common acumen. "Best" depends of the parameter of what-is-good or bad that comes from some architectural design.
Fern

karger
Posts: 196
Joined: Tue Feb 02, 2010 1:27 am

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by karger » Thu May 10, 2012 12:45 am

There was a version a few years where the author directly combined Grapefruit and Cyclone.

User avatar
fern
Posts: 8755
Joined: Sun Feb 26, 2006 3:07 pm

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by fern » Thu May 10, 2012 2:09 am

and it was better than any of those separated?

Fern

S.Taylor
Posts: 8484
Joined: Thu Mar 09, 2006 2:25 am
Location: Jerusalem Israel

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by S.Taylor » Thu May 10, 2012 10:14 am

fern wrote:Maybe it can be done if we could also combine Shakespeare, Joyce, Anatole France and Hemingway in just one novel.

Fern
Mix creationists plus non-creationists plus tomato juice.
Then you MUST get to the truth!!!
lol

karger
Posts: 196
Joined: Tue Feb 02, 2010 1:27 am

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by karger » Thu May 10, 2012 4:42 pm

Grapecyclone was never released by the author with source code and GPL license. It was only released on the author's website as an exe. It was tested to be about 2900 elo, with a very aggressive balanced style of play.

pichy
Posts: 2564
Joined: Thu Mar 09, 2006 2:04 am

Re: Can Houdini+Critter+ Stockfish+Komodo be Combined?

Post by pichy » Thu May 10, 2012 5:09 pm

fern wrote:To put any number of programs giving votes for a move and then having a simple device to play the most voted is not problem at at all, but there is in that nothing like a fusion of ideas or whatever, just a democratic process. And certainly what we could get is not the "best" move as a supposedly result of a common acumen. "Best" depends of the parameter of what-is-good or bad that comes from some architectural design.
Fern


I feel like joking a little bit too, but my memory is trying to remember that around 1993 a three brains consisting of Chessmachine King (programmer Johan de Koning), Mephisto Lyon (programmer Richard Lang) and operator Dr.Ingo Althöfer, a German mathematician and chess amateur rated 1960 at the time. Third brain Althöfer selected one of the moves computed by the engines or played the move both programs agreed upon. In the 1993 Aegon tournament 3-Hirn scored 4 out of 6, with a TPR of 2401. Win against Nico Kuyf, draw against GM Vlastimil Hort, draw against Alexander Münninghof (Euwe's biographer), win against WIM Jessica Harmsen, win against Michiel Wind and loss against Bronstein in the last round.

PS: That was with three weak engines, imagine what 3 top engines and a human brain selecting the best move can accomplish and for the human, GM Larry Kaufman replacing Komodo to select the best move would be just fine. :wink:

Post Reply