Random Posts

  • Anthony Santasiere Plays the King's Gambit
  • Tweaked Version of Stockfish 7
  • Palma de Mallorca 1966...I was there but missed it
  • Lechenicher SchachServer Ratings
  • Chess, Strikes, Baseball, WW2 and a Long Forgotten Game
  • New LSS “World Champion”
  • US Chess Hall of Fame
  • Dilemma of the Engines
  • Two Obituaries: Walter Shipman and James L. Harkins, Jr.
  • An Amazing Game by Geza Maroczy
  • Monday, April 14, 2014

    Which Analysis Engine?


         I found this question asked on a forum and the one of the “engine experts” gave his opinion which was based on the number of engines being used (I assume when using Aquarium’s IDeA).  What I found interesting was his evaluation of the strengths of the engines.  One note though: the post was made a little over a year ago and so is somewhat outdated but it gives a general idea of the capabilities of different engines, especially the free versions.

    Top choice for a single engine - Houdini 3
    Stockfish - sometimes better in closed positions
    Houdini 1.5a - reliable for an analysis in depth
    Critter 1.6a sometimes better in endgame analysis and the second best after Houdini
    Komodo 5 - sometimes yields some good ideas that are different from the others

    Ivanhoe B46 - like Komodo 5
    Deep Rybka 4.1 - it's good
    Zappa Mexico II - because of its unique options for solving tactical positions that Houdini 3 "tactical mode" won't solve
    Critter 0.90 - good for some endgame analysis and closed position analysis
    Deep Fritz 10 - the most "human" engine

    Houdini 2.0c is best on short time controls and it prunes lines recognized as bad better than others
    Houdini 1.5a - almost as good as Houdini 2.0c
    Critter 1.4a - some say it's better in closed positions and good at endgames without tablebases…very fast, but not as fast as Houdini 2.0
    Rybka 4.1 is…most balanced one in all kinds of positions
    Komodo 4 has original dynamic evaluation of the pieces' activity.
    Strelka 5.1 - Very fast and original but with wrong numerical evaluations (for example +1.47 in a += position) which can be trusted only for a suggestion of the best move
    Stockfish 2.2 or Ivanhoe 9.. - both good for a second engine
    Robbolito 0.10 SMP - good as a second analyzing engine, too. It doesn't use endgame tablebases though

    No comments:

    Post a Comment