User Guide

How to Play A Game

To play a game you need a Graphical User Interface (GUI). Refer to your GUI’s Help document for instructions on how to install a chess engine and start a game.

How To Weaken Playing Strength

MadChess is difficult to defeat, unless you’re an international master (2400 ELO). For a more enjoyable game, you may wish to weaken the engine. In the Shredder and Hiarcs Chess Explorer GUIs, set the playing strength of the engine in the new game dialog.


Not all GUIs provide a playing strength option in the new game dialog. In GUIs that lack this feature, you may weaken the engine by setting the LimitStrength and ELO parameters. If the LimitStrength parameter is enabled, all other parameters are ignored (except ELO).

MadChess ELO values have has been tested against Shredder, a chess engine known for its realistic, human-like play at reduced strength. See ELO Ratings for a list of typical chess players (of varying enthusiasm and ability) and their corresponding rating.

How To Create A Personality

Another way to weaken the engine’s playing strength is to create a personality by adjusting parameter values (without enabling the LimitStrength parameter). You may disable the engine’s understanding of positional factors such as pawn structure, piece location and mobility, king safety, etc.

In addition, you may restrict the engine’s search speed and force it to commit errors and blunders. The following engine parameters have a great effect on playing strength, much more than the positional understanding parameters.

  • NPS
  • MoveError
  • BlunderError
  • BlunderPercent

Refer to Engine Parameters for an explanation of all MadChess engine parameters.

Refer to your GUI’s Help document for instructions on how to modify and save engine parameters.

How Does MadChess Play Chess?

For a technical explanation of how MadChess searches and evaluates chess positions, see the Developer Guide.

How Does MadChess Play Chess As Poorly As Me?

For a technical explanation of how the strength of MadChess is reduced, see The MadChess UCI_LimitStrength Algorithm.

Comments are closed.