Another XBoard engine derived from Minimax Basic

[Roland Chastain]

Hello! For my pleasure I converted Minimax Basic (by Dieter Steinwender and Chrilly Donninger) to a XBoard engine.

I know that this has already been done by Thomas McBurney some years ago, but this new version will be open source, and multiplatform, because it doesn't use Windows API. So I believe it could make sense to share it.

I just wonder how I could name it, to avoid confusion. For now my best idea is Minimax Basic 2020... Or maybe Minimax Basic II...

You can find the source code here, with a Linux executable. Not yet had time to compile for Windows.

Minimax.zip

The XBoard implementation is unfinished. I am learning the protocol. But the engine works. I tested it with XBoard and CuteChess.

Regards.

Roland

[Roland Chastain]

I uploaded Windows binaries: minimax.zip

I still have to find a name. Maybe Minimax II...

Code: Select all

Rank Name                  Elo    +    - games score oppo. draws 
   1 Needle 0.53.1         977  288  211     8   88%   647    0% 
   2 Winglet               916  244  205     8   81%   663   13% 
   3 Irina 0.15            770  189  172    16   75%   494    0% 
   4 Minnow 20150301       742  174  145    16   88%   421   25% 
   5 Quokka 2.0            669  154  140    16   78%   420   31% 
   6 Minimax Basic 0.0.1   598   83   79    64   71%   392   20% 
   7 MSCP 1.4              516   99   97    40   61%   398   13% 
   8 Iota 0.1              460  147  143    16   63%   349   25% 
   9 Hippocampe 0.4.2      305  115  119    32   34%   474    6% 
  10 NEG 1.3               196  172  235    16   13%   538    0% 
  11 Sachy 0.2.0.1         150  153  172    16   25%   392   13% 
  12 Alouette 0.1.3        128  112  128    32   20%   402    9% 
  13 Durandal 0.0.7        125  123  135    24   17%   416   33% 
  14 Random mover            0  187  276    16    3%   460    6% 

[karger]

Roland , What did you do to Minimax basic ? I have played against Thomas Mcburney's compile & it is 1700 or 1800 elo at best. Your windows compile hits a depth of 20 in 1 second under Shredder gui and must be btw 2800 & 3000 elo , Great engine thanks , Karger

[Roland Chastain]

Roland , What did you do to Minimax basic ? I have played against Thomas Mcburney's compile & it is 1700 or 1800 elo at best. Your windows compile hits a depth of 20 in 1 second under Shredder gui and must be btw 2800 & 3000 elo , Great engine thanks , Karger

Thank you for testing Minimax II. I suspect that there was something wrong in the configuration of your test. I cannot believe that Minimax be so strong. I did not change anything to the original program, and my implementation of the protocol is minimal: the engine ignores depth and time commands, and plays at fixed depth. After my own tests (see my previous message), the engine is currently somewhere between Iota 0.1 and Irina 0.15.

[Guenther]

Roland , What did you do to Minimax basic ? I have played against Thomas Mcburney's compile & it is 1700 or 1800 elo at best. Your windows compile hits a depth of 20 in 1 second under Shredder gui and must be btw 2800 & 3000 elo , Great engine thanks , Karger

Not sure, if your post is serious? If yes, you are doing something very wrong, most probably you have loaded another program instead of Minimax compiled by Roland.

Note that he measured it himself below the McBurney version. This is no surprise, because it has no time management, it plays immediately and the McBurney version also has other enhancements. Also Rolands version does not show thinking at all.