(Not So) Brainless

This is a discussion on (Not So) Brainless within the A Brief History of Cprogramming.com forums, part of the Community Boards category; I'm actually very much into Zero-sum AI stuff currently. I have a chess AI too if you want to give ...

  1. #1
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,183

    Lightbulb (Not So) Brainless

    I'm actually very much into Zero-sum AI stuff currently. I have a chess AI too if you want to give it a try . It's rather strong, though (about the level of a FIDE Master).

    http://cyberfish.wecheer.com/Brainless
    Last edited by laserlight; 06-05-2008 at 11:44 AM. Reason: Moved part of the post to the original thread.

  2. #2
    Registered User
    Join Date
    Sep 2006
    Posts
    8,868
    Quote Originally Posted by cyberfish View Post
    I just coded it up in 20 mins... bugs might have creeped in, but otherwise the AI is unbeatable. It can already prove that the game is a draw with perfect play by both sides. (analysis from the initial position)

    I'm actually very much into Zero-sum AI stuff currently. I have a chess AI too if you want to give it a try . It's rather strong, though (about the level of a FIDE Master).

    http://cyberfish.wecheer.com/Brainless
    You're a wonderful programmer. I know it's unbeatable if there are no bugs. I've been a big fan of computer chess since Northwestern was world champs.

    I like chess, but it's very demanding of your time to be good at it, and that time is almost all completely sedentary-->sitting, sitting, sitting! It's wonderful for your mind, but hell on your muscles.

    Except for opening traps, I haven't beat a good chess program since Sargon II, I believe!
    The amount of look-ahead with today's hardware makes the programs *very* tough to beat, even for the best human players.

    They're probably working on the 8 piece end game tables, (Nalimov's), now? Amazing stuff! Thanks for the url, I'll have to give it a go.

  3. #3
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,183
    6 pieces Nalimov IIRC. 8 pieces would be several EB =).

    I am very bad at chess, too (~1300 on FICS). My program is ~2200, though =). It lost its only match so far with a FIDE Master (on FICS).

    I personally can't even take an engine to the end of its opening book... so opening traps don't work for me =).

  4. #4
    Registered User
    Join Date
    Sep 2006
    Posts
    8,868
    6 pieces are done right, they were working on 7 a year or two ago. Yeah, it's an unbelievably big job, even with Nalimov's brilliant code.

    I couldn't take them to the end of their book either, without special prep. Now, probably not even then.

    Several of those FIDE Masters are very tough, even for computers. I'm sure as your program matures, you'll start holding your own with them, though.

    Does bring back the memories. I long for the day when you could wipe the floor with the little silicon beasties!!

  5. #5
    C++ Witch laserlight's Avatar
    Join Date
    Oct 2003
    Location
    Singapore
    Posts
    22,303
    This thread was moved from Making Tic Tac Toe Smarter.

    It lost its only match so far with a FIDE Master (on FICS).
    Tsk tsk, and you said it was of FIDE master strength. Perhaps candidate master strength would be more believable

    Incidentally, how many games were played in the match, at what time control, and what was the result?
    C + C++ Compiler: MinGW port of GCC
    Version Control System: Bazaar

    Look up a C++ Reference and learn How To Ask Questions The Smart Way

  6. #6
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,183
    Code:
    		rating	 RD	 win	loss    draw	total	best
    Blitz		2036	 38.5	 1613	423	201	2237	2073 (23-Apr-2008)
    Standard	2191	 40.9	 1256	250	107	1613	2228 (06-May-2008)
    Lightning	2060	 114.0	 10	3	0	13
    The FM it played is vor (on FICS), rated 2346. IIRC the match was 2 games at 30 0 or so. So I guess it is a bit weaker than FM.

Popular pages Recent additions subscribe to a feed

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21