Thread: Sudoku Contest Invitiation

  1. #16
    Cheesy Poofs! PJYelton's Avatar
    Join Date
    Sep 2002
    Location
    Boulder
    Posts
    1,728
    I'm not necessarily against shortest code/obfuscation contests, I'm just bitter that that is all I can find. I think its great as a novelty and certainly people seem to enjoy them, it just annoys me sometimes that the only contests I can find all force such bad habits and ugly code instead of challenging the more problem solving and algorithmic abilities that would be more beneficial in the real world.

    Granted this is all subjective, plenty of people find obfuscated code beautiful and there is a level of problem solving trying to whittle down your code. Also I'm sure somebody can make a case for how this teaches you beneficial real world skills. And obviously its MUCH easier to grade these contests than anything I'm looking for.

    But really I'm just ranting, I'm not suggesting you change anything about how you run your contests. Give the people what they want! BUT if you want to give it a try, I'd suggest pulling some problems from either TopCoder or the UVA contest site. Most can be written quickly if you can solve it and by changing the word problem a little bit nearly impossible to google a direct answer for. Of course you'd have to subjectively grade the algorithms given to you.

  2. #17
    Tropical Coder Darryl's Avatar
    Join Date
    Mar 2005
    Location
    Cayman Islands
    Posts
    503
    I agree with a lot of your points and for the record I don't do obfuscate contest for the reason you suggest. Code should be readable. In the brevity contest, we encourage, good naming of variable and comments so as you said, it's arguable if it is providing real world experience or not. Yea, maybe some of the topcoder stuff would be good...especially if you allow contestants to challange others code. As for you, there is topcoder as you ment as well as POTM. Personally, I like to do AI contest (as participant, not host) , but those are really rare. I keep meaning to put together something for the Loebner Prize, but been too lazy.

    Anyway, i am thinking of hosting a Monopoly AI contest...would you be interested?

  3. #18
    S Sang-drax's Avatar
    Join Date
    May 2002
    Location
    Göteborg, Sweden
    Posts
    2,072
    Quote Originally Posted by Darryl
    Anyway, i am thinking of hosting a Monopoly AI contest...would you be interested?
    I'd be interested and I would probably submit, but IMO there are better games to use in an AI contest. Monoploy has lots of special rules and few decisions.

    A card game AI contest would be fun.
    Last edited by Sang-drax : Tomorrow at 02:21 AM. Reason: Time travelling

  4. #19
    Cheesy Poofs! PJYelton's Avatar
    Join Date
    Sep 2002
    Location
    Boulder
    Posts
    1,728
    I'd probably submit, although I agree with Sang-drax.

  5. #20
    Registered User samGwilliam's Avatar
    Join Date
    Feb 2002
    Location
    Newport
    Posts
    382
    How about Rummikub? Coding AI for that would be interesting!
    Current Setup: Win 10 with Code::Blocks 17.12 (GNU GCC)

  6. #21
    Tropical Coder Darryl's Avatar
    Join Date
    Mar 2005
    Location
    Cayman Islands
    Posts
    503
    Quote Originally Posted by Sang-drax
    I'd be interested and I would probably submit, but IMO there are better games to use in an AI contest. Monoploy has lots of special rules and few decisions.

    A card game AI contest would be fun.
    Few decision = easy to program, greater participation.

    I am looking for the type of decsion making that a computer can't easily do and not just because of high combinations like chess, but because of uncertainty of outcome. Monopoly offers this in way of decision like: do I buy a property or save my money for something else. Do I trade? When should I buy houses? hotels?

    Personally I like board games over card games because most card games have a "correct play" strategy that is easily programmed. Although, I think hearts or spades might be interesting.

  7. #22
    S Sang-drax's Avatar
    Join Date
    May 2002
    Location
    Göteborg, Sweden
    Posts
    2,072
    Quote Originally Posted by Darryl
    Few decision = easy to program, greater participation.
    Monopoly would not be easy to program. Think of the complexity of the interface: The players need to read a lot of information: Who owns what, what places are built upon, where are the other players, how much cash have they got?

    There also have to be a way of reading what the other players are doing, what cards they draw etc.

    And then there are the special rules, like the jail immuity cards, which all have to be handled separately.

    Even if you create a very good interface, it would take time to learn it and to create your program. This is a lot of time invested for quite few decisions in a game mostly controlled by luck.
    Last edited by Sang-drax : Tomorrow at 02:21 AM. Reason: Time travelling

  8. #23
    S Sang-drax's Avatar
    Join Date
    May 2002
    Location
    Göteborg, Sweden
    Posts
    2,072
    Quote Originally Posted by Darryl
    most card games have a "correct play" strategy that is easily programmed.
    A Texas Hold'em contest would not have this weakness. There are many more examples of possible card games.
    Last edited by Sang-drax : Tomorrow at 02:21 AM. Reason: Time travelling

  9. #24
    Cheesy Poofs! PJYelton's Avatar
    Join Date
    Sep 2002
    Location
    Boulder
    Posts
    1,728
    Simple game that involves a lot of chance but still requires a decent AI to win with is backgammon.

    Also, we've talked about blackjack competition that involves a variety of bet amounts and the object is to be the one with the most money amongst your competitors. Takes more AI than simply writing a card counting routine.

    What about Scrabble?

  10. #25
    Cat without Hat CornedBee's Avatar
    Join Date
    Apr 2003
    Posts
    8,895
    Scrabble sounds like a brute force method is possible, once you've got a dictionary filled into a smart data structure.
    All the buzzt!
    CornedBee

    "There is not now, nor has there ever been, nor will there ever be, any programming language in which it is the least bit difficult to write bad code."
    - Flon's Law

  11. #26
    Magically delicious LuckY's Avatar
    Join Date
    Oct 2001
    Posts
    856
    Scrabble. Now there's a fun challenge. In addition to forming words (obviously, they'd come from a dictionary of valid terms), you need to use as many letters as possible, attempt to acquire as many points per play as possible, attempt to place on "bonus" squares, and search the existing board for the different places with which you can place which different words.

  12. #27
    Cheesy Poofs! PJYelton's Avatar
    Join Date
    Sep 2002
    Location
    Boulder
    Posts
    1,728
    True, a brute force would probably be good, but you can put time restraints like say one or two seconds max so the contestants would need to optimize their AI. Plus a good scrabble AI would also need to play defense so as not to allow the other player optimum scoring chances.

  13. #28
    Tropical Coder Darryl's Avatar
    Join Date
    Mar 2005
    Location
    Cayman Islands
    Posts
    503
    Monopoly would not be easy to program. Think of the complexity of the interface: The players need to read a lot of information: Who owns what, what places are built upon, where are the other players, how much cash have they got?
    Yes.. I was really speaking of the decision making, but yes you are right, Monopoly is quite involved, interface-wise. So back to the drawing board... though I am thinking about dominoes now.

    As far as cards and texas holdem goes let me elaborate on my original statement. There is no strategy to playing the game, there are only strategies for betting on the game to maximize/minimize wins and loses. I am looking for games where the contestants have more control of the games outcome. However, Holdem might be doable, I even thought of how a computer program might have "tells" but won't reveal just yet just in case I do decide on a holdem contest.

  14. #29
    60% Braindead
    Join Date
    Dec 2005
    Posts
    379
    What about risk? Now theirs a challenge :P.
    Code:
    Error W8057 C:\\Life.cpp: Invalid number of arguments in function run(Brain *)

  15. #30
    Tropical Coder Darryl's Avatar
    Join Date
    Mar 2005
    Location
    Cayman Islands
    Posts
    503
    Quote Originally Posted by Blackroot
    What about risk? Now theirs a challenge :P.
    I agree and it's one of my half finished projects I have been working on for ages now....I havn't found the idea way to represent the board, but I have a few ideas i might try. Also, the AI for Risk isn't trivial, and so there's the risk of no participation, which I don't want to put in the effort if no one will participate, however, I think the game has enough appeal to still attract a few.

    As far as the AI goes, the thing I really like about Risk is that the player decides when their turn is over and there's a fine line between being over aggressive and too passive that can be hard to balance between. It's also makes it difficult to calculate your opponents moves.

Popular pages Recent additions subscribe to a feed

Similar Threads

  1. Expression Evaluator Contest
    By Stack Overflow in forum Contests Board
    Replies: 20
    Last Post: 03-29-2005, 10:34 AM
  2. Obfuscated Code Contest
    By Stack Overflow in forum Contests Board
    Replies: 51
    Last Post: 01-21-2005, 04:17 PM
  3. WANTED: Contest Master
    By kermi3 in forum A Brief History of Cprogramming.com
    Replies: 15
    Last Post: 01-23-2003, 10:15 PM