?

Log in

No account? Create an account
multidimensional tic tac toe's Journal
 
[Most Recent Entries] [Calendar View] [Friends]

Below are the 8 most recent journal entries recorded in multidimensional tic tac toe's LiveJournal:

Tuesday, May 12th, 2009
11:59 pm
[paedimonic_lim]
for ...



new aisusio./new pies...sulfuoxi ..ottop i.b irl.
http://www.myspace.com/tobeobuiousentomologie0o
Monday, August 13th, 2001
8:32 pm
[price]
A few points to note...
This post is largely based on hach's past posts, and one or two IM convos we've had.

The functionality of the game is eventually going to get to the point where we have AI opponents which can play with the user. This being so, all functionality (as far as rule checking and such) needs to be on the client side. Server smart, client dumb really does not apply in this case.
Protocol for multiplayer sessions should likely be TCP/IP, if only for integrated chat client functionality.
I also think that multiplayer games should be hosted through servers, as opposed to direct connects between players. This prevents cheating, and helps people find new opponents.
The GUI, I could care less about at this point. I think it would be interesting to play with 3d and stereoscopic representations of playing fields, but 2d obviously has some benefits. GUIs are always rough, though, as you have to make things apparent to the novice, but non cumbersomely simplified for more experienced users.

I plan to start hacking a shell for the client. If anyone with art skills could draw up potential GUI looks, that would be extremely cool and useful. Hach, you were also mentioning that certain games just arent playable. is there a general rule/algorithm for determining if game G is unplayable? If we dont know, any ideas on how to find out?
Sunday, August 5th, 2001
11:44 pm
[hachibambatar]
guidelines
ok, there are three things to talk about here :

gameplay

ai

ntic programs

or anything else even vaguely related for that matter... POST!!!!

Current Mood: mellow
11:44 pm
[hachibambatar]
i'm off to elsewhere
i'm learning about all manner of new maths and gradually developing extremely proper style for c. i'm off for vacation in yellowstone for a bit, perhaps i'll come back armed with some revelation or other, but i wouldn't cross your fingers... maybe though, i'll write a neural net to play 3d 3 player while i'm there though... i think i would just go 2 layer, the input layer accepting 54 inputs, corresponding to a 2 bit state for each square. (player 1, player 2, player 3, nobody)... following this, because i have a whole lot of computing power at my disposal (my dad is a scientist) i will have two 500 node layers... that number may change, then, an output layer which will simply describe where it moves. now my trouble is coding in the rules of the game... lets say early in training, it makes an illegal move... how do i tell it not to do that? indeed, what exactly am i training it to do? also, symmetry... there are several equivalent representations of a game of ntic... any good strategy would obviously be invariant under shifts between representations... there is something to the symmetries, and that should allow us to very significantly reduce the training set.

Current Mood: mellow
Saturday, July 28th, 2001
2:31 am
[hachibambatar]
code!!!
alright, so my plan is to write a c++ backend for a multitic system on the server... here's the deal though, i've /never/ written network or gui code before... if i have to, i will, but please don't make me, that or point me to a nice sockets library...

the storage for a game is as follows, much as the coordinates would be written as a list of numbers between zero and two, the game is stored as a list of moves.

here's the class, things will be simpler that way, hopefully...


class Game {

public :

// Constructors, destructors and the like...
Game(); // default constructor
Game( string name, int ndimensions, int nplayers, int gametype ); // this should make some sense, gametype is a flag variable...
~Game(); // the destructor is going to do some database stuff if the gametype happens to be rated...

// Setup
void addplayer(int playerid); // i don't quite know what the implementation of playerids will be, except that they will be 32 bit ints...
void removeplayer(int playerid); // not everything needs a comment...

// Gameplay related
void move(char *pos); // this function is nothing more than a malloc and a memcpy with some error checking so that people don't make illegal moves...
void undo(); // guess what this does...
int iswon(); // is the game won?
int iscatsgame(); // is it a cats game?

// File related
int save(string filename) const; // save the game to a file on the server
int load(string filename); // load up a game from a file on the server

private :

int gametype;
int ndimensions;
int nplayers;
int *playerids;
int nmoves;
int counter = 0;
char *history;

}

post questions about this, anytime, or if you have a better model for the program or anything to say, do go ahead and say so... cvs, bah...

Current Mood: awake
Wednesday, July 25th, 2001
2:15 am
[hachibambatar]
lattices
anybody here ever done anything with high dimensional lattices?
Tuesday, July 24th, 2001
12:19 pm
[hachibambatar]
further generalisations
in my experience, tic tac toe on 4x4 boards, row of 4 to win is not a playable game and in fact always ends with cat.

here is the most general definition of tic tac toe i can muster, which also includes connect 4 without gravity...

n dimensions, p players, l cells to an edge, w cells in a row or diagonal to win.

perhaps there is a topological method of looking at multidimensional tic tac toe. in n dimensions, how many cells border a given cell? (including along diagonals) obviously this is different on edges, different on walls, different in corners, and different in high dimensional analogues of such things.

i suggest using a bit of the language of go here, and refer to a cell's bordering cells as liberties.

clearly on a large board (length, not necessarily dimensions), the first player will always win if only 3 in a row are required to win.

i see you monitoring this, perhaps you should post...

Current Mood: blah
2:20 am
[hachibambatar]
hi kids
welcome to the multidimensional tic tac toe livejournal community. items to be discussed here:


  • High dimensional tic tac toe
  • Fair versions of the game
  • AI related to high dimensional tic tac toe with large numbers of players
  • General AI
  • A possible web interface for n-D tic tac toe.
  • Representations of high dimensional geometric objects
  • Group theory related to tic tac toe
  • Really anything that leaps to mind


Current Mood: accomplished
About LiveJournal.com