Texas Hold'em

[Darryl]

In a previous thread there was some discussion on holding a Texas Holdem contest vs blackjack and some 3rd choice...

Anyway, I was a Texas Holdem advocate and since I already have a poker framework for 5 card draw/stud, I figured I'd convert it over to texas Hold'em, but I have a question.
Built into the framework is the ability to evaluate the hands to determine the poker ranking ie. flush, straight etc...

The question is should I provide access to the evaluation functions to the contestants or since this is a learning experience as much as a contest, should I force the player to evaluate their own hands?

What's your opinions on this?

[Sang-drax]

Players should write their own evaluation functions, but anything that makes more people enter is perhaps a good thing. The focus will be tactics anyway.

[LuckY]

I agree with Mister Drax. A part of the assignment should be to successfully determine the state of a hand. If a contestant's code is unable to perform this accurately, it will tend to lose against better written submissions. This sounds to be a very desirable requirement of such a contest.

[kryptkat]

card game rules please. <learned that lesson the hard way meow>

[Jherry]

this should explain the rules of the game pretty well. You can count me in, just tell us some of the specifics (format of the cards, how we look at the widow cards, functions for betting, etc.). I think it would make it more of a challenge to make us evaluate our own hands, but If you want more players, then start out with us being allowed to use them, and if turnout gets better, you can change it to evaluating on our own.
-cheers-

[Darryl]

hmm, I have been debating both, on one hand, it would be a good excercise for the programmer, but on the other hand, the contest is really about writing the better texas holdem algorithm. Additionally, if I release the framework for them to test with, I won't be able to hide the function anyway and it's a required part of the framework to evaluate the hands. So I won't give access to it, but for those that want to copying/modifying my function to fit their needs won't be hard.

@Jherry, sorry I am not familiar with the term widow cards, but anyway, I will give you the details tomorrow. I will release the player class that the contestants will have to overide...that should be enough to get everyone started.

[Darryl]

Ok as promised, here is a preliminary player class, subject to change, although I don't speculate it will.

I also included the card class, as the player class depends on it.
Basically, you will take the file Your_Player.h and fill in the blanks. Rename your_player.h and the class to something you find more suitable (try to be unique)

The file player.h has been commented to explain how each function works.

The framework hopefully will be finished this weekend for you to be able to test your players with. I'll include a practice bot and a console/human player as well.

I am hoping to get at least 8 participants

[Jherry]

just a quick question, are you going to be logging the game for us? If not, can I add a function to do so in actionNotify, and have you send me the log? How many chips will we start out with? Also, the widow is just the community cards.

[Darryl]

just a quick question, are you going to be logging the game for us? If not, can I add a function to do so in actionNotify, and have you send me the log? How many chips will we start out with? Also, the widow is just the community cards.

The game will be logged and provided. I was thinking 10000 chips to start with

[kryptkat]

ok thank you.

I was thinking 1000000 chips to start with

i think i get the idea.

[misplaced]

is there an original post to this contest? when's the deadline? how often will the blinds raise? will there be antes or just the blinds? where in the code can you get your (table) position and the amount of players in the hand who have yet to act?

if i have time i'll get in. i say you hold a monthly tournament too. i think it would be kind of cool to have our players get more experienced over time. in other words, everyone can tweak their code to make their players better each month. a program like this could get very large and complex. it would be cool to go all out on it, but if everyone did, everyone would lose interest in the time it takes to do something like that. if we had monthly tournaments, i think we could keep up the interest and have adequete time to write some real good code.