Poker Kingdoms 2.0


  • Tiger Hat

    So I’m working on a 2.0 for Poker Kingdoms, the 1.0 never really did commercially (go vote for the Greenlight: http://steamcommunity.com/sharedfiles/filedetails/?id=406518503 ) but I think there’s a really solid idea there and yearn to make it something bigger.

    The premise of the game, is a blending of Match-3 games and Poker, tiles fall on the board, and you have to try and build the best Poker hand you can with them. The better the hand, the higher the score, tiles can be dealt to the board flipped, meaning the player can’t see the value, but your score is multiplied based on how many flipped cards you select.

    I’ve rebuilt the game from the ground up, using test driven development - Poker is a pretty large problem-set, especially when you factor in things like flipped cards, and jokers, TDD was the only way to go. I have a suite of several hundred tests, that is growing by the day.

    The game design also follows the Web - Focused MVC design, as much as is practical in a game. I have forced this on myself in Unity by neglecting to include the UnityEngine in any logic class. There is a singleton that controls the game, which spawns all the objects it needs, including unity gameobjects, these call back to logic classes, but logic classes cannot talk back to gameobjects. This means that if I play an animation, the logic unit needs to be told to wait, and to stop waiting by the gameobject. Not using GameObjects everywhere means my logic units are very lightweight, and I can spawn thousands of them without GC rage.

    This has allowed me to build an AI system that can play the game on it’s own https://youtu.be/gBmmZRB_Hkk

    There is some inspiration from @geoffb in the AI, as it generates a heuristic value for each hand it selects, the video shows it playing with a blank slate, but I can make it favor certain hands, suits, or face values - also, not shown, the AI can deal with flipped cards - and carries a risk taking stat. The detached design means that the AI is not aware of what flipped cards are, not by simply ignoring them, but because they simply are not available to it, so it can’t cheat, even if I wanted it to.

    The final goal is to allow a campaign mode, where you have to defeat a selection of play-styles, rather than just different boards.


  • LDG

    Upvote from me :) Hope it makes it onto Steam!


  • Tiger Hat

    This looks pretty cool–I’m downloading it on my phone now!

    Your Google Play announcement intrigued me though–you mention that the play-store was not a place where what you wanted to do was going to happen (paraphrased). What did you mean by that? This sort of game seems like a natural fit for phones/tablets, no? At least personally, this seems like the type of game I’d much prefer to play on a mobile device.

    Either way, hope it makes it through Greenlight!


  • Tiger Hat

    Yeah, it’s a perfect fit for mobile and I’d love to make a mobile version.

    The play store, is a mess though, and unless you fluke it, you’re competing with a mess of spam-voted clones. In retrospect, Poker Kingdoms was a loser from the outset - getting lost in the mass of Poker titles and Match 3 titles - despite being quite unique (Wildtangent has released a clone) it’s swallowed up and there’s no discovery.

    I literally have one active user, in Iran.

    My plan of action now is to build up a player base on Steam, and convert that to Mobile users.


  • Tiger Hat

    Ah, I hadn’t considered the discovery aspect of it. Makes total sense.


Log in to reply