AntaresTrader is the the game I want to play, but have never actually found. I've been writing it in some form sense I was in elementary school and learned that computers could play games. As I have gotten more sophisticated, I have learned a lot more about why people like to play games and how to design them. This post lays the ground work for a discussion of what makes a good game.

The Game Loop

Before we get to far, we need to talk about a basic mechanism of every game, the game loop. Whether they are called turns, rounds, innings, or frames, every game worth playing has a repeating series of steps. Many games have nested sets of loops.

Lets take for an example, a typical FPS. The smallest loop is the frame. About 30 times a second (we hope) the computer gets input form the uses mouse and keyboard, changes the state of the game (player location, weapons fire, AI, etc...) then redraws a new image that represents players view. This loop happens fast, but there are also larger loops. Each level is a variation on the theme of start somewhere, move through the level, kill the boss, advance to the next level.

Board games are often much more strait forward in their use of the game loop. Each turn a player examines the state of the board, possibly generates a random modifier, alters the state according to the rules of the game, then allows the other players to do the same.

The Game State

State is a word used in programming theory a lot, but it applies just as well to board and card games. The state is simply a description of the current condition of everything that effects play. In Chess the state is the location of every piece, whose turn it is to move, and possibly the time left on the game clock. In Candy Land the state is the location of the players tokens, the color designated by the spinner, and whose turn it is. In draw poke the state is the cards in all the hands and the amount in the pot.

There are two important things to note about the state of a game. First, how much information is in it. Because of the limits of human memory and physical space, board games tend to have a rather small state. Small does not mean simple. In chess the state can be described in a series of moves (e4 e5, Nf3 Nc6, c3 etc...) but the variations are staggeringly large and subtle. Computers on the other hand can hold a tremendous amount of information. Electronic Games tend to have immense states encompassing many many factors. From out FPS, the state might include the amount of ammo, health, strength, and armament of the character, the location of the character, the direction they are facing, plus the same information for 100 of baddies and the location of all the power-ups still in the level. Bigger states tend to make a more lush environment to play in, but do not necessary lead to a more engaging game as the chess example shows.

The other thing about state the varies by game is how much of it the player knows. In chess for example every player knows the whole state of the board, while the entire idea of poker is to control knowledge about the state of the cards in your hand while discerning the otherwise hidden state of your opponents cards.

An extension of this concept is the Game Space (or topography). This represents all the possible states of a game. A varied topography increase interest in a game because players will seek to explore states that they have not been in. To spite the rather small state of chess, the Game Space is large, and unlike many other board games different states present very different outcomes. A game can turn on the ability of a pawn to hold the center files, while in a game like Risk the difference of one army more or less does not usually have a significant effect on game play. Far too many video games today have huge game states, but a very flat unvarying topography leading to the feeling "Shot one monster, shot them all."

Determinism vs. Randomness

Have you ever noticed that some games get very boring once you learn them because they always play out the same way. Tic-tac-toe is one of those games. A reasonably intelligent player will discover that a certain paten of play leads to certain victory and that to avoid loosing the player mus block that pattern. In doing so each player can with force the other to a draw.

Games like this are said to be solvable and once a player has solved the game it is no longer interesting to play (I'll explain why in a bit).

One way to avoid this is to introduce some element of chance into the game. The toss of the dice or spin of a wheel prevent the game from becoming predictable. The basis of gambling is betting on the outcome of this randomness, and as the success of the so-called "Gaming Industry" has shown, it can be very engaging. On the other hand if you remove the large stakes from games of chance they become almost as dull as tic-tac-toe. Many games for young children have chance as a huge element in the game to percent it from being solved, while still keeping it easy enough for their target audience to play. Unfortunately, for the poor parent, uncle, or older cosine who is roped in to an afternoon of Candy Land with a five-year-old, their is not really any fun in being a piece mover and wheel spinner in a game the plays its self. Choice is a vital element of engaging game play as well.

The trick is to balance determinism and chance in such a way that the player feels like their play is effecting the outcome of the game. Having other players, some element of chance that the player must overcome, or a game too complex to be easily solved all work in this regard. Each of these thing adds incentive to keep exploring the game space.

Mapping Play to Plot

Games have a narrative. They connect the abstract manipulation of their state to some story about why the manipulations work. Chess pieces are shaped in the form of medieval war elements for a reason. The game does not need the piece on c1 to be shaped like a horse and called a knight for the game mechanics to work. It only need specify the rules of movement, but the players need that association to enjoy and even to remember the game.

Role Playing Games take this to its logical conclusion. The goal is to immerse the player into a story. While some systems develop overly complicated rules to go with this play, the best table top RPGs are the ones that eschew game mechanics for a well told story.

Even stodgy mathematicians map their games onto a plot. check out the following:

Two agents each choose true or false, if both choose true there is a small reward, if one chooses true and the other false the one who chose false is reward greatly while the one who chose true is punished greatly. if both choose false each is punished greatly.

Now with the usual plot from the prisoners delima:

Two people are arrested by the cops. Each is given the opportunity to rat out the other. If neither talks both will serve 30 days in jail, if one squeals and the other doesn't the nark gets to walk while the sucker spends 25-to-life in the Big House. If they both role, they will both spent 15 years in the same cell contemplating what they did wrong.

Our minds work with stories and setting, so even when a game is being studied in a formal way, having some plot to go with it is important. When we are supposedly enjoying the game, it is essential that it have at least a hint of a plot. even if it is as simple as "You are here to kill all the ugly monsters using cool looking weapons. Go!"

Rewards, Consequences, Choices and Goals

This will probably get a separate post, but for completeness sake lets talk about what makse a game fun and engaging.

First there has to be some form or reward system. Winning a match can be one, but so can getting items, completing tasks, leveling up, or watching a hell spawned monster explode into bloody chucks. Ideally there should be a reward about every minute, with larger rewards after longer periods. Games like chess that lack short period rewards feel slow particularly to new players. Games that lack sufficient intermediate rewards feel like they are a grind (WoW, we're looking at you). Games with poor long term rewards lack depth and purpose. Children's games are not attuned to these rewaeds because children are not oriented towards long term goals. But a game for intelligent adults needs these long term rewards to not feel childish.

Writers will tell you that in order for something to have meaning (that is be rewarding) it mush have some conflict around it. "The boy shoots the Rabid yellow dog" does not pay off unless the whole movie that came before it is present. In games, where the player drives the plot, this translates into consequences. Their need to be some possibility of failure for success to have meaning. Games like Nethack, where a players death is final, take this to the extreme. To spite years of occasional playing, you author has never even met the wizard of Yendor. Other games there is not enough. I have played far to may FPS games where as a first time player I played for 2 hours and was never in any serous danger of getting killed.

There also has to be choice in a game. Part of the reward is in the agency to receive it. Choice is all about how many different ways a player can manipulate the game state and how varied the rewards are for doing so. If every choice leads inevitability to the same or very similar states the game is boring. On the other hand if a choice made early limits the rewards far down the road the game is instead frustrating. There must be a balance in choice reward and consequence. As an aside, a game hat the player has solved, no longer presents choice or risk so therefore cannot present reward either. A game whose outcome is entirely or mostly determined by random chance must substitute large tangible rewards (like cash) for so make up for the lack of choice and agency. (No, pulling the handle on the slot machine is not agency)

Finally there need to be goals. This is the slippery part of game design. A clear set of winning conditions make clear goals, but when the ability to meet those conditions exceed the reward for them then the game is boring and the player stops playing. I can successfully complete a game of free cell in about 5 minutes, and can do so with about 95% consistency. The feeling of accomplishment that I used to get after working a game out has largely disappeared, and so I don't play free cell any more.

Conclusion

Now with some of the basics out of the way, I will state to get into some of the design choice in AntaresTrader. I will also refer back to this post in what I hope will be some forthcoming game reviews.