This is the letter Delta: Δ. Just in case.
Edit: This entry was originally entitled Story Delta and Mechanics Semantics, but I rather use the term Mechanic Vocabulary instead of Mechanics Semantics. It’s a better way to refer to the idea I was talking about.
If you’ve read the most recent posts of this blog, you should know by now the concept of Story Delta. I mentioned here and here. I will quote myself, because I’m lazy and it makes me feel kind of important:
The designers driven story (or Story-Story, as Valve’s designers call it; Embedded Story as referred by Schreiber) is the usual narrative we identify in games: the plot arcs, characters and dialogues. The players driven story (or Gameplay Story, as Valve’s designers refer to it, Emergent Story as called by Schreiber) is the sequence of actual and specific actions the players take while playing.
Tetris has no designer driven story: there is no narrative whatsoever. This title is what we call a non narrative game. Tetris is totally made of player driven story, you take those little tetris pieces, move and rotate it in specific way until the game finishes somehow. That sequence of actions makes up a story that depends exclusively on the player. On the other hand, a game like Portal has a sequence of actions that conform a plot, with characters and dialogues that are a companion for the player driven story.
The questions that arises are: how different these stories have to be? how much separation between them has to exist, or how much of it it’s permitted?. Valve’s designers talk about the Story Delta, how much the actions directly driven by the player trigger actions in the designers driven story. Their conclusion is that delta has to be as little as possible in order to engage players.
I wan’t to provide an illustration of what story delta is about and how we can implement it gameplay. Also, I want to introduce the concept that I usually call Mechanics Vocabulary.
If you haven’t played the game, spoilers ahead. This is the example:
Well, the actual example are the first two minutes or so.
The first minute and a half is the hook that connects Half-Life 2: Episode One to the previous installment, Half-Life 2. The next minute is the actual beginning of the game, the section we are going to focus on.
The question I want to answer is: Why is Dog (the robot) the one rescuing Freeman?.
I want you think about this question in terms of its alternative: why isn’t Freeman releasing himself from the debris?, we already know the kind of badassery he is capable of, so, why does he need help in this particular situation?.
There are two major answers: the one related to story, the one related to design. Both are, in fact, right; what I want to point out is how related they are and why.
The first answer, the one related to story, is that a catastrophic explosion happened and there is no way Freeman just survived without help. Actually, the Vortigaunts (the alien beings) rescued and protected both Freeman and Alyx. They are the first ones to help. Then, Dog appears and the game begins.
That’s cool. Is not a detailed answer, but I think the point is clear.
But why, from a game design perspective, is Dog the one rescuing Freeman.
As far as I can tell, there are two alternatives to the start sequence showed in the video. One is to, literally, show Freeman emerging from the debris (always in first person view), or performing any action that could help Dog and Alyx to localize and rescue him. The other option is to give the player some kind of challenge that triggers the aforementioned situations.
Both aren’t the most suitable given the context where the game is framed.
Showing Freeman releasing himself increases the story delta. No action performed by the player is driving the actions of the avatar. The designers driven story is going in one direction, but the player driven story is going in another direction (is going nowhere by this point, to be very specific).
The concept behind this is Agency. The capacity of an agent (the player, in terms of games) to act in the world it belongs is what we call agency. The sense of agency is diminished if what suppose to be the avatar (Freeman) is doing things without player intervention. The player isn’t actually acting in the game world, so agency is not being considered.
When the players actions transform the game world, the sense of agency increases. One way to transform the game world is to trigger the events that move forward the story. Then, if the actions performed by the player aren’t triggering the designers driven story, then sense of agency, again, diminishes. That fluctuation (or gradient) of agency that influence the story of the game is what the concept of story delta is trying to measure. Lesser the agency, bigger the delta.
If we want to attach some numeric values to the concept, then the delta is equal 0 when what’s happening on the screen is completely due player performance. Non narrative games are the case at hands (Tetris). The delta is equal 1 when the player is, literally, doing nothing even though there are things happening on the screen. Cinematic scenes are the best examples of this.
In other words, showing Freeman releasing himself in a cinematic way diminishes player’s agency, therefore increasing story delta.
It’s zero the ideal value for the story delta?. The answer is no. A delta equal zero means no narrative at all, and that is not always what we want. The story delta, as I see it, is a function depending on time (gameplay time) that measure the gradient of agency. Therefore, story delta is also related to the interest curve of the game. It’s beyond the scope of this post to explain this relation, but story delta can help us to make the game pacing to look like this. And that’s cool.
So, let’s give some agency to the player. What about making some button smashing sequence where Freeman can release himself as a result of that smashing? or something like it. The thing here is not directly related to story delta, but to Mechanic Vocabulary: what each button actually does in the game.
Let’s focus in one button, the left mouse button. Half-Life is a very classical FPS, the most important action in the game is to shoot at stuff and enemies. Therefore, as the most important action, it has a dedicated button, one that means “to shoot” and have no other meaning. Imagine that the initial sequence in the game is triggered by a “click as fast as possible the left mouse button” mechanic. You are giving to that button a meaning that is not related to its original intention. Freeman has no weapon in that moment, no shooting can be performed. Freeman will push instead of shoot.
Creating that kind of control in a game where no pushing (in the sense described) is needed afterwards is ambiguous, but moreover, it’s clumsy.
In games, as a medium where player intevention is required, the richness of the experience is to learn and master certain kind of controls that embody the core mechanics. In Half-Life, the core mechanic is shooting, not pushing. There are games where pushing things (whatever that means in the context) is the core mechanic. Half-Life is not one of those. Why create a situation where certain mechanic is needed, and then abandon it completely the rest of the game?. There is no learning nor mastering there. The richness of playing is lost.
That analysis applies to any other button in Half-Life’s layout.
Isn’t limiting the scope of what buttons can perform, in fact, limiting the scope of the game? yes and no. Half-Life is a good example. The mechanic is always to shoot at something, what changes is what you are shooting and what the action of shooting has as consequence in the game. The most obvious way to see this is looking at the arsenal: shooting the shotgun is not the same thing that shooting any other fire weapon (or using the crowbar).
But Half-Life goes beyond introducing the Gravity Gun. The mechanic of shooting remains the same, but with the gravity gun it has a lot of new consequences inside the game, from new ways to kill foes (which are challenging, therefore rewarding) to a new universe of physics puzzles. The alien pheromone, the one which lets you control the Antlions, is another way to give new meaning to the same shooting mechanic. Those new meanings can be learned and master in the game, unlike the pushing mechanic mentioned above.
I could extent a lot here, but if you see closely, you can push things (grabbing them) using the gravity gun. It’s particularly smart how this mechanic of moving things rise to the surfaces by changing the meaning of shooting
One thing to notice here is how designers at Valve never forget a particular tool they gave you. You get to the final chapters of the game (Half-Life 2 or any of the episodes) and you always find yourself using the crowbar for meaningful things, mostly breaking boxes to find resources and defending yourself for those flying thingies, so even though the skill is mastered you are constantly using it, they don’t fall into the “create this skill and not using it never again” trap.
Something similar happens with the Rocket Launcher. Have you ever wonder why, unlike other games, the Rocket Launcher in Half-Life 2 has a laser sight to drive the rocket? that’s because otherwise there is no skill to master. You usually launch the rocket in straight line and hope for the best, there is not much to learn and master there; Half-Life 2 gives you the opportunity to refine a skill every time you use the Rocket Launcher. Each situation has to be solved differently and it’s up to the player how to guide the rocket. You can learn it and master it throughout the entire game.
Portal is another example of giving new meaning to a classic mechanic. The shooting is there, but the meaning of it is quite different and it allows for a new whole bunch of actions in the game. Is not the intention of this post to discuss about it, but new actions open the door to new kind of emotions and feelings in the game, without relying heavily in tools outside the gameplay (cinematics, to name one thing).
Going back to the example case, it is not the proper way to design a sequence that increases the story delta, and on the other hand, is not actually smart to create mechanics not aligned with the intentions of the game. So, Dog comes in.
Of course it’s an animation, but not one where the avatar is doing something, so agency isn’t there for a good reason: the avatar can’t move, which is different than moving without player intervention. Dog is never used by the player, so its actions aren’t attached to a particular mechanic, it makes sense it has full freedom to perform any kind of job (applies to Alyx as well). And the agency is never left at zero, the player can move the camera during the sequence; looking at things is a skill that maybe is too basic for a game like Half-Life, but is there for the player to use in that moment. Adding the fact that the sequence is not that long, every decision made aligns properly with the mechanics and the story of the game (remember the first answer, it fits perfectly in this setting).
One thing you can point out is that reasoning is quite intuitive, you don’t need those kind of argumentation at the time of designing, which is somehow true. But remember, we are trying to create a body of knowledge that, to start with, can grow up from the understanding of intuitive choices. That understand will allow us to represent more complex situations.
I choose this particular sequence for the sake of the spoilers, but Half-Life 2 and its sequels are full of these examples, where the design takes into account story delta and mechanic vocabulary. You could apply the same analysis to say, the final sequence in Half-Life 2: Episode 2 and you will get the same result.
Of course, it depends on the game what the story delta should be in general, or how the mechanic vocabulary is laid out. You can use this post to compare to other game and try to understand why the design is created in a particular way.
I think good games, in a very general way, should maintain a fair low overall story delta and try to create a lot of meaning from a limited and small number of mechanics. In that way the agency is high and you can distribute the cycle of learning and mastering in a smooth curve. Both things ensures a gradual distribution of fun that the player will have from start to end.
I can summarize all of this in the quote Good games takes minutes to learn, but years to master. Agency and mechanic vocabulary have a lot to do with that.
Hope this post can help you to better understand how design should bring together story and mechanics.