Wrecking The Curve: An Analysis Of The Teach/Tell Connundrum

December 21, 2010

In my last post on Portal, I explored the idea that Portal, in matching its gameplay story and narrative story so well, has given us a great idea of what is required for a truly amazing game without giving us any real ideas on how to do it without either feeling derivative or constraining ourselves to a limited narrative pool.

Today I want to take this a bit further and talk a bit about curves, both in terms of difficulty and narrative. What follows is a description of what I call the ‘teach/tell connundrum’, and some thoughts of how we might deal with it.

A difficulty curve, as you may know, is a curve that maps the difficulty of a particular game against time (or sometimes levels). The curves are almost always theoretical in nature, and we look at them as a way of designing games so as to continue to be challenging, interesting and fun.

In general, a difficulty curve should be exponential on the large scale, whilst containing small-scale stages to allow a new skill to be learnt, utilised, challenged and then enjoyed (see Figure below). Failing to follow this sort of pattern often leads to disorientation, frustration or boredom as the player either finds the game far too difficult or far too easy to complete. There’s a whole array of literature on how to design difficulty curves, and a large amount of theory of exactly what constitutes a great curve. The rules I’ve stated above are by no means hard and fast, and you’ll find that different curves have different advantages and disadvantages that suit certain audiences or games.

The important point is that when designing an experience, you’ll probably have a difficulty curve in mind based on gameplay factors such as how many modes of gameplay there are and how many times you think the player should die. Modern story-heavy games are usually designed to be finished without too much hardship, and are thus often designed as described above (which I’ll call the ‘story-driven difficulty curve’), while arcade games will often have steeply linear difficulty curves to challenge and encourage repeat plays.

While this is a great thing for gameplay design, it’s a big problem for narrative. As mentioned in my post on Portal, the synthesis of gameplay and narrative is a vital aspect in making a game transcend either, and as such, the difficulty curve is going to have an increasing impact as games begin to utilise these aspects better. For the difficulty curve dictates the amount of tension in the gameplay, which should be reflected in the story that the game is telling.

Suddenly, we find that the narrative arc for a game is being shoehorned into a pattern that is both predictable and fairly detailed, considering that a game usually has a narrative complexity comparable to a movie. The pattern has a large number of mini-arcs contained within, each with their own establishment, climax and denouement. And as each of these mini-arcs is rather devoid of story-stuff, they will probably be rather predictable.

This is the basis of the ‘teach/tell conundrum’, which in many ways is only a conundrum if we use the prevalent sparsely-told model of game-based storytelling. For the narrative arc generated by the story-driven difficulty curve is the rather familiar arc of a TV show’s season. Each mini-arc is an episode, which often contains its own story that may or may not contribute to the overall plotline (depending on how serialised the show or game is). The issue comes when each mini-arc is not adequately linked to linked with narrative or packed with enough narrative to make it interesting in its own right. At this point, the game becomes driven by the gameplay alone, lacking that synthesis that can be so powerful.

That’s not to say that games aren’t doing this already. Bioshock was so critically acclaimed in part because each of its areas contained its own individual and rather compelling story that could be explored by the player, which built to a climax as the player completed it. Conversely, Assassin’s Creed (1 – sadly I’ve had a pretty awful year for playing games and am very far behind) built in some story into each assassination, but lacked individual story density for the climax to have narrative resonance.

In addition, the curves can be taken as an opportunity to create interesting synergies: altering the curve can easily create frustration, boredom, or any other number of emotions in the player which can be utilised to help tell the story (this assumes you don’t overdo it to the point where they stop playing). The conundrum exists only when you don’t want to use the gameplay structure to tell story. In reality, the risk of losing your audience will often be too great to make this an option in most mainstream games.

In the end, the difficulty curve is a fact of the medium: a somewhat malleable constraint within which games must adhere to allow for interesting interactive experiences. The detail involved in such a curve requires a similar boost in “story stuff” density as games develop into the future.