Thursday 30 March 2017

Videogames - too much fun for their own good?

As a medium videogames have been kidnapped by their easily-achieved engagement. Simple gameplay is so much fun on its own that storytelling has never been needed in order to draw an audience. Compared to films, the element of storytelling is seldom elevated in videogames. Is it time for a walk down that lesser-known path, leading to better narratives in videogames?

When the first films appeared 120 years ago, they were shown under amusement-park-like conditions. By peeping into a Kinetoscope, the audience (one by one) were able to get a short experience of moving pictures. For instance, as in Fred Ott's Sneeze, by W. K. L. Dickson (1894), anyone willing to pay could watch an engineer sneezing.

As you can imagine, these clips felt pretty boring quite quickly, which led to an immense pressure to make moving pictures more interesting. The first step was to find something more fascinating to film than a sneezing engineer. In the late 19th century, a steam engine arriving at the train station fell into this category, as the Lumière brothers proved in L'Arrivée d'un train en gare de La Ciotat (1895).

But there's a limit to how many moving objects people are interested in seeing on film, as well as in ways to top previous experiences. There was a need for something that caught the attention of the audience, something that would make them come back for more. Enter - narration!

The first attempts to tell the audience a story built upon simply filming theatre plays. At the same time, film developed into its own medium, and filmmakers started creating new tricks suitable for film, which increased the audience engagement in the story. This is when a lot of interesting things started happening. For instance, filmmakers realized that not every part of the plot had to be filmed. By simply implying actions and events, the audience would still keep up with the story. Film editing became a crucial part in tinkering with the narrative, and storytelling in film evolved greatly.

Let's compare this to how games work. This is what one of the very first video games (Pong, 1972) looked like:

Contrary to the first attempts at film, this game is still quite fun to play. In fact, there are still new games being made based on Pong's gameplay.

Flag N Frag, by EDEVOX

Obviously this version has a lot of new features and graphical decoration, but it still relies on the same concept as the original game. This differs widely from the evolution of films. No one would consider making a film based on the same concept as Fred Ott's Sneeze. There is just something inherently fun about interaction that makes an old game like Pong still worth playing. This is not just true of the very first games and films; if you compare works like Pac-Man (1980) and The Birth of a Nation (1915), the former stacks up far better.

Dealing with video games it was pretty clear from day one what the interesting thing was - playing is simply fun! The medium itself presented new ways of making gaming even more fun, which were being used in all possible ways. Compare this to a scenario where a similar evolution took place in film production. The film equivalent would be movies still being all about watching small clips [1] and most of the effort during the last 120 years would had been on extending this aspect. In the games industry, this is the progress that basically took place.

Before I continue this post I need to clear up a term. When I talk about "the fun of interaction" I mean a very specific case of gameplay where you have a core mechanic that you base the whole experience around. Examples of this are: shooting spaceships, jumping over chasms, avoiding incoming hazards, leveling your RPG character and so forth. This is the type of gameplay where the aim is to make it feel as "fun" as possible. Even if the graphics are just made up of simplistic shapes, this sort of gameplay remains very engaging. I will henceforth refer to this as "classical gameplay" in order to differentiate it from other forms of gameplay (e.g. exploration, anticipation of monsters, dialog, etc). I also need to make it very clear that I think the future of the videogame medium lies in interaction and play, and we should not strive to remove this - quite the contrary. With that said, let's continue.

From a pure classical gameplay perspective there is nothing wrong with focusing on the fun part of gaming. Most of what we currently see in the videogame world is there thanks to this focus. However, it has held back videogames as a storytelling medium. In movies, it's crucial to get the storytelling aspect right as you really don't have anything else to fall back on. Sure, there're always blockbuster movies that can make the audience overlook so-so storytelling by offering a visual spectacle. However, these are the rare cases. The vast majority of films relies foremost upon having good storytelling, and it's needed even in order to make spectacle work. The narrative, however corny, still has to be front and center. Not so for video games; as long as you get the core gameplay working, the audience will be happy.

Storytelling which is just added as a sort of extra spice has long been the standard in videogames. In fact, most attempts of storytelling often feel like they get in the way of the classical gameplay. Not only is storytelling something that isn't really needed; it can even worsen the experience by acting as an impediment to classical gameplay. From this perspective, it's no wonder storytelling has had a tough time to progress in videogames. For a long time, it hasn't really felt needed and has been seen more as a hindrance than an opportunity.

This sort of thinking still permeates game development. "Make sure your core [classical] gameplay gets done first" is one of the most basic pieces of advice given to any aspiring game developer. And once you get that basic classical gameplay working, only then should you try and make your story fit into it - if you need a story at all, that is. It's important to note that classical gameplay-wise this makes a lot of sense and is a fundamentally good thing. If you're aiming to make your games as much fun as possible to play, getting your gameplay loop working first is a wise move. It also makes sense from a commercial perspective, since classical gameplay is the easiest way to get the audience's attention. Narrative-wise though, this is far from an optimal strategy.

Historically there have been two game genres of note that have resisted this trend: adventure games and horror games.

Adventure games share the same issue as movies: the core gameplay isn't that much fun. The players basically (through text or using a mouse-based interface) give commands that a character might carry out for them. And unless there's some sort of greater context involved, this gets boring quite quickly. One way of fixing this is to put more effort into the storytelling. When the character you are indirectly controlling is part of an engaging narrative, it becomes a lot more fun to control them.

Despite this, storytelling-wise, adventure games stopped evolving quite quickly [2]. There are a bunch of reasons for this, some of which I outlined here. Another especially important reason is that almost all adventure games revolve around puzzle solving. They haven't really given up on their game legacy. The player can always go into "I am just doing this for the puzzles"-mode, and thereby avoid much of the game's attempts to tell a good story. So we're back to the initial problem - classical gameplay standing in the way of progress in narrative.

Horror games take on this issue from a different angle. This is one of the few (possibly only) bigger genres where classical gameplay turns into a nuisance. The most basic example of this is: if monsters are too much fun to encounter, they stop being scary. So horror games have been forced to tone down on one of the core engagements that has been a cornerstone of many other types of videogame. By giving up on the most fun part of the medium, the genre had to turn to something else in order to keep up the engagement level - storytelling. Many horror games - Silent Hill is a great example - feature clunky combat, and much of the time it is more stressful than fun to encounter enemies. But by offering a story that ties into the player's actions, you can take something that is not so much fun on it's own and turn it into an very engaging experience.

To convey horror by purely system-based means is hard, and therefore a narrative is crucial in order to provide the right experience. However, crafting these sorts of experiences is also hard, especially if the storytelling is supposed to carry the heaviest burden. As a way of making up for this, instead of putting more focus on the narrative aspects, horror games have always added all sorts of other systems to provide a basic engagement loop. In the end, this is what made the golden age of the PS2-era horror games come to an end. When the genre started to stagnate, Resident Evil 4 came about, putting all focus on gameplay and becoming a huge success.

Resident Evil 4 is an amazing game on its own, but it really did a disservice to the horror genre as a whole. Just like we have seen in the past with classical gameplay being the cornerstone to fall back on, the horror genre ended up doing the same. And with it much of its narrative-based ambitions never got a chance to properly evolve.

What this all leads me to is the following: When it comes to storytelling, games are inherently just too much fun for their own good. I think the problem comes down to being stuck at a local maximum.

What I mean with this is that as you are developing a game, you will come across a bunch of ideas that you can choose to follow. There will always be a lot of tension between getting the game's gameplay and storytelling to work. Scouting the territory of the possible design choices, the ones where the gameplay wins are the ones that will almost always come out victorious. Following the path of narrative-focus will almost always decrease the perceived engagement. Think of these gameplay-focused solutions as going upwards towards a peak, and the story-focused ones as going downhill into a valley.

But that doesn't mean that focusing on gameplay is optimal in the long run. It just means that given the solutions at hand, most of the time, the best one will seem to be the ones with gameplay-focus. There could be another, much higher, peak further away, but the only way to reach it is through tough terrain and deep valleys. By this I mean that a method will not show its value until you let it evolve to a certain amount. But in video games the classic gameplay is so interesting on its own, that it's is unlikely anyone would want to make this journey.

I think that a lot of features of modern film have been sitting on a distant peak, but because the simple joy of the medium wore out so quickly, people have been forced to take this treacherous path. Video games have never been forced to do this, and this is likely why we, narrative-wise, haven't been able to evolve to the extent I think this medium is capable of.

Traveling down this path is not easy, and just walking it blindly will not generate anything useful. You will just end up lost but not found.

Heavy Rain (2010)
One way of approaching this problem is to take another medium as a springboard and to use all of its core engagement as a foundation to build upon. The best example of this is in interactive movies, which use film as their base and then build a game on top of that. This works fine at first, but you will run into similar problems as with normal games; you get stuck with a local maximum. These games rely on the language of films to provide the core engagement, and this is bound to break once you step too far away from those foundational aspects. And just as in games, every nearby path in solution space will give you a worse result.

Dear Esther (2012)

I think a much more fruitful approach is to break down games into their basic elements, and then start building from there - now with the core goal of achieving better storytelling. Games like Dear Esther have been great pioneers in this regard, and have shown how building engaging experiences without a lot of features, thought to be crucial, is possible. Sure, these sort of experiences are far from perfect and not everybody's cup of tea. But to dismiss them would be very foolish indeed. We are now starting to gather knowledge about what makes games tick in a way never seen before. Now it's time to figure out where to go next.

It is my belief that in order to make more progress, we need to start analyzing what makes games special and, instead of just applying these findings in classical ways, figure out new ways by which they can increase our sense of interactive storytelling. The path ahead will be harsh, unfamiliar, and filled with challenges, but at the end we shall reach a peak greater than what we have ever seen before.

[1]: I guess one could argue that we are back to the good old days of Fred Ott's Sneeze with Youtube and gifs, but I don't think that is true. When people watched Fred Ott's Sneeze, they watched it for the "cinematic" experience that it provided, for seeing things recreated on a screen. But when we watch a clip of something silly happening, we are watching it for the sake of the event itself. People played the original Pong because it was fun to interact, and the same reason is still valid.

[2]: I am sure that people will disagree with me on this, but to me adventure games reached a peak, storytelling-wise, with games like Full Throttle and Broken Sword and it has not really improved much since.

Tuesday 21 March 2017

Traversal and the Problem With Walking Simulators

To keep the player focused on the game's world is crucial to every game creator. While the player is traversing a space this is even more important, but at the same time harder to achieve. So how do you keep your game interesting and avoid turning it into a walking simulator?

The Wizard of Oz. Dir. Victor Fleming. Metro-Goldwyn-Mayer. USA.1939.

This blog post is based on a conversation that I had with Brian Upton at GDC a few weeks back. Most of the basic stuff here comes from the discussion with Brian, then I have added my own ideas on top of them.

Our basic problem was stated as the following: Where is the fun in simply going from place to place?

This is a problem that is very unique to games. In a movie we rarely see a character actually going places. Instead we witness the intention of going to another place, possibly see the mode of transportation, and then we're at the destination. Unless narrative-related hardships happen along the way, we never see the character actually traveling. Why? Because it simply isn't very interesting.

Games work differently. In games we have to show every single step that the player takes. There are a couple of reasons for this.

The first, and the most obvious, is that it's very hard to know what the player's intent is. When you enter a car in GTA, the game can't possibly know where you're supposed to be going. You have to express your will by actually driving your car where you want it to go, every inch of the way. When a game features cuts, like in fast travel systems, it's all based upon the player first expressing their will to go to a certain place.

In games we want the player to take on the role of a certain person. If a game simply decides where the player will go when they enter a car or start walking, that aspect is violated. There are a few games that do it, e.g Thirty Flights of Loving, but these games are usually short and made in a way where this phenomenon becomes a part of the gaming experience, or they simply contain very little player agency overall (e.g. interactive games like Heavy Rain). In this aspect, traversal is more than simply "empty travel time", it's a crucial expression of the player's agency, cementing their role as the protagonist.

The second reason is a bit more subtle. As mentioned, part of what makes games interesting is the expression of will. To achieve this, the player must know what they are able to do within the game's universe. In a movie, a character can reach for an object we've never seen before, or exclaim "I saw that shop on my way over!" despite the viewer never seeing it. This isn't possible in a game. In order for a player to know a game's possibility space, both in spatial terms and in terms of what actions are available, they need to familiarise themselves with it. They have to go through the boring process of walking about in order to form a mental picture of the surroundings. If they don't, they can't possibly know what their options are.

However, this activity is not very interesting at its core. Sure, it's fun to look at fancy environments for a bit, but after a while it gets tiresome. Most games solve this by introducing some sort of activity to the player at this point.

Sonic: Lost World (2013)
In a platformer the player always has obstacles of some sort to get past. For instance, pits to jump across or objects to avoid, During moments of traversal (when the game is not meant to pose a direct challenge) these are not very hard to get past. Still, they do require some attention. So when you are going from A to B and not really accomplishing much, you are still involved in a basic muscle task that relates directly to the game's world. This means that part of your brain is actively engaged in the activity at hand.

Think of how you sometimes zone out when you perform an activity at a certain level of difficulty. For instance; driving, knitting, or just walking rugged terrain in the woods. This is the same thing - you are engaged just enough not to get bored by the traversal.

Metro 2033 (2010)
Another way of doing this is by making use of our sense of anticipation. This is how stealth, tactical combat and horror games work. When walking towards a door you are not simply engaged in the activity of walking. You are also constantly thinking about what might lie ahead. "I need to make sure I don't make too much noise", "What might attack me from behind that door?", "When I get to the door I need to make sure I sweep the room for hostiles", and so on. So when walking, you're also engaged in the activity of planning ahead. You're still in the game's world.

Virginia (2016)
However, a walking simulator lacks this sort of engagement. Walking forward is just a matter of pressing down a key or stick. And unless you are my dad playing a game, this doesn't pose any sort of challenge at all. Your brain is basically unoccupied and the chance of your mind starting to drift is very high. Instead of being immersed in the game's world you might start thinking of what to cook for dinner or something else that is totally unrelated to the experience the game wants you to have.

I know there are some people who argue that "walking simulator" is not a fair name, but because of this issue I actually think it is quite appropriate. What happens during traversal is quite closely linked to the core of the game. In a 3D platformer your activity during traversal is still about platforming, in a horror game you are on the lookout for danger, and in a walking simulator - well, you are simply walking.

This doesn't pose a problem to everyone who plays walking simulators, and I think the "trick" is to put yourself in a sort of meditative state where you simply block out any intruding thoughts and just focus on the essence of being in the game. One way of achieving this is through stuff like music. It's one of the reasons why The Chinese Room's titles have been so successful. Their amazing music often becomes front and center during these moments of just walking, and by doing so keeps the player in the world.

Still, I think this poses a problem and it's something that anyone making a narrative-heavy game needs to think about. It's similar to how scenes are constructed in movies. If a scene simply starts and ends on the same note then it falls flat and gets boring. Just like some walking simulators can get away with just walking, some movies can get away with this for a portion of the audience. But that doesn't mean it's the best way to approach the problem. In the same way as film scenes thrive on there being dramatic motion, so should games try to find an interesting activity to tie together all of the traversal.

If you are making a game that uses a classical game mechanic, then this doesn't pose a huge problem. But it's when you want to go off the beaten path and try something different, especially when the focus is on storytelling, that this becomes crucial. You need to consider: when the player is simply walking around, what keeps their mind in the game's world?

In one of our upcoming super secret games, we want to explore new ways of telling a story through gameplay. This makes the issue of traversal really high on the list of things we need to make work. A key component for us in solving this has been to focus on what sort of fantasy it is that we want our players to partake in. The trick is then to make sure that our players focus on this fantasy at every single moment. We want to make sure that the players are preoccupied with things that relate to this fantasy, and that these actions require their attention.

The way we intend to do this is by packing the environment with narrative- and gameplay-important information. The more of this information the players have, the easier it is for them to create plans for overcoming upcoming obstacles. On top of that, the information changes over time, so players need to keep up this mental exercise even when entering previously visited locations. The crucial bit is to avoid making this procedure too difficult, as it would otherwise be exhausting in the long run. It should lie at the sweet-spot where it becomes barely conscious, coming into full focus only when important, when new information is discovered. On top of this, the information needs to be interesting in itself, not simply dull collectibles or similar. As I mentioned earlier, it is important that this task reinforces the player's fantasy.

I know this sounds a bit fuzzy, but going into greater details would be too spoilerish at this point. It's also worth pointing out that this is still in an early state, and we haven't had time to see how well it works when put in practice.

So, this is far from being a solved issue. But by simply recognizing it and gathering modes of attack, it feels like we've taken steps towards a solution.