The thread linking the games Mr Lee praises is that they are, to a large extent, sandboxes. Yes, some of them have endings, but they get the player there by dumping them in a world and letting them do what they like.
Portal is the least sandboxy of the games mentioned (maybe apart from Tetris!), in that it has highly structured, intentionally designed levels with well-defined goals, but it still manages to give the player a huge amount of the responsibility for getting through them.
A weaker thread is procedural generation. Dwarf Fortress and Brogue have procedurally generated levels. Journey doesn't, but probably could have done. Portal doesn't, and i doubt could. Tetris kind of does. But that's a higher incidence of procedural generation than in the general population of games. Is there something about procedural generation that intrinsically puts narrative responsibility back in the hands of the player? That seems plausible; a game designer who gives up the power to shape levels explicitly gives up some power to shape stories explicitly. Think also about Civilisation and SimCity; the software is entirely story-free, but any given game of either of those is bursting with stories.
What tests of this hypothesis are coming down the pipe? No Man's Sky is the one that springs to mind.
Thank you, I was going to say that one of the main reasons I play Civilization is for the often strange stories that can develop from the procedural content and simple sets of rules. "The rebel sons of nearby Valetta who embark on a hundred year campaign of foreign shores, sometimes dashed on fortunes rocks, but always rebuilding, and becoming the first people ever to experience firearms."
An interesting article, gave me much to think about as I'm planning out a game to make.
My thought on it tho, I think the article misses a GOOD use of cutscenes: A good story is all about tension and release, and a cutscene can provide a necessary release from intense gameplay (not that it's the only way).
Am I wrong on this? I'm sure there are differing opinions on this depending on what game genres you enjoy (I'm quite a fan of traditional RPGs).
Yeah, when well-timed a cutscene can be part of a syncopated rhythm, or be a satisfying moment when you can throw down your controller in exhaustion after an intense piece of action, or they can be used to frame a shift in gameplay.
That's not to say they're not overused in general, but they can have their use.
On the subject of overuse, one reason is that a linear story and movie-like cutscenes are easier and cheaper to write and make, at least compared to the alternative. The processes are well-known and so they're easier to budget for.
Half-Life 1 did this so well. It was primarily a shooter, but if you followed closely you could get a story arch. You never lost control of you character and every cutscene was optionable( except the very first and last, you still need some pacing ). This is why I consider Half-Life 2 an inferior game, even though it does everything else better, you constantly get into situations where the main characters talk for several minutes and you can only watch and wait. HL1's gameplay flows, HL2 is a stop-and-go.
I agree, except that even on modern machines HL stops the game every 2 minutes to load the next area (this is especially painful when the train is moving through the intro cutscene.)
Dark Souls and Demon Souls, two games normally only noted for being extremely difficult, are some of the best examples of ludonarrative assonance I've ever seen. The mechanics of the games not only reflect the events of their plots cohesively, but also serve as metaphors towards larger aesthetic and philosophical themes shared by the stories and settings.
I find it rather strange that the author asks for open mindedness but seemingly dismisses the whole idea of the cut scene narrative. On the other hand he praises the generated, "natural" narrative of Dwarf Fortress, Brogue and Journey. One of his remarks about Journey is rather interesting:
> When you find another player, there are visual cues that underscore their presence and introduction. When you communicate with them through singing and body language, all sorts of imagery forms in your mind about the other player’s personality (that’s character development!)
The character development part is a very important part of a narrative, and is in its way, only possible because of other, real players. His other two examples, Brogue and Dwarf Fortress, have huge short comings in this regard. A true, emotionally deep and complex character development and interactions between those characters is only possible if the other characters are played by humans, the characters are predefined or if there would be a sophisticated AI. The first variant is used in Journey, the second is used in "cut scene" type games and the latter is currently not technically possible. So when the author claims:
> At least silent films are excused by their technical limitations – no comparable excuse exists for games.
I take that with a grain of salt. Dwarf Fortress, Brogue and Journey are great examples of how to expand the narrative experience and where games could be heading to in the future. But in my opinion, they are an addition to, rather than a replacement for cut scene narrative games.
I think it's not really about dismissing the idea of the cutscene narrative - cutscenes can certainly be very well done, but a challenge for game developers to develop the medium's own mechanisms for storytelling instead of continually borrowing from film.
The best storytelling we've managed so far is temporarily suspending the "game" part of video games and putting on a Hollywood production. We're not really blazing our own trails, we're just getting better and better at emulating filmmakers.
There are few games that seem to be able to tell a story without completely stopping gameplay or removing agency - the very thing that makes video games video games. Developers realize this limitation, but our only response to it so far is to let the player move the camera a bit and throw in quick-time events (mash X to escape! oh too late, let's try that again!). That's not really agency.
One game that's interesting is an indie game called Kentucky Route Zero - it's a point and click adventure but with level design borrowing heavily from stage theater. It's a nice breath of fresh air, even if it isn't a complete reinvention of in-game storytelling.
I think Bioshock Infinite needs a quick nod for some advances in storytelling. While it's not free from player-control-robbing cutscenes, a huge portion of the game's story is communicated in its environment - something that film is less capable of doing, and games are uniquely positioned to exploit. It's a good example of something playing to the medium's strengths (if only the gameplay was as good as the storyline...)
I think the overarching point is that "storytelling" simply can't work in the same manner as it does in books and movies.
If you tell a story in a cutscene, you then aren't playing a game at that point. Or if you plot out every possible interaction (kind of like The Walking Dead games) you have to limit the outcomes to a manageable level, which hamstrings the gameplay. So what's the solution? The article is arguing that you have to abandon narrative and focus on what you want the player to feel directly.
The problem is that "game literacy" is focused on all the wrong elements. A game like Doom has much better, tighter, more interesting gameplay than something like The Last of Us. (which I still really enjoyed!) And you can be moved by the story, characters etc. from The Last of Us, but it's a lot more like how you feel watching a movie. With Doom, people create their own narratives by interacting with the game, and those narratives have a different kind of power. People still tell stories about playing Doom for the first time 20 years ago with their friends. The objective, I think, is to make a game where those personal stories are directed to achieve a specific effect.
There's one example of this effect that I've read about before. There was a game in the 80s where you play a war game as either the U.S. or Russia. If you play as the U.S., you stockpile arms and try to win the war as you'd expect. But if you play as Russia, you have a whole different set of concerns - starvation, winters, limited resources, aggressive neighboring countries and other factors which lets you experience some of the desperation Russians must have felt during the Cold War. It's a pretty powerful story to tell through gameplay alone, and a story that was told through gameplay without narrative.
You raise a good point, but even if sophisticated AIs are not technically possible yet, predefined characters can still be done well without (ab)using cutscenes.
I loved the contrast of Tomb Raider and Portal - interactive, fast thinking rather than "press X to not die" nonsense during a "cutscene".
I also really love the way they break down storytelling axes - dungeons and dragons adds the "interactive" axis without the "AV" one, which presents its own challenges. (if you want a good example of this, I recommend the "critical hit" podcast[1], although be warned you'll need to listen to literally hundreds of hours of people playing D&D, which is not ideal if you're only interested in the story's interactivity.)
I thought they missed a trick not analysing Dishonoured a bit more though, as the game world reflects your playing style (high/low chaos). Which reminds me, I should probably play that game in kill everything/high chaos mode at some point.
Author here! Dang, I browse HN daily and didn't notice when my own article gets posted until it's too late. Anyways, glad you guys liked it. One thing that I want to clarify is that the examples and analogies I give are just to help illustrate points. A lot of people have read too much into them, and that's not my intention.
Also, I don't mean to disparage any particular game, or style of games. My goal is to show that video games are a very unique medium for presenting narrative, and so far we have mainly just tried to emulate cinema. There's a lot of experimentation out there, and still more that needs to be done, before we can say we've really understood the medium.
Books have more senses than movies and games, because in books, you can describe touch feelings, smells, temperatures, etc., thus giving readers a multisensorial experience. In movies you can only rely on sound and image. Zero other senses. Smell in books is superior to smell in movies.
I wouldn't say zero sense. When I watched sin city, I could smell the yellow guy. It all depends on how the story is presented. The color scheme of sin city definitely has sensory effect beyond what you can see.
Portal is the least sandboxy of the games mentioned (maybe apart from Tetris!), in that it has highly structured, intentionally designed levels with well-defined goals, but it still manages to give the player a huge amount of the responsibility for getting through them.
A weaker thread is procedural generation. Dwarf Fortress and Brogue have procedurally generated levels. Journey doesn't, but probably could have done. Portal doesn't, and i doubt could. Tetris kind of does. But that's a higher incidence of procedural generation than in the general population of games. Is there something about procedural generation that intrinsically puts narrative responsibility back in the hands of the player? That seems plausible; a game designer who gives up the power to shape levels explicitly gives up some power to shape stories explicitly. Think also about Civilisation and SimCity; the software is entirely story-free, but any given game of either of those is bursting with stories.
What tests of this hypothesis are coming down the pipe? No Man's Sky is the one that springs to mind.