Check out some of the critiques for games like No Mans Sky. The short of it is that procedural “story” doesnt really work. You still need writers for engaging content, and that doesnt scale. If you remove written story you get something closer to a tower defense/tactics/arena shooter/grinding genre, which is “fine” but substantially different than most open worlds people envision.
See dwarf fortress for a counter-example. I’m of the opinion that story-writing is a red-herring in video games; an attempt to apply “tricks” from other mediums, by disregarding the “interactivity” of video games. A story can be sufficiently created by means of the player’s operations, if the simulation is sufficiently complex and built to support it.
Another example is competitive fps games like CoD/Halo/Quake — the story (and singleplayer mode in general) is wholly unimportant; players will continue to have, and create, stories, despite the multiplayer modes offering little more than a setting. Interaction between each other becomes the story you tell.
No man’s sky (NMS) was simply a bad game, and poorly implemented procedural generation (procedurally generated games are appealing in that you’re exploring the system/simulation, and trying to get a handle on it. No man’s sky had a lot of “elements” to its generation, but its system was very basic; it didn’t take long to mostly understand it, and it was sufficiently complex to allow interesting new phenomena to emerge). NMS’s failure says very little about what procedural generation can offer, because it didn’t make a very good attempt at it. Spore is another example of a failed attempt. Borderlands is another, to a degree. None of these games understood that it’s the system thats interesting, not the combinatorial explosion. Thus their procedural generation is totally unsatisfying, and uninteresting.
Dwarf Fortress is a different kind of game from No Man's Sky or The Witcher 3. It has a huge learning curve and takes a lot of effort to play. For every amazing story that people crafted there are hundreds, maybe thousands of failed game attempts.
That's fine and purely procedurally generated games can be amazing (I'm working on one myself), but they are a different type of game play, and often attract a different kind of player. Not everyone wants to be a musician, some people just want to play rock band.
For open world exploration games in particular, in every single example I've seen, mixing in hand crafted world building with systemic game mechanics always provides a better experience.
The learning curve and effort to play are more due to UI and the authors' own stubborness-slash-artistic vision. Rimworld is a somewhat-simpler DF without as difficult a UI and is still quite capable of generating stories for its users, organically as a result of interacting systems and random events.
For those who would like to play Dwarf Fortress, but are turned off by the incredibly steep learning curve or the offputting visuals, I suggest you look into Rimworld. Its similar to Dwarf Fortress in that it is a procedurally generated story with creation and survival aspects. But it is easier to learn to play, and has modern (albeit simplistic) graphics. I've enjoyed it quite a bit.
I don't think the difference for Dwarf Fortress is the learning curve, I think the difference is that Dwarf Fortress is better designed and it is designed in such a way that it's goal is to produce "stories" that Tarn and Zach have explicitly written. I think for No Man's Sky, they had a much more vague idea of what they wanted the user to experience, and thus the experience is more vague, and less interesting.
That's a good point. That is a big difference with Dwarf Fortress. It's definitely a masterpiece of game design.
To donavanm's point that you need writers to create engaging content, as you've pointed out Dwarf Fortress has writers--they're just operating on a different level.
But I still think the learning curve is a big part of it. Just look at how many "Get started with Dwarf Fortress" tutorials are out there.
I'm pretty sure the learning curve is more a result of DF's total lack of interest in reducing the learning curve, than any kind of necessary function of it. Obviously the complexity of the game leads to complexity of the interaction, but 10% of DF's learning curve is its nonsensical UI, probably another 30% is the fact that it has no tutorial or manual of its own, 20% is all the additional tooling surrounding it, which should be but isn't native to the game. And if my percentage allocation means anything, then only 40% of the game's current learning curve is essential (that is, unavoidable given the game's complexity).
I'm also not sure its fair to consider DF's writing as similar to donavanm's -- they're not operating at a different level so much as a totally different task. donavanm's writer as I understand it is trying to lead the player through a story, whereas DF's trying to produce a story through the player. The former produces an overarching theme, plot, characters, etc. The latter produces pieces that could fit in a story (or stories), and hope it will occur as an emergent phenomena of play.
It's like writing a protocol versus using one; you could say both are doing software engineering, but they're very, very different kinds of software engineering, and require totally different thinking.
Notably, story-component-writing probably also scales pretty well (the engine develops it out), while story-writing does not. Which of course is true of procedural games in general. You get the engine and the pieces right, and you've got exponentially many scenarios to explore (the problem is that, if you do it wrong, you have exponentially many scenarios that you don't want/care to explore ~~ No Man's Sky)
And while it isn't anywhere near as deep as DF, the mass attraction of Minecraft shows that procedural generated worlds where there are no real goals can be very attractive and approachable, even to children.
That's very true. But again I think it's a different type of game. Very few people, in my experience, spend much time exploring Minecraft for the sake of exploration.
The genius of Minecraft is that it's a vast canvas for you to shape, not an interesting world to study. The purely procedurally generated world that it generates are fun to build on they just aren't that interesting after you've experienced it for a bit.
>I'm also not sure its fair to consider DF's writing as similar to donavanm's -- they're not operating at a different level so much as a totally different task. donavanm's writer as I understand it is trying to lead the player through a story, whereas DF's trying to produce a story through the player. The former produces an overarching theme, plot, characters, etc. The latter produces pieces that could fit in a story (or stories), and hope it will occur as an emergent phenomena of play.
I don't think they were only talking about an overarching large scale story, but also the small scale side stories that happen throughout the world. Allowing the player to overhearing the backstory of 2 train robbers right before they start the robbery, or coming up with a back story for an abandoned house and telling it through the items left around. Small additions that add flavor and detail to the world. When you add to that some game mechanics around choosing to aid or stop the train robbers, choosing to find the lost treasure hinted at by a journal found in the abandoned house etc..., and then the game world reacts and changes based on those actions, to me, you're doing exactly what Dwarf Fortress is doing. To me the difference is scale (granted it's a huge difference and I think much more difficult to pull of right). Dwarf Fortress does it on a much grander level that allows for more variation and more genuinely unexpected emergent behavior.
The creators of Dwarf Fortress write small stories that add flavor to the world and then they add systems to make those stories possible. It requires much more human effort to develop than just piecing together systems and hoping that the interactions are interesting, which is what I think donavanm meant by being less scaleable.
Ah, Dwarf fortress. I loved hearing the stories generated from it and resolved to learn to play it, but that resolve quickly faltered in the face of the interface and learning curve. Sometimes I think to myself "I should try again", but I believe that it (along with Eve) is a game I truly enjoy reading about rather than playing myself.
As the sibling said Most people dont see dwarf fortress as the same “open world explore & build” genre. And the arena style FPS was called out explicitely as an interesting, but different, type of game. “No story” doesnt mean no fun. But when people say “I want to go and build and explore” they usually, unintentionally, miss the draw of a well done story component. There are plenty of interviewers with producers of games like warcraft, witcher, or skyrim who explicitly call out the limitations/benefits of a written story system in an “open world.”
I think the issue is that you’re thinking of open world rpgs, which have trended towards very heavy-handed story telling. Mostly because they derive from dnd 3+.
But you also have dnd/adnd to derive from, which put much less effort into story telling, and more into setting/emergent stories. Roguelikes, dungeon crawlers, etc offer very, very thin stories wrapping the gameplay (dcss for example boils down to: you want to steal the orb of zot, and its on the 15th floor of this dungeon). Minecraft also gained most of its popularity before it too added a very thin story on top.
But the main thing to realize is that most games don’t have very compelling or interesting stories; at most approaching a marvel film. Yet the succeed regardless.
And even warcraft (I’m assuming WoW), I’d bet good money that 90% of its players don’t have any idea what the plot is, except at the most superficial level.
I think its mostly that the open world genre happened to really only have RPG’s currently available and well funded (I cant even think of any major ones atm beyond bethesda, witcher and bioshock); but there are plenty of games that put zero to little efforts into any kind of story, that it’s not at all obvious to me that its required in open world games.
I’d like to see some of these interviews you’re referencing, if you have links handy.
Indeed, as a once-heavy Warcraft and WoW player I can confirm that Blizzards story writing is to be actively ignored, even mocked. Great games, terrible writing.
> I’m of the opinion that story-writing is a red-herring in video games; an attempt to apply “tricks” from other mediums, by disregarding the “interactivity” of video games.
I’d offer Braid and Bioshock as clear counter examples, where a significant part of the emotional punch of the storytelling comes from the interaction. Brothers: A Tale of Two Sons too.
Much like “build your own story” as a genre can’t be judged from the failings of NMS, it’s just as true that games as a conventional storytelling medium as a whole shouldn’t be judged based on the “make it cinematic” approach that a lot of studios aim for.
