The CRPG Project

I’ve been thinking about computer RPGs a lot recently, and trying to play a bunch. There’s a quality that I’ve always loved about especially older computer RPGs: you really felt this sense of ambition behind them. You could sort of feel that there was someone who was super excited, who kept asking the question, “what if the player could also do X?” Often, when some of the older computer RPGs would come out, they would be super buggy on release. I feel like this is because there was a prioritization on increasing the possibility space (which invites bugs) and not prioritizing polish (which involves bug-fixing).

I think I have a fascination with CRPGs because I feel like games like Arcanum, Fallout 1 and 2, Daggerfall, Temple of Elemental Evil, Wizardry 8, most of the Ultima games, some of the older Gold Box games, and many others, embody some of the original promises of videogames better than anything else I’ve seen. These games were made, it seems to me, with a real spirit of ambition toward some goal of “fantasy simulation on the computer”.

The map view in the original Fallout. You could move anywhere, at any time, encountering random events and secret locations along the way.

Perhaps I, like every other human being, am simply romanticizing the media that I played when I was coming of age, the games I played in my teens and early 20s. If I had been born 10 years later I would be crowing about Mass Effect instead. I guess it’s hard for me to completely disprove this, but many of the games on my list were made well before my time. Some were made before I was even born, and some of these games, like the early Ultimas, I haven’t even played myself, I’ve only read about them, and even in reading about them the sense of ambition is clear. I would also put a game like Divinity Original Sin perhaps into this category, as I feel that it is also a good example of what I’m talking about, and I only played that a few years ago.

“The Project” of the CRPG

The story of the computer RPG, as I see it, is that in the late 70s, you had all these solo developers trying to more or less reproduce D&D on the computer. Computers were, and still are, extremely limited in what could be done, especially for a one-player game. We don’t have generalized AI that can make NPCs who act and respond the way that a person would. We don’t have a thoughtful human-like AI who can do the work of DMing. So, even if developers didn’t want it, what they really had was a process of designing some interactive system, much like if they were designing a strategy game or a puzzle. And frankly, that’s cool! That’s way more interesting to me than if they really were able to just re-produce D&D faithfully.

I think there was this project of working on that and building on what had come before. The development of the Ultima series is really the “main character” in the story of the development of the computer RPG: here was a project spanning almost two decades that culminated in Ultima VII, one of the most interactive and deep (if also deeply flawed) RPGs ever made. Ultima VII is seriously nuts. From its Wikipedia page:

The gameworld of Ultima VII is renowned for its interactivity: virtually everything not nailed to the ground (and not excessively heavy) can be moved, taken, or interacted with in some way. It is possible, for instance, to bake bread, to forge weapons, to play musical instruments, to paint a self-portrait, and to change a baby’s swaddling. The Avatar and his companions, if not fed regularly, will complain of hunger pangs and severe thirst, and will even perish if these matters are not attended to eventually. If they come across a disgusting or gruesome scene, they may groan and vomit; sufficient intake of strong alcoholic beverages will also result in visible nausea.

Before I go further let me give a nice clear explanation of what it is I am calling the CRPG “project”: a collective mission to pursue new ways to create fantasy world or adventure simulations. These games push the boundaries in terms of interactivity in ways that increase real immersion*. Objects that respond to physics, tools that have many uses, a wide variety of ways that NPCs can be interacted with and related to, and systems that have emergent/toy-like properties all are good examples. CRPGs that are part of “the project” are usually very good toys, to use my nomenclature. Most games these days have RPG elements, but something like a JRPG, as a point of comparison, is a much more tightly controlled pre-authored experience that’s often much more linear and has fewer emergent systemic properties.

* not the video technology stuff that gets hyped as such. Get the idea that “4K 120 FPS lush visual landscapes is where immersion comes from” completely out of your head. It is marketing propaganda, it’s “blast processing”, it’s like this never ending Ouroboros of hype and nothing more. Of course visuals, audio, and everything else about the application matters, but as anyone who has played older games can tell you, they’ve been every bit as engrossed in the games of their time as anyone else.

You just want a murder hobo simulator

Before I go any further, I think it’s important to talk about patterns we find in the existing manifestations of games that were indeed part of “the project”. Very early in the Ultima series, designer Richard Garriott noticed that when he gave players the ability to interact with the world, players started behaving very badly.

Garriott tried to minimize the murder-hobo-ism with his Virtues system (but not so much in Ultima VII)

The classic manifestation of this is the “murder hobo”: players soon become little more than a marauding godlike power rolling into various towns stealing everything and killing anyone they feel they can get away with (up to and including possibly “everyone”). I’ve definitely gone this route myself more than a few times. RPGs and strategy games both suffer from the same cultural, or language problem: we understand action in videogames largely through a language of violence and war. Everything is framed around violence, even the “non violent” possibilities in Fallout are still framed in relation to the naturally assumed violence that would normally be there. “Adventures”, as a concept, are often largely built around violent dehumanizing themes, as well as colonialism, domination, exploitation, and other similar problems.

So, when we conceive of a game where you have a lot of freedom, the natural inclination of both players and designers is stuff like “you can kill any NPC”, “you can steal from anyone”, “you can destroy things”, “you can seize power by force”, etc. Even just writing this stuff out right now, I’m getting a little bit excited, because of a lifetime of this kind of thing being so hyped, and the ways that I’ve been trained to think that “fun comes from obtaining power at someone else’s expense”.

I don’t want to downplay how much this “kill everyone in town” aesthetic is a problem. Like, we call it “murder hobo simulator” and it sounds all cute, but it doesn’t sound as cute when you call it “school shooter simulator”, and this is really not that far off of a characterization, if you think about it.

It doesn’t have to be this way, of course. In life, and in the fantasy worlds we create in other mediums, we do many things with other people. And it’s not just violence that can be abstracted out into numeric or systemic expressions; the Sims, dating sims and many other kinds of games have done that for years. I’m not against violence as an option in these games, but I recognize that right now, if there’s going to be interactivity of *any* kind at all, it’s going to be along the lines of violence. With that said, “can kill or not” really should not be the only way you can interact with NPCs in these worlds, and the fact that it is is a terrible indictment of our society and our imagination.

 

The end of the project

It was really around 2000-2005 that the project was killed off. The games industry got bigger and bigger, and the expectations for content got more and more expensive to produce. Numerous great companies involved in this project at this time faced a choice: sell out, or die. And by “sell out”, I mean, produce video games that are first and foremost consumer products that are “CRPG-flavored”. No companies, as far as I know, were able to continue the project through this time. Companies like Bethesda switched to making games that were essentially more like Diablo or one-player Everquest. Elder Scrolls 2: Daggerfall was one of the most ambitious videogames of all time. It has, I think still, the largest overmap in any videogame, and the dungeon maps are randomly generated labyrinths that you absolutely can and will get lost in.

Actually, Daggerfall was arguably a little too “the project” for its own good, and I personally find Elder Scrolls 3: Morrowind to be a much more playable but still pretty project-y RPG that I would still recommend to people. But that’s also kind of true of most of these games, which again, when they came out they were buggy messes, and most of them continue to have pretty massive flaws. That’s why this is a project that’s larger than any one game.

Elder Scrolls IV: Oblivion was the first “post-project” Bethesda game. It’s a very good example of the opposite of a sense of ambition, if you ask me. With Oblivion, the focus was on making it easy to use with a controller, making nice water effects and pretty landscapes, and making sure everyone has a relatively “balanced” and predictable experience. Oblivion introduced “global leveling”, so that no matter what level you were at, all of the creatures in the world you would encounter would be at your level. It also massively un-complicated the magic system and many other systems. Overall, Oblivion was much more of a “reliable market product” that was vaguely RPG flavored, but really has about as much in common with the old CRPGs as one of these mobile gacha games do.

In a general sense, I think that bigger budget games are sort of orbiting around becoming mobile gacha games, regardless of their genre. At this point the “retention” stuff of the Farmville era already got introduced, embraced, criticized, and now finally just accepted. It’s now just… the water we’re swimming in. No one really bats an eye when they see loot boxes or arbitrary “leveling up” schemes all over their games.

I wouldn’t go as far as to say that the CRPG project is fundamentally at odds with the “commercial retention product design”. I think it’s true that all RPGs ever made, pretty much, rely somewhat on Skinner box style “reward scheduling” to keep players playing. I think the real difference between the old games and the newer games is, now that’s kind of all they are. A little hamster wheel to run on and some reward pellets.

The return of the project?

Around 2012 or so when Kickstarter started kicking, a bunch of these 1990s devs—people from Black Isle/Troika, as well as Interplay people like Brian Fargo—came out with these big Kickstarter campaigns to remake games like Wasteland and new spiritual sequels like Pillars of Eternity. A lot of people backed these things. Probably most of the people backed them because they want (or at least, they think they want) “more games like Fallout 2“. Many are definitely doing the thing where they think “the games that were coming out when I was coming of age just so happen to be the best, I want more like that”.

Some of these games, like Pillars of Eternity 2 (especially now that they added turn based mode), really do succeed at the idea of “re-creating Fallout 2“. If you want more Fallout 2, you really can’t find anything that meets that criteria as well as PoE2.

But, is this really a continuation of “the project”? I don’t think so. Because the project was not to “re-create” anything, it was a process of pursuit of something, something possibly unobtainable. The project’s mission, I think, whether the developers themselves would conceive of it this way or not, was to ambitiously suggest that things which had never been done before, could be done. Pillars of Eternity 2, by contrast, absolutely has been done before.

Last year’s The Outer Worlds is arguably an even clearer example of “a commercial product with a ‘The Project’ flavor”. It is a slightly more on-rails re-release of New Vegas with a new skin. It’s actually shocking how much The Outer Worlds just is Fallout, again. Like, down to the kinds of guns that exist in the world, it’s all just… re-creations. And on one level, there is a place for that. Maybe people do just want more of the same thing. But I also think the same forces that lead developers to re-produce this “game that already has an established player base”—not just capitalism, but an extremely harsh capitalistic, over-hyped, consumerist market—these forces also lead these same developers to make these things actually even more boring than their 1997 counterparts. The Outer Worlds is Oblivion-fied: there’s loot everywhere and it’s all in nice clearly designated Yellow Loot Boxes. The enemy Marauders (who you can’t talk to or anything, they just attack you) stand next to explosive barrels all day, waiting to be killed by you. The Outer Worlds is “extruded videogame product”, to quote my friend Holographic Doll.

So many people told me I needed to play The Witcher 3, but to me the Witcher 3 is more like GTA than a CRPG. Not because it’s 3D, but because it’s massive and almost completely empty. There’s tons and tons of space and nothing to actually do in the space. You go into a city—a whole city—and there’s like, five doors you can interact with and five NPCs you can talk to. While it is open world, which is an important aspect of games like this (the ability to truly explore, to roam, to wander), the feeling of “I can explore” dies off fast when you get the sense that there is not really anything to discover except for more of the same. To call The Witcher 3 “Fantasy Simulation” feels like a stretch, because it’s clearly not simulating anything any more than GTA or World of Warcraft is. It’s definitely a good videogame, and maybe it’s a good RPG in some ways too. But it’s not a continuation of the project.

A better example of a return of, or a continuation of the project would be something like Divinity: Original Sin, which really tries to push the boundaries of what’s possible in combat. But also, a lot of the kind of stuff that was being pursued in these older games has just moved to less hyped, smaller corners of games. Some of the Paradox games, or something like Dwarf Fortress, is where a lot of that kind of ambition is still living on.

People are still making CRPGs, but The CRPG Project is more or less dead, I think. I have some high hopes for Baldur’s Gate 3, and a much smaller amount of hope for Wasteland 3.

Text in CRPGS

In interviews I watched, the designers of The Outer Worlds talked about something that’s kind of been a mantra in the CRPG world: the concept of “choices & consequences”. That sounds good, right? But I actually think that it’s the wrong framing, that’s asking for the wrong things.

So, text in RPGs is kind of the big elephant in the room I haven’t talked about much yet. Firstly I should say that when a game has really great text, I love it. Planescape: Torment is still one of my favorite games of all time, pretty much entirely because of the text (the gameplay was very bad). So, by no means do I doubt the amazing amount and kinds of value that great text can bring to a game.

I guess what I would say is that the “text route” is a very distinct route from what I’m talking about. “The project”, as I think of it, is a sandbox. It has emergent properties, it is surprising. In my view, it’s difficult to achieve anything like that using text in games. Text and text-based choices and their consequences are all pre-authored. There is nothing that can happen in an RPG whose primary value comes out of text interactions that can surprise those who wrote all the text. They know literally all of the things that are possible. Sure, there may be some really great, brilliant, moving and powerful moments that the author wrote into the game, but this is something distinct from a “fantasy world simulator”.

I recently played Shadowrun: Dragonfall. It is a generally well-made RPG. It is very linear, you can’t interact with anything at all in towns pretty much, you can’t interact with NPCs at all except to click on them to get more dialogue. The dialogue is good: I think it’s a good example of a game with a well-realized universe, good characters, and most of the things one would like to see in any text-based game.

But dialogue trees can’t have emergent complexity. They can’t result in outcomes that the developer did not manually write in themselves. They do not give much space to explore. And frankly, Dragonfall might as well have been a visual novel (or, if you like, a visual novel plus XCOM 2012 combat missions). The same could be said of Planescape:Torment, one of my favorite games of all time. I wish it was a visual novel, so I could actually show it to people!

There is nothing wrong or inferior about the “visual novel” as a form or a type of value, but it’s very distinct from the value of games like Ultima and the old Fallout games. These games do have dialogue trees, but crucially, an NPC isn’t merely a dialogue tree. What these games need is NPCs that have behaviors. Oblivion, ironically, touted their “radiant AI” system, which apparently the other games have as well, and it’s actually not a terrible example of what I mean.

Alright well, it’s a little terrible maybe – but the point is, you can see that there’s some effort to make NPCs actually have behaviors. It was supposedly improved somewhat in Skyrim as well, which I didn’t play much because it seemed very terrible in every other way. From the Elder Scrolls Wiki:

I’m actually surprised that there’s this kind of stuff in the game, and happily surprised. Because to me, even if this stuff probably isn’t as sophisticated as what they were doing in Ultima VII (I’m not certain, someone feel free to correct me; in any case, I would have hoped that in 14 years we could do better than tread water), it still kind of counts as someone not just trying to reproduce a Diablo or Everquest “quest engine loot/experience delivery mechanism”, but actually trying to make some sort of believable fantasy world. This project of NPC behavior is something that Dwarf Fortress is incredibly good at and it’s made by 1 or 2 people total! Why can’t we have some more of that in our RPGs?

I don’t expect games with high levels of this kind of interactivity and world-simulation to be smooth. I understand they will be buggy, and possibly not even that great as games. I don’t expect these games to be even adequate, but I do expect us to try. That’s the project. I am asking for games like Skryim to be more like what they advertise themselves as.

I want to reiterate the point that I am not asking for more of the 1990s era CRPGs. That’s what Pillars of Eternity 2 was, and… it was fine (except the loading screens, holy hell), but it’s not what I am looking for. Despite whatever romance I may have for these games, they were deeply flawed, and I want things that are much better than those things. I want new games that push the boundaries in new ways. I’m told that Disco Elysium does a lot of new things, but because of its reliance on dialogue, I’m not sure if it’s a part of this specific project or just a really cool game of its own (I’ll find out eventually, once I get through my backlog).

If you care about RPGs, I think there are two ways to go:

  • Visual novel or a text-dominated, highly authored thing, if you want like, a story
  • What I’m talking about

As I said, I’m really excited to see what Larian does with Baldur’s Gate 3, above anything else going on. I’m also interested at some point to find a way to make my own kind of game like this, perhaps after Gem Wizards Tactics.