What is an RPG?


I threatened in another thread to write this one, so… in PAX South, a panel of developers was asked “What’s an RPG?”

This was a panel on online RPGs, which I sadly only noticed after the fact, or I might not have screwed up so badly that I will facepalming at the memory of it for years to come, but… the panel was chaired by Dallas Dickinson (of QC Games: Breach), and consisted of some fairly heavy hitting game design stars:

  • Gabe Amatangelo (also of QC Games)
  • Chad Robertson (of Bioware, Anthem])
  • Matt Rhoades (of Stoic Studio: Banner Saga)
  • Scott Hartsman (formerly of Trion Worlds: Everquest, Rift…)
  • Marsh Leflet (of Monster Squad games, Torchlight)
  • Jack Emmert (of Daybreak Games: Lord of the Rings Online…)

Gabe spoke first, saying he felt there were four things that made an RPG, whether tabletop or PC:

Collaborative Storytelling - he said that for a computer RPG, the collaboration would mostly be between the player and the developers and AI.
Character building and progression
Combat, with loot
Exploration - he said this was an uncertain one, and the rest of the panel, after brief discussion, agreed it was not necessary.

To me, this seems a fair start, though what’s missing there is worrying. It’s basically just a rephrasing of Bartle’s four character archetypes (Socializer, Achiever, Killer, Explorer). And as a hardcore Explorer, I was pained to see my segment so swiftly dismissed!

Chad spoke next, saying that for him, the agency you had with your character mattered. I perked up for a moment, but then he defined that as the ability to define a representation of yourself, via XP curves.

Then Jack spoke, saying he felt that persistence of the progression was important, and ideally the gameplay should not be twitch play.

Matt said that it was solely about character progression through stats. Jack pointed out that tabletop RPG was becoming became less about stats and more about story, and suggested this might be because computers RPGs had the stats-based gameplay sewn up, so tabletop had to go for the niche play

Scott said that, while they were making Everquest and such, they had thought they were making RPGs, but in fact all that players really wanted was fighting and loot.

Jack pointed out that this was just Monty Haul gaming, which has existed since the days of tabletop.

(Looking back at it, I really like Jack’s continuing efforts to draw the parallels between tabletop and online RPGs).

To me, thinking that players want Monty Haul campaigns feels like conflating what players complain about, with what they would actually enjoy playing. Players will tell you where the pain points are, and how to make the game incrementally less unpleasant to play. They will generally not make suggestions on how to make it exponentially more enjoyable to play. So if you make a grind game, they’ll whine endlessly about the combat, and the loot drops, but they will not, in general, say “deprecate the grinding and looting, and put in an epic storyline, with relatable NPCs!” – particularly if you’re not making a sequel, so people can’t compare to what went before and say “this is missing, and should be there”.

Marsh said that basically, he felt that the focus should be to make an amazing game that you could lose yourself in, and that the RPG trappings should just be a layer on top. That the grind days were over, and that nowadays, we had got to the point where we could actually make games that are fun to play, rather than forcing players to “grind to make the pain go away”. (Damn yeah! Preach it!)

He also added, and this is kinda relevant to UA, perhaps, that it takes time for an RPG to come into itself. It takes time, keeping it alive and developing it, building the community, before it starts to become itself. Jack pointed out that KotOR is still alive and well and profitable, but if you ask most gamers, they’ll say they think it died after a year or so. Hype cycles, he said, are a thing!

But after all that discussion, the overall answer they came to, almost unanimously, was “skill trees/character development, which persists between gameplay sessions.”

They were asked if they considered MOBAs to be RPGs, Battle Royales, and they replied, sure, if the character could persist between sessions. FPSs, they said, were not RPGs but only because they required twitch reflexes.

Freaking MOBAs and Battle Royales!

Not one person in an hour long panel even mentioned NPC interaction, object interaction, world interaction, morality exploration, or having an impact on the world that persists. Story was briefly mentioned, but immediately discarded as being “really only for context.”

And, forgive me, I was tired. I’d been pushing my wife around in her wheelchair for about two days by this point. I had, for the most part, been faking human pretty good. But I fear I lost it. When they asked for questions, I stood up at the mike and went full-on fatass middle-aged neckbeard. I ranted at them until the poor helper at the mic had to tap me on my shoulder and say “so what’s your question?”

And I groped around in my head, trying to sum it all up into one point, and asked: “If RPG is not the name for this type of game that I seek, then what tag should I be looking for?”

And they looked down at me with pity. One of them, I think it was Jack, bless him, leaned forward and said… “I get where you’re coming from. I was a fan of Ultima Online too.” My face fell: this guy didn’t get it at all. But he continued anyway. “But nobody nowadays wants to make that kind of game. Maybe if you look at some indies…”

Shaking, my stomach in a tight knot of social embarrassment at what I’d just done, I returned to my seat. I was an idiot to even have asked, and I’d been rude and made an ass of myself. But boy, it’d been revealing.

To me, the Elder Scrolls Online does “RPG” really well. It’s a full-on massively-multiplayer online RPG experience by a AAA studio. So… how can a panel of luminaries look at the tag “RPG” and decide it’s just persistent skill trees?

Which is all a long way of asking… what is an RPG to YOU?


Nearly every single thing you said could very easily have been me. So for whatever comfort it may be, I can tell you with absolute assurance: you are not alone. I am very nearly the stereotype of an Explorer, and I could easily have been the guy at the mic. I get your astonishment.

It’s appropriate you should mention Richard Bartle, and in at least a couple of ways. As a creator of MUD and someone who’s thought long and deeply and well about why MUD mattered and what today’s MMORPGs owe to it, he’s why I can point out that EverCrack was a descendant of DikuMUD and took its “roll-playing” emphasis from those genes.

But more importantly, it’s the very idea that there are Explorers, and Socializers, and Achievers, and (as I prefer to call them) Manipulators – that there is no such thing as “players” who all like the same thing when it comes to game development feedback – that needed to be emphasized at this panel. The devs who’ve made DikuMUD descendants have in their games emphasized what Achievers like. Cool! I don’t think anyone would say fun mechanics (loot, progression, etc.) are unimportant. But when did the AAA RPG makers get together and decide that it’s not an RPG if anything else (world-simulation/dynamics, storytelling/aesthetics) is emphasized more than the in-game accumulation of status tokens? When did the Explorers and Socializers get relegated to second-class citizens in the RPG world, pitiful figures reduced to begging for scraps from indie soup kitchens?

From your description, it sounds like the devs on that panel were so steeped in seeing RPGs from the perspective of Achiever gamers, who judge every design choice through a Mechanics lens, that the idea that the interests of other kinds of gamers matter barely registers any more.

I think there are some exceptions, notably The Witcher 3. It’s hard to imagine anyone claiming seriously that this game isn’t an RPG… but it’s also a fact that it emphasizes playing a character experiencing an emotionally engaging story more than most games called RPGs today. Why doesn’t the fan and critical acclaim this game has justly earned get other RPG developers (or, at least, their publishers) asking, “Hey, shouldn’t our game have more heart?”

The role-playing experience I want is broad-spectrum; it engages the hands, yes, but also the brain and the heart. IMO a good RPG supports my desire to define my character not just through attributes and skills and gear but also through defining that character as a person, with interests and failings and beliefs that emerge from important experiences, and then through interactions with a game world that responds dynamically to the player’s choices for that character.

When did that become too much to ask for?


I’ve always been partial to the definition offered in this article:


RPG - As I said, the acronym RPG does not mean anything. You can call whatever you like a role-playing game. Halo 3 is the best example. But I like to use two definitions – my own and the historical one. The historical definition of an RPG is that if the designer of the game thinks his game is an RPG and he gets a lot of people to believe it, then an RPG it is. The historical definition is what I like to call the Peter Pan phenomenon – clap your hands if you believe in fairies and there will be fairies.

And now on to my own definition. RPGs are brilliant, psychological devices that allow the player to discover himself in a safe environment. This psychological part already existed in “pen and paper” games. But when it comes to RPGs as video games, the first designers just transferred numbers and dungeons from the “pen and paper” experience into a video game equivalent. However, it was not long after the early days (in fact it was in the early days) when a game was about the player himself. This game was Ultima 4 and the year was 1985. Since Ultima 4, it has taken a long time for the genre to fully realize this concept, but with the arrival of Fallout in 1997 a lot of significant people in the industry started to focus on this aspect of RPG design.

An then there is a third definition – RPG stands for Repetitive Procedure Generator.

RPGs are brilliant, psychological devices that allow the player to discover himself in a safe environment. This psychological part already existed in "pen and paper" games. But when it comes to RPGs as video games, the first designers just transferred numbers and dungeons from the "pen and paper" experience into a video game equivalent. However, it was not long after the early days (in fact it was in the early days) when a game was about the player himself. This game was Ultima 4 and the year was 1985. Since Ultima 4, it has taken a long time for the genre to fully realize this concept, but with the arrival of Fallout in 1997 a lot of significant people in the industry started to focus on this aspect of RPG design.

While I don’t fully embrace this definition myself – I think RPGs are about more than just identity-forming because some people play them with goals other than the enhancement of identity – it’s not far from Richard Bartle’s own view about the nature of virtual worlds as places that enable the player to undertake a Hero’s Journey:

Conjecture: Playing virtual worlds is a kind of hill-climbing activity through identity space; the hero's journey is an algorithm for finding a very good local maximum, if not necessarily a global one.
-- Richard Bartle, [i]Designing Virtual Worlds[/i], p. 440.

Less metaphorically, Bartle writes:

[T]here is a split between the physical and the virtual self. Furthermore, I believe that it is a purpose -- no, a [i]duty[/i] -- of virtual worlds to facilitate the reconciliation of these selves, for the benefit of the real-life person. ... Personally, I view an individual as having a core self (or an agglomeration of selves, it doesn't really matter), which projects itself through the filters of action and word to present images to other individuals. The filter can be changed according to circumstances, but in the real world it's difficult to do this because of the anchors of family, community, and society. ... Virtual worlds allow the filters to be changed experimentally and (as a consequence) experientially. Through playing them you can find the filter closest to clarity: The one for the person you feel yourself truly to be.
-- Richard Bartle, [i]Designing Virtual Worlds[/i], pp. 510, 514 (emphasis in original).

In this model, RPGs are particularly good at enabling this identity-realizing function because they’re closest to being fully-formed virtual worlds. It would be pretty tough to undergo a hero’s journey in Tetris; an experience of that depth requires a more complex environment. An RPG, where action is expressed through human-like characters operating in (and on) an aesthetically and dynamically and mechanically responsive place, is much closer – if perhaps still not “close” in an absolute sense – to a system within which an identity-finding hero’s journey is possible.

Which I think perhaps says something about immersive sims, and about the Looking Glass games specifically. Here, I think System Shock is a more pure game, but as such – as less of an RPG – it’s less able to help players explore their identity. By contrast, Ultima Underworld with its much stronger character-defining elements (and its connection to the core Ultima games with their emphasis on letting players explore their personal ethical/moral viewpoints through their characters) is much more effective at letting the player self-define.

Which brings me to a couple of questions:

  1. Does this partially explain why Underworld Ascendant, which in its initial incarnation was visibly more “game-y” and less of a place for ethical decision-making than Ultima Underworld, was an unhappy surprise for some gamers?

  2. If a more complex world, with stronger RPG (role-playing + roll-playing) and dynamic elements, is more satisfying for many gamers because it’s better at letting them explore their identity, can you imagine what the reaction might have been had Looking Glass made a modern open-world game with Underworld Ascendant’s world-reactiveness? Or Warren Spector’s “one city block” game? Or – hold onto your hat – a persistent MMORPG?


There was a bunch of psychologists from places like takethis.org at PAX South who gave a set of very interesting panels on this stuff (which, frankly, I enjoyed a TON more than the RPG panel!): they talked about how people play as themselves or as their character, or some symbiotic mix of the two. They categorized these as:

  • Aspirational self (playing what you’d like to be)
  • Looking-glass self (playing as you see yourself: an old concept, coined by Charles Cooley in 1902)
  • Cathartic self (where you save the game, massacre everyone, then reload!)
  • Exploring/breakout self (where you “try on” moralities, roles and behaviors that are NOT you)
  • …and they left the list open, since there are doubtless other reasons people play ;)

An interesting exercise is to think back on all tabletop RPG characters you’ve had in your life, and how they related to you as a person - which of the above are they? What about computer RPGs? Do “who you are” and “who you want to play as” affect the types of games you play?

One panelist noted that in his studies, in the first playthrough, people tend to play as themselves (mirror or aspirational) and in their next playthrough, they tend to try the opposite, a breakout self, but often don’t complete it because it’s uncomfortable and just plain not fun. And a good question for psychologists to ask about that second playthrough is, who do they know IRL who is like that breakout self they don’t enjoy playing?

All choice comes from the player, rather than from the character, even if trying to play AS the character. And even random character generation isn’t really “random”, because we always project onto whatever character is generated.

Morality in gaming is one of my core interests, and any game which lets me explore it is near the top of my list of greats :)


A long time ago I was in a D&D campaign that lasted several real-time years. Although I myself am a very behind-the-scenes person, because the character I rolled up was a charismatic fighter I wound up being the leader of the whole party.

Which is how I discovered that role-playing a leader taught me things about leadership that I’ve been able to use in real life. So I definitely endorse the view that there can be some real value in using RPGs to play at being someone we’re not.

(This also taught me that computer-based RPGs are still light-years away from being as responsive to human choice as a tabletop game with a clever human DM. Separate thread. :smiley: )


One of the major facets I associate with RPGs is the ability to freely customize my own toolkit. I know this is a minor quibble, but everybody wants a) progress and b) customization as part of their experience as an avatar in a roleplaying land. I know for certain I favor extrinsic rewards, Diablo is positive proof that grind is a natural part of an RPG, and need not be tedious, but fun. I dunno if this adds much to the discussion, but I find more rewarding gameplay in terms of replay value in a Diablo clone than a Baldur’s Gate clone. I want my character, not only in personality, but also in skillset, to be an extension of self.

Part of my major ritual when creating an RPG party is to pick very specific names behind each character, each with their own motivations and background. It lends a lot to an experience, and is it is almost ritualistic, picking an avatar and assigning him a name. You then imbue that vessel with your positive and negative traits, or some arbitrary traits you have settled on, and roleplay as that character would.

Unfortunately, a lot of modern RPG narrative design negates the ability to really play a role. Sometimes interactions are as simple as, “can you fetch me that onion?,” where the only response is a single yes, or a single no. What if I want to refuse a fetch quest to be the bad guy? Does the game support alternate pathways through conversation that are not simple yes/no binary switches?


You’re not alone in that at all. Back when I used to hang out at Terra Nova, Liz Lawley wrote a wonderful short piece: “In Praise of the Grind.” It was such a perfectly on-point description of the pleasure of occasionally switching off the heavy lifting and just Doing a Simple Thing that I never forgot it.

This is one of the reasons why I’ll always be disappointed that Storybricks didn’t click with enough people to meet its Kickstarter funding goal. That idea of many alternate pathways – something more representative of the richness of possible character interactions, including some that have nothing to do with a specific practical goal – was something we were really interested in delivering.

Maybe someday…


I think limiting options sometimes helps dialogue to overcome the “encyclopedia entry” effect. The closest game to nailing dialogue for me was Vampire: the Masquerade - Bloodlines, with honorable mentions of Shadowrun Returns and Ultima Underworld.


https://www.filfre.net/2011/08/defining-the-crpg/ offers this:

A computer role-playing game (CRPG) is an approach to ludic narrative that emphasizes computational simulation of the storyworld over set-piece, “canned” design and narrative elements. The CRPG generally offers the player a much wider field of choice than other approaches, albeit often at the cost of narrative depth and the scope of narrative possibility it affords to the designer.

At least for now, I think I’m going to leave it at that. Most other definitions tend to emphasize character-building and leveling elements as a prerequisite, but, while I certainly acknowledge their presence in the vast majority of CRPGs, it seems limiting to the form’s possibilities to make that a requirement. Of course, I could have also simply used the definition we used in the 1980s: in adventure games you explore and solve puzzles, in CRPGs you explore and kill monsters. But that’s just too easy, isn’t it?

I like that definition a lot better than the “if it has a skill tree it’s an RPG” definition.
I don’t agree with it… I think that an RPG is best if it emphasizes narrative over sandboxing and combat.
But that just highlights the disagreements over what “RPG” means.


I forgot to mention the “type in a word” dialogue systems like Wizardry 8 are honorable mentions.


https://imgur.com/gallery/zzxOhHQ I found an “RPG-em-up”.
Basically, “shoot-em-up with skill trees”.
Perhaps I need to just accept that RPG will never mean Role Playing Game ever again. :frowning:


My go-to definition of an RPG for the last few years has been: A character growth simulator with a plot.

That seems to get to the heart of what is, and what is not, an RPG.


I like that a lot. So naturally I need to poke at it a little. :smiley:

Specifically, the “character growth” part. Looking at most games generally considered to be RPGs, I think it’s very fair to say the great majority of them feature character progression in some form. In most it’s visibly mechanical; your character is represented by numbers and those numbers increase. Online MMORPGs are all about this. In some, progression is more about gaining distinct skills – the Deus Ex games try to take this approach, and I think the intention was for UA to do likewise. In a few RPGs, progress is more closely tied to the story; as the character gains knowledge about the world (some of which is personal growth, some of which is collecting plot coupons), this enables solving more difficult story challenges – detective games follow this pattern.

What we don’t see many of are games that do away almost entirely with the idea of character growth, of innate aspects of a person increasing in utility or meaning over time. But there are a few such RPGs, of which my favorite example remains Traveller.

In Traveller, you roll up a character complete with attributes and a vast array of skills – and that’s it. You don’t play the game to level up; you just play the game as that character. Although there is a way to increase levels in skills, it takes so long that it’s not someone any player ever focuses on, as opposed to simply enjoying doing things in the game world. What progress there is in Traveller lies in getting better gear… but that’s not intrinsic to a character.

Yet I don’t think any serious person would try to insist that “Traveller is not really an RPG.” There are character stats, and gear progression, and players certainly can role-play their characters – but there’s essentially no “character growth” gameplay. It’s certainly not a character growth simulator! But it also certainly is a roleplaying game.

So how can the existence of static-character games such as Traveller be fitted into the definition of roleplaying games?


This brings up a good point, because most developers seem to view grinding as a bad thing. But take for instance Diablo, where the entire game is about the grind. There is little or no verbal interaction. Everything that happens is through “doing.”

The challenge, I believe, is to make the grinding fun, and not repetitive.


Diablo is generally considered a Roguelike, not an RPG (or CRPG, to be more precise).

As for Traveller, I’d just chalk it up as a weird edge case. If it were a computer game I’d consider it more an adventure game than an RPG.


It’s actually more of an aRPG. Rogue-like connotes a game with perma-death.

Most RPGs have the treadmill in some form or another. The point is to make it interesting.


One more thought: one thing I liked about the first Diablo was the concept of findable skill tomes. It tied exploration directly to progression. If only the tomes had been class-specific, that system could have proliferated into the second Diablo, and onward (imagine the possibility with gems modifying spells here.)


It’s actually a Roguelike. There’s a standard for judging how Roguelike-like a game is, known as the Berlin Interpretation (though not everyone agrees with it). Permadeath is just one possible aspect of Roguelikes. Random level generation, dungeon crawling, exploration, hack-and-slash, grid-based, are also important attributes.

An action-RPG is something like Deus Ex, SS2, VtM: Bloodlines, Fallout 3, etc.


I’m… not sure Traveller can be dismissed so easily as within the RPG category.

Also, there have been Traveller-based computer games. There were the two MegaTraveller games by Paragon (MegaTraveller 1: The Zhodani Conspiracy and MegaTraveller 2: Quest for the Ancients), as well as an AR game that didn’t get enough traction (which I was involved in testing). I wouldn’t describe any of those as more like an adventure game than an RPG.