Generation Game: Mark R Johnson on Ultima Ratio Regum

When I saw Mark R Johnson talk at the Norwich Gaming Festival a few months ago I knew I had to speak to him again in more depth. He is most certainly a man with a mission in the form of ten-year, open-world, roguelike project Ultima Ratio Regum.

Although roguelikes aren’t generally my bag, I can appreciate new ideas, creativity and vision regardless of the genre, and Mark Johnson has that in spades. The generation of vast worlds is no new thing, Elite managed something like it on a few kilobytes of data back in the day, but in order to create depth as well as simply generating breadth, he’s developing what he calls ‘qualitative PCG’ (procedural content generation, the act of programming a computer to generate believable content based on a rule set). The ambition is nothing less than to distil culture down to its core components and rearrange them like Lego. Not only that but to maintain a continuous thread of connections throughout it all.

Thanks to the success of indie games like Spelunky, which riff on the approach to gameplay established in the original Rogue, and ambitious projects like No Man’s Sky and Elite: Dangerous, procedural generation is a hot topic at the moment and is being explored as a means to solve content generation problems in the industry’s continuing love affair for massive open worlds.  Ultima Ratio Regum might not be as aesthetically crowd pleasing as No Man’s Sky, its roots are after all very much in the classic roguelike space, but its approach to procedural generation brings something incredibly new and ambitious in scale to the table none the less. Rather than planets, Mark R Johnson is dealing with something far more complex and illogical: human beings and the culture we produce.

Ready Up: Let’s start with that name – Ultima Ratio Regum – which I believe is a Latin phrase. Why did you chose that and how does it encapsulate the project?

Mark R Johnson: Excellent question, and one that I’ve never been asked before weirdly. It means The Last Argument of Kings and it’s a phrase that, in general, is used to refer to warfare. You have trade and diplomacy, but the final thing that only kings, or rulers, in general can use is war. Originally I had this idea that it would be an RTS game where you played as a single unit, rather than as an omniscient person able to view the entire battlefield. You would kind of have to control your armies from the ground rather than from the skies, so to speak. But that didn’t really work out and it didn’t really interest me. When I shifted onto what it is now, this cultural, puzzle, mystery, decipherment genre – of which it might be the only game – the title, in fact, is still quite applicable but maybe as a type of challenge to the concept. War is not the be-all and end-all, but rather being able to control history, or to write history, is more powerful than any army.

War is not the be-all and end-all, but rather being able to control history, or to write history, is more powerful than any army.

RU: Wasn’t that phrase etched on Napoleon’s canons?

MJ: Yes, it was Louis XIV’s canons, actually!

RU: What it’s turned into now is a roguelike, but it’s not the kind of roguelike most people (certainly myself) are familiar with in indie games like Spelunky and Rogue Legacy. Can you tell us a little about this independent roguelike scene, which uses ASCII art and symbols, and how your game fits into it?

MJ: Those are all what you’d call classic roguelikes. I mean there’s this really tedious thing at the moment of: “what is a roguelike?” Are games like [Binding of] Isaac or FTL roguelikes or roguelikelikes or roguelites, or something else entirely? But in general classic roguelikes have a few major differences from their more modern cousins. Firstly ASCII, clearly, which has a lot of impact, I think, on both the gameplay and how players make sense of that world. Player learning and mastery are quite different in those games. Also they have no meta-game aspect, so you don’t unlock new guns, new ships, new levels for instance, but everything is there from the start. I think that’s a lot stronger from a game design perspective. [Not long ago] I posted a rather controversial piece on my blog about that and that’s got a lot of discussion on Reddit and Twitter and NeoGAF about whether meta games, like those in FTL etc, are good for roguelikes, but that’s a different discussion. I personally played Nethack and Crawl and the rest long before I played Isaac, FTL and Spelunky. I enjoy both classics and moderns equally, but I think the older ones do some things well that the newer ones don’t.

Whereas a single play through of Isaac and FTL is an hour and a half, and Spelunky probably a bit less, a single play through, for a player who’s skilled enough to win, is something like 12 hours in a classic roguelike, or maybe more like 18 hours. And it’s still permadeath, of course, so I think you get this massive investment in the world and your character in that single play through. That really appealed to me a lot as it provokes a very different subjective, phenomenological feeling in players. That’s something I’m trying to reproduce. When there is a world as vast as there is in my game, there’s something neat about this idea that when you die there will be cultures that you never saw and there will be nations that you never visited. Likewise, even if you win, my intention is to balance it so that there will still be parts of the world that you’d never have time to visit.

RU: It’s interesting that the idea of permadeath is coming back, but in other genres like the RPG. I think The Witcher 3 has a permadeath mode and of course there is Dark Souls, which uses some roguelike elements. Do you see this as a trend that’s coming into mainstream gaming more?

MJ: If anything I think it’s a trend that’s going out actually. Well, hmmm… I think it may be coming in and going out equally, but in different areas. I think the fact that all the modern roguelikes are all still permadeath but are far gentler on the player is a sign that it is going out, or that it has to go out in order to appeal to a larger crowd, but on the other hand you do have things like the Souls series, which are not permadeath but you lose a lot with every death. I think they are designed, in a sort of roguelike way, to tell the player you can’t just level up to get better at the game (though you can a little bit) but you have to go away and ask yourself “how do I learn get better?” And that, I think, is a good change. So maybe it’s that permadeath in the genre we expect permadeath from is getting milder and milder, but the implicit philosophy behind permadeath is becoming more widespread at the same time.

RU: So perhaps we can talk about what you’ve already alluded to as the large scale of your game, which I believe is all procedurally generated. Can you go over a little what you’re actually trying to achieve?

MJ: [Laughs] You mean beyond intellectual masturbation? What I’m trying to achieve is a new type of gameplay, almost, where rather than having a roguelike about combat and monsters and dungeon delving and so forth, having a rogue like that is more… if you look at the classics, which are ASCII right? You have to do a kind of semantic detective work, almost, in terms of what each symbol means and how class structures are laid out. So everything with a lower case ‘a’ might be ants and an upper case ‘A’ might be angelic. In the classic roguelike genre there is a sense that you have to read the game, and the act of reading the game is less clear than in other games. In most games once you’ve seen one dragon you’ve seen every large, flying, fire breathing creature.

What you “do” is far more dispersed through the world and far more abstract than exploring a certain area, it’s more like exploring ideas.

What I’m trying to do is spiral that out into the entire game, so basically in order to succeed you will need to figure out the game’s cultures and religions and nations and social norms and cults and wars and histories and all these types of things. I think that’s an extremely fertile ground for procedural generation in general, since most PCG is dungeon layouts and so forth. What I’m trying to do is get the game to produce very detailed puzzles drawing on all this cultural data, and have those created anew each time so you can’t just learn “oh I see I press the button behind the kings throne and a secret door opens”, because that might not be there every game. Indeed there might not even be a throne or nations with kings on any given playthrough. What you “do” is far more dispersed through the world and far more abstract than exploring a certain area, it’s more like exploring ideas.

RU: Can you go over what types of things will be procedurally generated?

MJ: Nations, cultures, religions, social norms, cities, city districts, each building in a city, each item within a building, everything from chairs and tables to religious altars and thrones are generated, and those have their own clues in them. So you might read in a book that a king that sits in the jade throne holds item X and when you find him you know you have the right person. And then things like caravan and ocean trade routes, and roads, towns, jails, asylums, theatres, galleries, guilds, blah blah blah. Then on a smaller level things like novels, poetry and art works. I think it will be particularly intriguing when I set up the game to create paintings, which will also have clues about the world. So if you see a painting of a certain battle, and you’re trying to find out where that battle is, then by looking at the painting you can determine the battle took place in a tundra area next to two rivers with a city nearby. Then you look at your map and try and work out where that might have been. Right now I’m working on clothing and clothing styles and having those linked to broader aesthetic preferences and iconographies of each nation. As well as there being a lot of PCG stuff, I’m trying to weave in details which are tethered to a particular culture and then visually reproduced across many different things In that culture.

RU: How do you get the computer to create all that stuff but make sure it’s in harmony, or is that way too complicated to go into?

MJ: It’s a surprisingly tricky question! A lot of data – as in a large volume but not a lot in terms of memory. Each culture, religion or civilisation has a lot of high level data that needs to be reproduced, and the computer looks to that high level data. If it needs to spawn a holy book for a certain religion, then it will look at the nation that the book was printed in for how it should look and how it was bound, and then it will look at the religion to determine what the book says, what the title might be what kinds of symbols it will have. Everything that the game spawns, it will draw on every bit of data relevant to that thing. Every person, every item and so on cuts across lots of different axis and that makes for really dense connection between things.

Every person, every item and so on cuts across lots of different axis and that makes for really dense connection between things.

RU: How are you doing all this single-handedly?

MJ: Good work ethic. I do say that kind of flippantly, but at the same time that is probably the best answer. I’ve been doing it for three and a half years now and I’ve just about finished the world building part of it and am now moving into the gameplay, at last.

RU: You mentioned the game is a mystery, how does that aspect work?

MJ: Well, the game’s objective is that there are nine extremely strange historical items hidden throughout the world somewhere. But the world is about two and a half billion tiles down each axis, and each of those objects occupies one tile. So how you find them will be the basic quest. The basic idea is that you can follow trails of clues, which will all be generated by the game, but these trails will be hidden and embedded within the world’s cultures.

Let’s say one item is owned by a king and he keeps it behind his throne, and you are many, many miles away from him.  You might hear about a mythical type of creature that lives in the land, maybe you get closer you realise that the creature is the symbol of this nations flag, then you get even closer you find a bit of poetry about the creature and the object you’re after, and as you get closer and closer the data, plus visuals, plus conversations, plus everything else you encounter should be more and more specific about which tile, in which building in which district, in which nation this thing is stored. It’s almost like a [Dark] Souls game in that a lot of the game takes place in the players head, in the sense of working out what the game wants you to do and how the game “thinks” is a big part of the game. I think it will be interesting to see how skilled people get at doing this.

RU: You mentioned Dark Souls there and one of the things that surprised me the most when reading your list of future features for the game is a Dark Souls style combat system. Can you tell us more about that?

MJ: Combat I intend to be extremely rare, so even if you win the game you will wind up fighting no more than a dozen people in the whole play through. So it’s kind of like a classic roguelike with all the popcorn foes taken out and the bosses left in. The thing I love about the Dark Souls combat system is it’s about timing first and foremost, and you and the foes are very heavy and very physical. It’s not like Skyrim in that you stand there and hold a button and your guy swings an axe into their face over and over again, and they swing an axe into your face over and over again and whoever’s face is more axe-resistant wins. That seems to me agonisingly dull, but in Dark Souls every hit will stagger you, every hit will take a huge amount of your health, or can be blocked by a shield or dodged by a roll. So there’s no middle ground: either you take loads of damage or none. That very stark mechanic, where the armour and your weapon and your movement really matters, I’m trying to turn into a turn based system where different weapons will have different move sets – like Dark Souls – and combat will be like playing out a well-timed dance almost. I do have a very early trial version here in a separate programme, and although I’ve only worked on it for a week it is incredibly cool and different. Hopefully in a few months it will work extremely well.

It’s not like Skyrim in that you stand there and hold a button and your guy swings an axe into their face over and over again, and they swing an axe into your face over and over again and whoever’s face is more axe-resistant wins.

RU: That sounds fascinating. So players can play this now, but the game is part of a 10 year project. What can they expect to find in the game now?

MJ: Ten years is a safe bet, yes. So right now playing means wandering around the world. I don’t think I’m being too arrogant in saying that it’s the most culturally detailed PCG world ever made, as far as I know. I don’t want to use the term “walking simulator” but right now it’s “have a look around this super detailed world, which a game will shortly be built in”, basically. Which some people complain about, and I get why, clearly, since it’s a long time to wait, but on the other hand I am just one person, I am working extremely hard on it, and I can’t have the game until the world is there. I can’t spawn an NPC until I know what culture they are from, what gods they worship, what clothes they would wear, how they would respond to you. All this kind of uber world detail needs to be there to then create the game on top of it.

RU: It’s the ultimate early access experience I guess.

MJ: [ Laughs ] Something like that.

RU: You’ve just finished studying, has this been part of your PhD?

MJ: It’s not actually. When I started my PhD three and a half years ago for some reason I thought hey let’s give myself a second massive project! [ Laughs ]

RU: You’re very interested in the academic side of games and have spoken at a lot of conferences recently. How do the two halves of your identity – a scholar and a designer – balance with one another? How important do you think it is for those two worlds to influence and inform one another?

MJ: I feel very confident in saying that I’m a better game academic because I’m making a game and vice versa. I think those two things have enhanced one another and will continue to enhance one another in the future. In terms of the way I keep them separate, in many ways what I write about academically is not very close to my game design. What I’ve written about in games is mainly about player mastery and competition, which is related to roguelikes but not as much as things like e-sports.  And the post doc is about rules and glitches and rule breaking and creativity and cheating, and that’s not really a roguelike kind of topic. Roguelikes are generally glitch free and very rule bound systems.

RU: Do you feel the kind of experiments you’re undertaking will have a wider influence within game design? What kind of impact are you hoping it will have?

MJ: I think those seem like two slightly different questions. So am I hoping it will have impact? Absolutely. I hope that a few years down the line it will be as well-known as Dwarf Fortress, with a large player base and that a lot of people enjoy it. Naurally I wish that, or I wouldn’t be doing it. I hope it has a lot of impact as an indie game/freeware game that is extremely different, extremely weird. Because the central mechanic is designed to be so different, I hope that has some impact in moving games away from combat, whilst still making them extremely challenging. It’s hard to imagine a game being challenging without that coming from a combat mechanic, and that’s not depressing so much as “ha, why is that the case? And couldn’t we make some other kind of challenge instead?” So that’s kind of what I’m trying to do there.

The other part of your question, in terms of whether it has more influence than impact. A lot of people have asked me if I’d open source it and I’ve said no, but I will write that into my last will and testament. But until then no one will be getting their hands on it and that goes back to the impact thing. In terms of PCG systems this is the only one that can create cultures and I kind of want to keep that secret, because that’s the thing that makes it distinctive and unique, and if everyone knows the code for it then it loses some of the noteworthiness.

I do kind of adhere to the auteur theory of authorship and I’m not a great believer in the death of the author, and by contrast I think the author’s intentions do matter a huge amount. I don’t want to dilute it by allowing it to be modded. I want it to have some impact but I don’t want it to be copied. I suppose if it is successful then it will inevitably be copied, because if you look at things like Dwarf Fortress and Minecraft we see the endless litany of games like Game of Dwarves or Mine of Thrones or Thronecraft and other meaningless, vapid, identikit garbage. My apologies to those who made those games, but seriously – how can you live with yourselves? I hope it gets people thinking about topics and has impact in its own right, but “influence” matters much less to me than impact, because I find it hard to see how it could be influential without it being copied. Only time will tell!

You can read more of Mark’s writing and follow the development of Ultima Ratio Regum on his website Mark R Johnson will also be hosting the International Roguelike Development Conference 2015 at the National Videogame Archive 27-28 June