Brendan Greene, known to gamers simply as PlayerUnknown, hardly needs much of an introduction, but we'll do it anyway on the off-chance you've been under a rock these past few years. Once a photographer and freelance web designer, PlayerUnknown was arguably responsible for spawning the Battle Royale genre when he started modding DayZ (itself an Arma 2 mod) to add a last-man-standing mode. His first work was DayZ: Battle Royale, inspired by the Japanese movie Battle Royale.
Later in 2013, Greene moved to the newly launched Arma 3 and released another mod called PlayerUnknown's Battle Royale. This mod introduced for the first time the now ubiquitous feature of dropping into the map via airplane. The success of his Battle Royale mods caused professional game studios to take notice of Greene. The first one to do so was Sony Online Entertainment, which hired him to create a Battle Royale mode for its zombie survival game H1Z1. The mode was called H1Z1: King of the Kill. But it was with Krafton (then still called Bluehole Ginno Games) that his vision was finally realized.
CEO Chang-han Kim brought him to South Korea as Creative Director for a new project that would ultimately become PlayerUnknown's Battlegrounds, or, as it is more commonly known, PUBG. The rest, as they say, is history. The game launched in late 2017, literally taking the industry by storm and establishing Battle Royale as the biggest gaming trend. According to estimates, it is the best-selling game ever released on PC with 42 million units, and on Xbox One with 9 million units. Across all platforms, it is estimated to be the fifth best-selling game of all time with 75 million units, behind only Minecraft, Grand Theft Auto V, Wii Sports (which was bundled into the Wii console), and Mario Kart 8/Deluxe. On Steam, PUBG still retains the top spot in the all-time concurrency peak chart with 3.2 million users playing at the same time.
Over time, the game's immense popularity diminished in favor of Epic's Fortnite. However, PlayerUnknown had left his mark and was about to move on to his next and even more ambitious project. In early 2019, he relocated to Amsterdam to establish a new studio called PUBG Special Projects, revealing they had a blank slate to explore new concepts with no strict deadline to follow. Later that year, Greene began sharing that he wanted to create a massively big world, way beyond current standards, in a game called Prologue.
We didn't hear from him for a couple of years until it was announced that the studio had been spun off into the independent PlayerUnknown Productions, with only a minority stake from Krafton. Later in 2021, Greene revealed some adjustments to the plan: Prologue would be a tech demo for a subsequent planet-sized game called Artemis. That wasn't the end of the changes to the plan. A couple of months ago, the studio said it had switched to a three-game plan that started with Preface: Undiscovered World, a free tech demo of the Melba engine now available on Steam.
Prologue: Go Wayback! (also available for wishlisting on Valve's platform) would instead be a single player survival game where players would explore a detailed world where emergent systems influence gameplay, facing unpredictable weather events and ever-changing landscapes as they attempt to escape the untamed wilderness. Finally, the third and final game that should realize PlayerUnknown's vision is codenamed Artemis, a massive multiplayer sandbox game that is, however, still far away, given that the updated plan has a 10-year roadmap.
I recently had the opportunity to have a nice, long conversation with Brendan 'PlayerUnknown' Greene himself, where he opened up to discuss the games, the studio's plans, and his view of the metaverse, which is starkly different from others we've seen so far in the industry. You can read the full transcript of our chat below.
You've had a big impact on the Battle Royale genre. Now, you're doing something completely different, and you're building your own studio. How is that going so far, and what was the drive behind this new initiative for your own career?
PlayerUnknown: The drive was - the reason I got back into gaming was that I fell in love with games with the DayZ mod and ARMA 2 and that idea of this emergent survival space, I really fell in love with. But the thing that I didn't like was just the scale of the world, that there was always an edge to a map.
We've been kind of trapped in these small 20-by-20 or 8-by-8-kilometer worlds for quite a long time. After PUBG, I thought, well, I'd like to do a 100-by-100-kilometer survival game. I thought that would be interesting. You could have trade routes with a world of that size. Imagine Rust on that kind of scale.
We found the way to do a 100-by-100 meant that we could do any scale, because doing it by hand means involving a lot of artists, and it takes a long time to do everything handcrafted. At the time, we had a research director who had a PhD in machine learning, and he thought this would be a good way to generate these worlds rather than doing it in a more traditional way, because then you could do them in real-time. I was told at the time that doing worlds in real time wasn't really a serious thing, but now, four years later, we have a demo on Steam to show that I'm not so crazy, thankfully.
But the whole drive was just to create these kinds of emergent play spaces that aren't bound by scale. They should be able to scale up to planet size as we've done in Preface. But ultimately, it was always just about creating these bigger, emergent spaces where players can get together and do stuff, play games or interact in other ways, but really in a shared environment. I loved playing DayZ and then having that sort of camaraderie with your squad and never knowing what was happening. Trying to recreate that on a massive scale is just something that excited me back then and still does now. The vision has kind of moved on a little from then, but that's where it came from.
How long have you been working to realize this vision?
PlayerUnknown: PUBG Special Projects started probably 3 or 4 years ago now. Oh, no, it's more than that, five years, because we have people in the studio that have been with me for five years. So it's been a long time, but as an independent studio, just over two years. We had our 2-year anniversary last October, so we are into year three now. It's been 5 years since the idea was established and I started building a team. The core team we have now is pretty much complete. We have 60 people on the team; it's a good, solid studio. It's taken some time, and I've made some missteps along the way, but not having a huge amount of experience when I got into this, it happens. Thankfully, I've met some great people in the last year and a half who have really helped stabilize the idea and get it into something that we can actually do.
Did you experience any trouble managing a growing studio while also developing the project?
PlayerUnknown: Of course I did. I was a photographer, graphic designer, and freelancer who was then given a team to build a game. I then did PUBG and a press tour for about a year and then got some funding to do this project. I'd never really run a studio. I relied on the intelligence of others, and I made some bad decisions with certain hires in leadership positions, but thankfully, after meeting David Polfeldt from Massive, he became a good friend and really helped me see where I was doing stuff wrong and where I could improve. He also helped me hire some great people in leadership roles.
Now, we have Kim Nordstrom as the CEO, Laurent Gorga as our CTO, and Scott Davidson and Petter Sydow on the game team as producers and creative directors. These people really believe in the vision and understand what I'm trying to do. They drilled down our three-game plan.
We thought this was how we should do it: by taking the tech we needed to solve at the time and building games around that. They came back with a plan very similar to that, so I was happy to hear I wasn't super crazy. However, I needed that sort of leadership team that could implement a plan to make it happen rather than just telling me it's fine and never really moving forward.
Do you have the full team over there in Amsterdam, or is it also kind of hybrid or remote?
PlayerUnknown: I'd say we are hybrid more than fully remote. We have a studio here in Amsterdam that accommodates about 60 people, but we have maybe fifteen of those working remotely. Some work remotely from Amsterdam and other cities in the Netherlands, but others work from other European countries. We hire mostly from Europe because we want to be in the same time zone. With the management team, we've come up with this seven-week sprint, and it's happening this week actually, where we bring the whole team into Amsterdam. We also have a planning week where there's no work done. We all just get together and figure out what we're doing for the next sprint and what's achievable.
We're trying to come up with a system that works for us because I think that fully remote is not ideal, especially when you're creating an idea, though some teams have made it work. We're really trying to make a system that works for us, and I think the seven-week sprint that we have with one week on-site, and again, the people who live in Amsterdam can be in the office if they want to. It's been a challenge to find the right process, but I think we've gone on to something that's pretty good now.
I think one of the key components of the whole project is the engine that you're making. It's interesting because lots of AAA developers are actually moving away from their proprietary engines to Epic's Unreal Engine 5. Then again, they are making traditional games that are not nearly this large. To make a planet-sized game, you definitely need a proprietary engine. I only know of another game that is attempting something like this, which is Light No Fire from Hello Games, also based on their own proprietary engine. Can you talk about the Melba engine and what are the technical challenges of trying to make such a big vision come to life?
PlayerUnknown: It gets back to why I'm doing this. I hate using the word metaverse, but I have heard a lot of talk in the last few years about what people are vying for the metaverse.
You see, Epic is trying to make the metaverse with Fortnite and all these other companies are making their metaverses, and I never really understood them. It just looked like IP bubbles to me. It didn't really seem like anything like the metaverse I thought in my head should be, which is an open digital place. It's like a 3D internet with worlds instead of web pages, and you can just travel between them. I never really saw that happening there. With the way people were talking about it, having a server-client infrastructure, it just doesn't work. You can only have maybe a few thousand or tens of thousands of players together rather than millions of players.
The whole aim of building this engine is to create this world generation platform engine that is going to be open source, and that's why we have left Preface relatively open. You can already mod it to hack the fly speed or the travel speed and shrink the world size if you want.
We're trying to be as open as possible with everything we can and trying to build it with the community, because I think for something to power the metaverse, whatever it is, it has to be open source. It can't be managed by a corporation. It has to be like HTTP, an open-source protocol that everyone can build on and create 3D worlds.
That's what we're trying to achieve with Artemis, and that's why it's 10 years away. We're trying to solve many steps, like building Earth-scale terrains locally first and then adding lots of interaction to that. Again, we're trying to keep it as local as possible and adding identity and marketplaces and all this stuff on top of that.
This is why we have this three-game plan. We recognize that each of these systems is a big challenge to solve. If we take them in small chunks and build on top of each other, I think we have more of a chance of achieving this world creation engine that will hopefully provide some kind of metaverse that isn't hardware-locked.
Kids in Africa should be able to access the metaverse as easily as kids on the West Coast. I don't see these things with the current iterations of what people are calling the metaverse. To me, it looks like strapping rocket engines onto a horse to get to the moon instead of rethinking how you do it and building something new. That's why we're building our engine, if that makes sense. And I hate to use that word, metaverse, because it's been corrupted. To me, it's a fifth place to go. I can create a world and what we're doing with Artemis is like a wilderness. It's like Minecraft survival, right? It's that layer. And then the engine is just open for you to create whatever world you want, and then all these worlds just work together, like a multiverse of worlds.
You should be able to create your own character and dress them however you want and not be bound to just play as your favorite comic book hero. There's so much creativity out there that needs to be given a platform to happen, but I think people are just still thinking in these small bubbles and not about truly big spaces where lots and lots of people can interact together.
It's an interesting subject, for sure. Personally, I've always been a big MMO fan. The way I see it, it's going to be like a massive virtual world, right?
PlayerUnknown: Yes, but I don't want to tell people to do something. I can leave systems in place that they can, resurrect the town and create borders and do stuff with the land that you take over by doing something in that town, or you do nothing and just wander around and see things.
I don't want to say this is the game. It's just a world or a world of civilization; mechanics are there if you want them. But if not, you don't have to do anything. You can just run and explore. With Preface, we have this fun thing called Deplex. You can be anywhere in the world, then press F5 or F6, and it brings up a deep link panel where you can just create a hyperlink that brings other people to that point in the world.
I thought, isn't that kind of what an HTTP link is? A hyperlink is just this, a link to a place in the virtual world. So you can go exploring by just friends sending you links rather than having to physically move in the world. You can do that if you want, of course, but it just gives that sort of freedom of exploration to everyone, and I think for worlds of the size we're building, that really needs to be necessary.
You mentioned that the world is going to be very large, but you are also aiming to have lots of people eventually online at the same time, on the same server. What's your solution to that? So far, in a 3D world, I think the maximum amount is like 10,000 or 20,000 players at once on a server. If you want to go higher than that, you have to create a custom solution.
PlayerUnknown: Exactly. And the problem is the server. There's hardware. Well, actually it's not the server as such, it's where the simulation is gone and where it's validated, and that's the issue because you're passing so much data around.
What we've done with Melba is that we do the world simulation locally so that the world is generated locally on your GPU rather than relying on servers to do it. Every agent and every hardware will generate the same world. We know then we can do stuff via ML agents without servers. I don't think servers are a solution to the metaverse or to creating millions of people in the same space. I think it has to be peer-to-peer.
We know we can do secure peer-to-peer with stuff like BitTorrent and blockchain as well, but it's more that peer-to-peer with a hybrid solution, where maybe we have peers that maybe you host, or a service provider hosts, where they do a more taxing simulation like weather or physics, and then people on lower end devices just get sent the data to recreate it on their devices.
But I think as long as the simulation is done locally before the network tick, then all you really have to share is player position and not much else, and that opens up to doing, hopefully, millions of concurrent players. But I don't know yet, because I'm not the tech guy. I have a dream and I try to pursue it.
But really, I think we have to do some kind of hybrid peer-to-peer system because I think doing server-client, as you said, there's physical limits there to how many players you can get. And I think chasing that server client... I think you still need that for certain game modes that will take place in the world. I see it just like sharding off that, okay, there's an FPS game happening here, and we need a different network protocol for this. But for the main world itself, it's peer-to-peer. I think that's probably the only solution to get to the kind of player numbers we need.
Yeah, that makes sense. Developers are usually wary of bringing stuff from the server to the client, as there is a legitimate fear that the player may exploit it to cheat.
PlayerUnknown: For sure, but this is how you look at what we're building, right, because we're not building a game; we're building a world. There will be cheating in the world, but I think the core here is how you offer moderation and how you empower the community to make sure that people who are cheating are easily banned.
And I think it comes down again to one of the core problems we have to solve, which is identity. How do you build a solid identity system that allows people to know that there's consequences to their actions? If you act like an idiot, communities can kick you out. It's like the real world, right? If you go into a supermarket and start shouting slurs, you're going to get kicked out pretty quickly.
There's not that sort of barrier yet online, where people just have anonymity, which I still believe is super important. I don't want to get rid of that, but I still think you need to have a system that ensures consequences for actions. Again, it's about empowering the community. The Internet doesn't ban people; communities ban people, but ensuring that players do have consequences for their behavior, I think there is your solution. But again, it's one of the big problems that we have to solve, and there's many other smarter people than me working on it. So, I don't want to go too deep into it because then I might sound like a fool.
Do you envision people potentially putting others in virtual jail or something like that?
PlayerUnknown: I thought about this and I thought for moderation, if you're running around shouting slurs all the time, then we can just make sure no one sees or hears you, and make sure you can't interact with anyone. You can still just float around the world like a ghost, and you can see what you're missing, but you can't interact with anything because you've been labeled a f*cking idiot. There's probably many creative solutions, but again, I think my core priority here is putting tools in the hands of the community to allow them to do it.
Because ultimately, if I'm providing a platform for something like the metaverse, then there can't be an authority banning everyone. It can't work like that. It has to be community-driven because that's what great platforms are.
It sounds very sandbox-driven. It's clearly a sandbox-inspired game.
PlayerUnknown: I love what Garry Newman does. I love what he's done with Rust and the whole dev mentality that he has. Facepunch is just great, so if we're compared to that, it's fantastic.
Let's talk a bit about Preface since it's out now. How was the community's reception?
PlayerUnknown: In our Discord, it's been great. I have a message from a CTO saying thank you for, like I've never felt this feeling before where the community have taken apart your demo and shrunk the world or made Millennium Falcon speed when moving or they're now adjusting how the hey like when we're switching between waypoints, how that camera movement and and working with the community.
It's been really lovely, seeing that the devs on the team are now posting challenges to the community and going, can you do this? It's really exciting to see people who really get it; they understand what this is. Yes, it's a bit empty for now, but they're willing to work on it and mod it and then try to see what they can create already, which was one of my hopes. It's meant to be an open platform, and it's quite empty right now.
I shouldn't call it a game. It's not a game. It is very much a tech demo. I say it's almost like the early Internet. It's empty right now with very few biomes and not very much to do or see. But I hope people can see the possibilities. If we can generate worlds of this scale for 3.6 GB of space, I think that's super exciting for what creators can do in the future once the tools are there.
I'm reading that it will be in early access for about a year, correct?
PlayerUnknown: For Preface, this is an engine demo. It serves a purpose, which is that we'll keep iterating on it. It may stay in early access for a year, but it'll just keep being iterated as we keep improving the engine. There may be a full release, but I don't see it like that. It's an engine product rather than a game product. It'll always be iterated on. This is the visualization of what the engine is capable of, and as we improve the engine, the demo should also improve.
Right now, it's free. Once it leaves early access, do you plan to add a price tag to it?
PlayerUnknown: No, it has to stay free. It's like the HTTP, right? It can't be charged for. The engine for the metaverse has to be free and open for everyone. That's our view for Preface. Having someone just download it and run it on their PC is fantastic because it just gives us more data as to how we can improve the engine for everyone and what we have to take into consideration. I don't ever see us charging for preface because that's not the point of it. The point is to have an open source engine for everyone.
Moving on with your plan, then there is Prologue, right? That one is a bit different because it's going to be a single player game, which is a bit new for you, I guess.
PlayerUnknown: Yeah. Again, it was more about how we leverage what we're doing and create something simple to test, right? I like survival games. When I first started, we discussed the possibility of doing a new map every time you press play. Now, after plugging our machine learning tech into Unreal Engine 5, we can do that. I think it's like 4.2 billion maps, mathematically. Maybe a million or a few million of those are interesting maps.
But the team has enjoyed exploring and discovering how it all works and how these maps are created. It's super exciting to see the progress. We're discovering how this tech works, how this new way of creating terrain can bring us advantages, and what we can do with it. We've just launched playtests on our Discord, where about 200 people are playtesting a very early demo of the game. It's broken, but people are having fun. As I said, I want to build games with the community, not for them. These are our very early steps in doing this.
I keep mentioning the three-game plan because it's about solving the three core problems: building the world, populating it, and filling it with interaction and millions of players. Each of these three things is being tackled by a game. Prologue is tackling the terrain where we generate new terrains every playthrough by using our machine learning tech. That'll help us prove out and mature the ML tech. Then, when we go to game two, which is hopefully on our own engine, we can have that ML tech matured, go bigger, and keep scaling up. That's the intention for Prologue. It's quite a simple game. It's not my next opus, but it serves a purpose, which is that of a testbed for our ML tech.
I still want to make a great single player experience and I'm excited about emergent gameplay in a single player world, kind of like Don't Starve and these games that are unique in every playthrough. Hopefully, we can create an 8x8 world that looks pretty good. Having a unique world every time, I think, gives us something interesting to play with.
That's why I want to work with the community as well, because I don't know how exciting this can be. This is so new and breakthrough that I want to work with them early to get feedback and think, okay, what can we do to make this a better experience from the get-go rather than trying to put band-aids on it like 6, 7 months in the future.
You mentioned that Prologue will have some survival mechanics. Is that going to be fairly traditional or are you going to maybe introduce some twists to the formula?
PlayerUnknown: Yes, we're not trying to reinvent the wheel. We're trying to make it not realistic, but as intuitive as possible, right? Light fires, find food, eat food, get across the map. As I said, the survival mechanics are not the point of this. I think it'll be interesting what we put into it. But I think the point is to test the ML terrain and make it feel good, make it feel realistic, make it feel playable. That you're not always losing because of something the terrain has done. I don't plan on reinventing survival mechanics here because that's not the point of the game.
Let's say the community asks for a co-op mode for Prologue. Would you be interested in adding it, or do you see it strictly as single-player?
PlayerUnknown: I'm open to anything the community wants, but we have to make sure that we have the resources to do that. I'd love to add co-op because I'm sure I can do it, and it's interesting, but it means adding a multiplayer layer to something we haven't planned to do. If we enter early access and there's a good reception and people are super interested and the game is doing well, then maybe that allows us to scale up and think about it.
But it would be diverging from the purpose, which is trying to make this a solid early access product first, and then we worry about that. But we do plan to release updates every 6 or 9 months or so. I'm not really sure about the timeline yet because we're still working on getting the product ready. We really want to use Prologue as a testbed, not only for the terrain but also for the minute-to-minute, hour-to-hour stuff that you would do in the survival layer of any game.
Maybe we'll scale up the map eventually, maybe we'll add more biomes. There's a lot of stuff we can do over the next few years of Prologue to help us get ready for Melba coming online, and we can learn a lot from the data we get from Prologue.
What's your opinion on machine learning for NPCs? I don't know if you've seen the Inworld demos.
PlayerUnknown: I think micro LLMs and LLMs for NPC interaction could be interesting, right? I want to be careful. I really think that when used correctly, like we're using ML to do our terrain, that's all trained on NASA data or open-source Earth data. I think as long as ML is used in the right way, it can be super useful. It's useful as a tool to speed up an artist's hand or a writer's drafting, but I think using it just to fully replace storytellers and creatives is not the way we should be using it, because I really don't think it's there yet. I think if you're in any way good at anything, you can see the mistakes it makes, you can see that it's not really good. It serves as a flash in the pan, sure.
But I think when leveraged right, it's a very good tool. We're using it at an engineering level to speed up simulation and do simulation locally, and it's very good.
I understand that the platform of reference is, of course, PC, but do you have plans to even potentially release them on consoles, whatever they may be eventually?
PlayerUnknown: Well, Artemis has to be platform-agnostic. If it is a digital world platform, then it has to run on everything from phones to consoles to PC to VR to anything. At least, that is how I see a platform like that existing. It can't be bound exclusively to high-end hardware. It's just not like that. Even Preface already runs on pretty average hardware, I would say, not super high end, but it still works and you have ray tracing and it looks pretty good.
The idea is that we will scale. The world doesn't necessarily have to look exactly the same for every player. It can have lower LODs or lower textures for one player and other players. As long as you have the interaction, I think it's fine. But when you move into game spaces within the world, then you have to think about these things a bit more carefully. But ultimately, I don't want this Artemis as a platform or that is a world to be bound to any particular hardware because that doesn't serve the purpose of the metaverse. It has to be an open source platform that everyone can use anywhere.
You said that it's a 10-year plan, right?
PlayerUnknown: Probably 15, but yes.
Are you considering this timeframe from when you started the project?
PlayerUnknown: From about a year ago. I would say that that's when we really formalized the longer timeline and it came to about 10 years. But that may be quicker depending on whether we solve some of the problems sooner and depending on how the games do. It's trying to be realistic about what we're trying to achieve, which is pretty massive and has to be done the right way.
I want to give the team the freedom to do it the right way rather than doing it in a way that serves to make us money because then I don't think you ever get the Internet, or the metaverse, right. It has to be done with the right spirit or else it just ends up as a monetized piece of garbage. It takes a long time to do something like this.
Since you mentioned monetization, of course, it's very, very early days, but what do you envision monetization to be like in Artemis?
PlayerUnknown: I think it just has to be a digital world. There'll be stores where you can buy stuff. I think for me, as a platform holder, I'll have just a financial services fee or something like this, that if we provide a financial layer and you use that to sell goods and services and do stuff in the world, then we'll just take a small tax on that.
But I don't think Artemis is a platform that will be monetized by me, because that's not what it's meant to be. It'll be an open platform for everyone to come, you know, hawk their wares, rather than everything coming from me because I don't think that that's the metaverse. The Metaverse is everyone comes, everyone can do work and get paid. Buy other stuff.
So players could sell an item for real money.
PlayerUnknown: Yeah, they have to, right? I mean, that's the metaverse, at least in my head, the one that we've been sold in books and stuff like that. I just see that the current metaverse is everything Ready Player One warned us against, right? It has to be something more open and for everyone and done with the right spirit, or else it ends up like AOL. I think it was a stab at monetizing the internet.
Do you plan to use blockchain to track transactions and everything?
PlayerUnknown: I don't know yet. Blockchain, hashgraph, or whatever other decentralized ledger solution, it's an interesting financial layer for sure, but we really haven't thought about it that seriously. I want to get the world working first.
Even getting a marketplace system that works doesn't need blockchain. It doesn't mean this. It needs a lot of time and economists to try to figure out how we build a marketplace system that works for millions of people trading. That alone is a challenge. There's lots of interesting tech out there, but right now, we're focused on building the engine rather than worrying about anything like that.
I don't know if there's anything else that we haven't touched that you can or want to discuss today.
PlayerUnknown: I would encourage you and whoever is reading to join our Discord. We have a pretty good community, still quite small but really active. We're doing play tests in there with Prologue already, and we're scaling up pretty well. If you want to get involved and see what we're doing, come join and help us build something great.
Thank you for your time.
About the author: With over two decades of experience in gaming journalism, Alessio Palumbo has led the gaming vertical at Wccftech since August 2015. He started working at a young age for Italian websites like Everyeye.it, Gamestar.it, Nextgame.it, and Multiplayer.it before kickstarting the indie English-language publication Worlds Factory as its founder and Editor in Chief. In the last decade, he has coordinated the overall output of Wccftech's gaming section, managed PR relations, assigned reviews, produced daily news coverage, edited gaming content as needed, and delivered game reviews. Arguably, his trademark content is the long series of exclusive developer interviews that have been cited by Wikipedia and by the biggest news media and gaming publications. His passion for technology also makes him knowledgeable when it comes to gaming hardware and tech. His favorite genres include RPGs, MMORPGs, and action/adventure games.
Follow Wccftech on Google to get more of our news coverage in your feeds.
