Uber's Jon Mavor Explains Planetary Annihilation
“a million units in a game”
Planetary Annihilation was undoubtedly one of the strongest Kickstarter projects to have appeared since Double Fine set off the goldrush. A stellar pitch, a cap doff to Total Annihilation, and the tech lead from Supreme Commander pouring all of his ambitions into it: few games can boast anything like this, and consequently this game threatens seismic effects in the RTS genre. The man making that happen is Jon Mavor, and I thought it well overdue for us to catch up with what he was up to, and why he is making one of the most exciting games of 2013.
RPS: So I'm really interested in how you came to lead this project. I know a few people have been a bit “hey, that's not Chris Taylor” when they've realised the heritage of this project, but you have been heavily implicated in all these games over the years – what's the story?
Jon Mavor: I've been involved with these games as “the guy behind the guy for a long time”, and mostly that's because I've been on the engineering side. I've honestly avoided being in the public eye. If you look at the Monday Night Combat stuff we did when we founded Uber, you really won't see my name come up that much. I've honestly avoided the spotlight! But with Planetary, well, I am really passionate about making these kinds of games. I realised that with this new era of funding, with Kickstarter, and even in general, it is important with the person who is making the game to be really involved with the community. I decided to change my tack a bit, and come out of my shell, talk to the press. I've tried to hide a bit – and that's why you've never head of me!
Are you actually interested in the story of how I got here?
RPS: Absolutely.
Mavor: Well, back in 1996 I was up in Canada and I had just finished the first game project I ever worked on, which was Radix: Beyond The Void, and me and some friends created that title. It didn't really make any money, but it was a great lesson in making games, the business of games. It was apparent, though, that my high-school buddies weren't looking to make a career out of it. They've all gone on to do really well, and I was still friends with them, but it was clear that I was going to move on.
So I went down to GDC in '96, and I met Chris Taylor in the bar. They'd given us these plastic guns and we had to go and find and “assassinate” people, and I ran into Chris during that, and he was recruiting a team for Total Annihilation. He'd just come down from EA Canada, and of course he couldn't hire any of the people there to make TA. So he basically had to find a team of new people, and he is pretty good at that, and happy to take some risks. I was twenty-one with a speciality in graphics programming, and pitched him my great big game idea – the sort of all encompassing Battlefield thing that wasn't really possible in 1996 – and during that he asked me to come and interview in Seattle in April, and that was Chris, Clayton and Ron Gilbert, who was exec producer on TA. They offered me a job.
Now, as a Brit, you might be familiar with getting into the United States? Well, they were all “you just need to go get this visa”, and it turns out it was a bit more difficult. I had a long gap where they wanted be on the team, but I wasn't able to because of the visa stuff. I finally got down there at the end of the summer. What I'd seen of TA's rendering tech was really cool, and I really liked that, and wanted to work on it. During the time when I accepted the job and got the visa I got other job offers because of my Quake editor, Thread – old news now, but it was kind of a big deal at the time - but I was like “no, I want to work on this crazy Total Annihilation game!” Even with just being shown some white boxes driving around on terrain, what Chris told me was so compelling that I had to work on that game. So that's what I did.
Anyway, as TA was winding down I moved to Cavedog and worked on an FPS, Amen: The Awakening, and there's a whole story you could do about that, but at the end of the day I worked on that for a couple of years, and Cavedog got cancelled. So that was tough.
After that I went over and worked in console technology for a while, which was incredibly useful to me in terms of learning skills and working with new technology, but I started getting bored around 2003. I pitched to that company, Amaze, and said that we could make a really good RTS title. I said we should go to Atari and pitch to make Total Annihilation 2. Unbeknownst to me there were a massive number of people who had the same thought, and no-one was successful. We had a number of meetings with Atari, but we weren't successful at all. I felt pretty sad, because I knew I wanted to make an RTS, and to make a TA style game. And there was a huge TA community!
There's a lot of unfinished business in the RTS genre, right?
RPS: Yes, there probably is.
Mavor: So I leave our meeting with Atari at GDC, wander down to the show floor, and who do I bump into? Chris Taylor. He asks what I am doing, and I say I'd really like to make another RTS game, but it doesn't look like that will happen. He looks at me and says: “Why don't you come work on Supreme Commander?”
I went to GPG to work on that game. I didn't know what had been going on with that game, especially because there was a good gap before I could transition to GPG. I had a lot to sort out. I was hired as the graphics programmer, exciting because I can concentrate on graphics, rather than worrying about the management stuff from my previous job. But Supreme Commander was a little off the rails when I got there. There was a lot of stuff that was an impediment to what we needed from an RTS game. The guy who was making with the engine as soon as I left as I got there. Literally he left on the Friday and I got there on the Monday. There was a lot of work to do. We rebuilt the engine, and rebuilt the team. I ended up leading the engineering team, rather than just concentrating on graphics, which was unexpected.
Anyway, we went through a lot – including a cancellation! (I don't really know what happened with that) - but we had a lot of fun. Now, I never considered myself a game designer, I came in from organically growing something and doing interesting things with technology. But I started to get interested in guiding the gameplay, and I clashed with the design side fairly often. I wasn't terribly enthused with the game that came out. It could have been a lot better. We fixed a lot of things with the expansion pack, Forged Alliance, which is the high-point of this kind of game, especially if you talk to the community – tonnes of mods and stuff like that.
RPS: It's certainly a high point for RTS games generally. What happened next?
Mavor: So I left GPG and 2008, with Supreme Commander being off the table. And, well, I wanted to work on some other types of games, even if that interest in RTS games remain. If you look at Monday Night Combat, you can see elements of RTS in the game. But I did not design that. Anyway, we started Uber, and wanted to work on different stuff. But I still wanted to make a spiritual successor to TA.
RPS: And that was Planetary Annihilation?
Mavor: I wanted a different take on it than Supreme Commander. It's my version of what the spiritual successor to TA should be. Having been exposed to that environment for so long, I knew there were a few things that no one had tried yet. Trying something with multiple planets seemed like the obvious way to go. Also there was this game called Risk 2210, I don't know if you have ever seen it?
RPS: I've not played it myself, but I've heard the stories.
Mavor: It's basically boardgame Risk, but it has this moon which is another play field, and which interacts with the rest of the game. I saw that and was all “hmm, I wonder if you could take that, take orbitals, and build that up as a platform to attack the planet below?” So that was the genesis of Planetary Annihilation. I was also really interested in Jon Ringo's series which begins with Live Free, Or Die, in which they have solar system combat with giant battle-moons and stuff like that. And I thought: “let's take the RTS into this 3D realm, smash asteroids into planets, and do big interesting things that I've read about in these books, or played in other games.” Total Annihilation combined with expanding the playfield up to planetary scale.
RPS: Your Kickstarter pitch was super strong. You seemed to know exactly what you were doing. Did you know what you were doing?
Mavor: Yeah, I think we knew what we were doing. Bob Berry, the CEO here at Uber, came to me after Double Fine Adventure and said “I really think there might be something in this Kickstarter thing” and we're not a company that works with publishers – we're self-funding, so it made some sense. We're not sitting here with a pile of cash to make whatever game we want. So Kickstarter was great not for funding so much as for validation: Is there a market for a TA-style RTS? It wasn't clear to me that there were enough people who cared about the product. We had to prove it, and Kickstarter was a bunch of things coming together.
I think you need three things for success on Kickstarter. We had this game idea which fits what works on Kickstarter. If you look at all of the really successful game Kickstarter projects they hook into a previous game or genre that people understand and appreciate, but that is being under-served. Double Fine Adventure – who is making adventure games? A millions of people played them in the past. Obsidian's RPG, you know what you are going to get with an Obsidian RPG. Hook into nostalgia, and you are going to be really successful. Star Citizen is “I am going to do another Wing Commander”. It's a no-brainer.
But you also need a proven team, and plausibility. People have to believe you can pull it off. And finally you need a great pitch. We felt like a two-minute gameplay visualisation would be the best pitch. We had those three elements in place, which was enough to be successful. I would define successful by the way as raising a million dollars on Kickstarter, if you can do that, you are pretty successful.*
If the Kickstarter had not succeeded, we would not have been working on this game. You, the gamers, decided that this guy would be made.
RPS: How did the studio respond when you blasted past your target?
Mavor: (Laughs) Well, I think they responded pretty positively! I know I did. But I am one of those guys who feels like if you do the execution right then things will generally work out. I wasn't super-super shocked, and we even had a pool of guesses about where we're going to end up, and everyone guessed like two or three million. And that was because we'd seen the pitch video developer, I am not going to lie, we knew it was hot. We saw it and said “yeah, this is pretty good”. It's the steak and the sizzle. The sizzle is that eye-catching gameplay, and the steak is the second half of the video where we really talk about the project. And it all came together, so we were ecstatic. Fifty thousand people on Kickstarter is an amazing opportunity. That's very motivating. If you ever need some fuel in your tank, you just have to remember how many people are waiting for this game.
Of course you then enter the phase where you have to knuckle down and make the game, which is the phase we're in now.
RPS: Earlier in this interview you said you felt there was “unfinished business” in the RTS genre – what is that?
Mavor: The promise of this kind of game has always been that we're going to get something on a crazy huge big scale. And if you look at where computers are now, versus where they were in the TA era, well, my TA machine was at 32mb of RAM. It was hot! Look at what we have now – why aren't we playing these huge RTS games with crazy battles with loads of people? Where are those games?
Also I think the spiritual successor to TA should turn out a little different to how Supreme Commander turned out, I want to provide another take on it. From a game design standpoint, and a tech standpoint, it seems like the right time to try and do something.
RPS: There's no campaign mode, is there? This is straight up robot army warfare – I would say “skirmish” but that sounds too small to be accurate... Do campaigns not matter?
Mavor: I think people really care about campaigns. People want campaigns. The issue with campaigns is that they are unbelievably expensive to build. You have to build the game first, and then you have to build on top of that, and then you have to bug fix that, you have to test that. Think about a bug that shows up hours into a mission. And you have to replicate it... think about that. And do you want videos to introduce it and tell the story? That's money. It would be nice to do a campaign, and that might expand the market of people who would buy the game, but it's beyond our budget. The other things is that I think that building a replayable meta-game that can take the place of a campaign is a strong, well proven solution – look at the Total War games. As long as you can implement good AI, you can create something that is different every time.
RPS: Can you talk at all about how this galactic war is going to work?
Mavor: I can't really go into a great amount of detail, but there's basically different ways of playing it. There's the single-player approach, where you want to conquer the galaxy by yourself, and then there's the multiplayer online approach, where you're playing with a lot of other people. We'll be making some announcements about that stuff as we go forward. But if you look at the Galactic War stuff from The Boneyards, then you can see the direction we are going in. That was a TA metagame that Cavedog ran back in the day. It was really cool!
RPS: Can you talk a bit about how the inter-planetary structure of the game is going to work? It's quite a complex thing for players to read and swallow, isn't it?
Mavor: Players already have to swallow a lot of complexity to play these games. And I don't ever want to treat players like they are dumb, what I want to do is to give them tools to play interesting games. There are a number of things we are doing with the UI, like the ability to have multiple windows, and auto-labelling of bases – more meta-information at a higher level to give you an idea what is going on. It's going to be more complex than playing on a square map, of course it is, but you can play on one planet if you want to - if you want a 20-minute game play just on a smaller map! Scalability is key to the whole thing. Players are going to be able to control this stuff so that they can determine where those sweet spots are. They are going to determine with a system editor how complex a game is.
RPS: Can you compare this to Supreme Commander? I mean, on the biggest maps with multiple players that's hours and hours of play. Does PA go up that far? Can you make a comparison?
Mavor: I want to have a lot more capability than that. Hopefully you've heard me talk about 40-player games that take place over tens of hours. The client-server technology that we are developing means that it is possible to have a game that is persistently there, and have teams come in and join a game in shifts. We could start up a game with four different armies and twenty players, and you might leave go have dinner, have someone else control your forces, then come back, and the game is ongoing. It could potentially persist. We are trying to push the engine way beyond what you would have seen in the past. This cuts both ways of course, because for a game like that you will need a beefy server. But hey we have this thing called The Cloud now, and there are beefy servers for rent right now. To play on the biggest game your server might have to be in a data-centre with an ultimate machine. Conversely, if you want a small game you can run it on anything. The level of scalability is going to be unparalleled. The number I throw around is a million units in a game. Whether we reach that... I dunno. But that's a goal.
RPS: It sounds like an amazing goal. So I could run my own server like that?
Mavor: I fully expect players to run their own servers, especially for mods, custom setups and stuff like that. I feel confident that we could run enough servers for the community if we had to do that, or wanted to do that, but I don't think we need to do that. People are going to want to run servers. People want LAN games, too! Get yourself a beefy server on a gigabit LAN and see how crazy we can make it.
RPS: LAN is a dying technology as far as games go.
Mavor: Well everyone is locking things down for micro-transactions. We do with Monday Night Combat. It's a free to play game, that's how it has to work. And, well, as soon as people can run their own servers, piracy becomes rampant. We have just accepted that will happen. We just want to incentivise people to log in to our network, so you can track stats and so on. I'm completely against the kind of DRM that makes it more difficult for legitimate players to play the game. The other thing is that we are going to be updating this game for a while, like probably a weekly basis for a significant period of time. If you want those updates you'll need to log in. Look at League Of Legends – they add content on a regular basis. Minecraft does the same, it sold nine million copies and did that not with micro-transactions, but just by making the game better on a regular basis.
RPS: So what has happened since the Kickstarter?
Mavor: A tonne of interaction with the fans – we got the forums going, we got the livestreams and updates going out. So communication is big! But in terms of your real question – what are we actually doing on the game? - the first part of it was the procedural planet generation, which I personally worked on and am still working on. Then there's pathfinding and that sort of stuff, which is being developed by someone I hired who is an expert on that subject. We've also started work on the client-server architecture. So at the point we're at now we basically have a lot of the pieces in separate test labs, and there is a very simple playable version of the game that uses these pieces. We are currently integrating the planetary generator, the orbital simulation, the pathfinding, the unit motion, the client-server stuff. So there's a lot of technical stuff to work on.
RPS: How big a team is that?
Mavor: [Turns around and counts head in his office] One, two.... eight programmers full-time. Add the art guys... we're at about twelve the current time?
RPS: That seems like a great size, as opposed to the huge 100-plus teams.
Mavor: That's actually part of our philosophy at Uber: we keep the teams as small as we can. You pay a cost in overhead when a team gets beyond a certain size. We also tend to hire very experienced people. There's not a lot of new blood, so for the most part it's veterans. I'm effectively the lead designer and lead programmer, for example, and William Howe-Lott is the engineering lead, and he runs the day to day team, and he's been a professional programmer since 1988. Elijah, who is doing the pathfinding, did ten years at Gas Powered. It's that sort of super-experienced team that makes it possible to keep the team so small. Our philosophy is “small team of highly experienced people”.
RPS: It showed in the pitch. The Kickstarter video itself is one of the most impressive pitches I've seen from anyone. How close is that gameplay visualisation to how the game will actually look?
Mavor: I think from a visual style standpoint the actual game is going to look a lot better. And that's because for the video we didn't really have time to go to town on how it looked. From a gameplay perspective it's also fairly accurate, but well... because you can look at the video and attempt to extrapolate gameplay specifics, people have done that, but I have to say “no: it's conceptual”. The specifics will vary! But things like selecting the engines that are on the asteroid and then clicking for where you want it to go, that's straight out of how we intend for the game to work.
But it will look better than the video, I think. We're already starting to push past that. The video will look quaint by the time we're finished. And that's because that is the kind of game we want to make: TA-style gameplay, multiple planets, smash asteroids together, big explosions... fundamentally it's not that complex to describe it, and that's why we could get it across so well in the video. Conceptually, you understand what an RTS game is, and you get the concept of smashing stuff together.
RPS: We do. And it looks like a glorious concept. Thanks for your time.
*A single tiny tear drops from Jim's eye at this point.
Planetary Annihilation can be pre-ordered here.