If there’s one thing that the gaming industry now has in common with the industry from ten and even twenty years ago, it’s that making a truly path-breaking, innovative and exciting game is still supremely difficult. Chris Roberts – who created perhaps the best space simulation of all time with Wing Commander – knows this all too well. But after nearly a decade out of the industry, the man returned to reveal Star Citizen – an open world, massively multiplayer online space sim that lets players explore the vast reaches of the universe and craft their own destinies. After initial crowd-funding, Roberts and Cloud Imperium Games would eventually surpass $35 million in funding. A figure like that, without the backing of a major publisher, is simply awe-inspiring in its own right.
GamingBolt recently spoke to Roberts about Star Citizen and the team’s modular approach to development. How will the game handle so many players online? What will the ships and physics be like? Roberts reveals all and much more below.
Ravi Sinha: It has been revealed that Citizens will enjoy benefits like reduced tax and have easier access to inter species trade. You stated before that there would be a self-sustained economy. How will this actually work in the game?
Chris Roberts: The model’s really similar to the system we have in today’s world. For instance, there will be different taxes on those coming into a country as compared to those who already live there. So the idea of citizenship isn’t about greatly reducing taxes. You’re a citizen and have rights that are different than people who aren’t citizens. Perhaps you could have different licenses for trading and you could also have a better chance of being protected if you’re a citizen. So it’s sort of taking modern day practices with nations and extrapolating them into the future. Some people may not want to be a citizen. If you want to operate on the other side of the law, you probably don’t want to be a citizen.
Ravi Sinha: With regards to the self-sustained economy, what does it say about micro-transactions? Will they be present in the game or will there be something else?
Chris Roberts: The goal of the game is to have you be able buy everything in the game with in-game credits. There’s no…the design of the game doesn’t have things you can only purchase with real money. It’s also not to make it that you have to have purchased in-game credits with real money because the game is sort of blocking you till you spend a certain amount. So the concept we have for it – and I don’t know whether it will work out but we will see – is that once you bought the game you should be able to play it forever for free if you want.
Of course, if you don’t have the time to play it and want to do something else, we can allow you to buy some credits with real money but we’re going to sort of cap that on a monthly basis. There might be some people that back this game that don’t have 20 hours a week to dedicate like a student would, right? They may have a few hours per weekend and they may have family and jobs but you know, they may want to fly fancy spaceships and not want to spend the time earning the credits to buy it. They’ll buy some credits and fly a spaceship that way. So basically, you have to pay with either time or money but paying with time is not a punishment because that’s actually where the fun is. The whole point of the game is running around and making money.
I think that’s a good enough balance and I hate games which force you to buy stuff and I don’t want to buy. I don’t really want that in this game and between that and bringing new people in the fold, hopefully we’ll generate enough money to maintain servers and add new content to the game.
Ravi Sinha: Star Citizen started on CryEngine 3 and is running on the 4th generation of CryEngine. What kind of benefits has this given to the game’s development?
Chris Roberts: It’s incredibly powerful and visually as good as any engine out there. If you look at the next gen games, in my opinion, the best looking next-gen game is Ryse. Ryse is a fairly simplistic sort of God of War mash-em up but just in terms of visuals and looks, it’s by far the best looking…its better looking than Killzone [Shadow Fall]. I’m a big Killzone fan but just technically if I look at Ryse and I look at Killzone, Ryse is superior.
So I think that gives you a big leg up to design in a way that allows for the content creator like artists and designers to do content fairly quickly. It’s a big, really full featured engine that also comes with everything. It comes with physics, everything you would normally add or imagine, and we just build on top of that.
Ravi Sinha: You have also mentioned that you want Star Citizen to replace Crysis as the next benchmark for new GPUs. There’s a concern that the final PC requirements for the game will be quite high. Given the scope and depth of the game what is the development team doing to ensure that optimization is their top priority?
Chris Roberts: Couple of things we’re doing – one, we’re merging a whole bunch of simpler meshes that use the same material since they’ll render at the same time on GPU level which looks like it will give us a decent frame rate. On top of that, we’re doing things like obviously supporting Mantle which is from AMD and basically that makes the arithmetic go straight to the GPU hardware and bypasses DirectX, which is a big bottleneck.
Direct X was built during a whole different time and they didn’t really conceive of the machines of today with parallel, multi-core CPUs. There’s two of the major areas we’re investing our R&D into. So then of course we’ll be doing a lot of standard stuff like level of detail. One of the nice things about being so strongly funded at the moment is that I can actually take some of the money and invest it in R&D in a way I probably wouldn’t have been able to for various areas to make the game better.
Ravi Sinha: Star Citizen will be utilizing AMD Mantle. Can you give us an example explaining the benefits of Mantle which otherwise would have been made possible?
Chris Roberts: The thing with Mantle is that it lets you go straight to the hardware so you can have actually more draw cores. One of the big bottlenecks in PC game performance is how many draw cores you can actually get through the GPUs. Normally if you profile a modern based PC game, its CPU drops a little bit. Now you can have a whole bunch of Titans in it and they’re not really doing much because they can’t get all the information from the computer from the CPU side down to the GPU because it’s getting bottle-necked in Direct X.
DirectX is really a multi-threaded API so at most I think it deals with two threads, whereas something like Mantle can basically exist on 8 threads…So that’s sort of the best thing about it, something like Mantle, is that it does a little multi-processing and it’s a much better architecture to work with multi-processors and then it allows you to sort of bypass the OS along with the middle bloatware of something like DirectX.
Ravi Sinha: Given that the game has reached almost insane funding numbers, does it create added pressure on you to deliver on the promises? How do you tackle such a situation wherein you are actually ‘over’ funded?
Chris Roberts: To be honest, to be funded at this level is really relieving and reduces pressure for me. When I first started out, this was always going to be an ambitious game so I didn’t actually think I would ever be able to raise the money to build the game I wanted to build. Just from the crowd, it was like…okay, the crowd wanted a space sim. We started to do so well that I thought maybe we could raise the 20 million that I think we needed to build this game, but we did that and more this year. So now it gets to the point where the extra money isn’t like, “okay, great I can buy a new holiday” but about building a bigger and better universe. And then doing things like investing in sort of longer-term graphical performance research and we’re also investing in some procedural stuff. I can afford to have a team that is just doing some procedural work in research.
It may or may never ever pay off in the game but if it does, it would be really great. So it’s allowing me to do that and sort of allowing me to scale resources sooner than I would have done…if I didn’t have as much money I would’ve been a little bit slower about it, because you actually want to be more cautious so you don’t over-spend. But now I can sort of scale up teams to be working on every feature in the game right now. That just means for the game itself, the full-featured game, is going to have a better chance of being delivered when – and one hopes it’s going to get delivered – and it’s certainly going to be in a better state.
Ravi Sinha: You stated in an interview with Massively, that you want the motion capture up to par with James Cameron’s Avatar. How close do you think the development team has reached in achieving the same?
Chris Roberts: I think on the motion capture and facial capture side, we will be pretty close. The difference between, say, what happens with something like Avatar and what will happen with us is that we’ll capture it and put it back in real-time and they’ll capture it and take that capture and also do finesse and animation on it and put more detail in it and then render. We still won’t be as good as sort of an upline rendered thing but I think we’ll be , we’re definitely going to be, better than anything that’s out there currently. I know the tech we’re using. I can see the stuff so I would say a game like Ryse is sort of a high point for the capture and animation stuff and that’s like a standard that we’re going able to have.
Ravi Sinha: Will most of the exploration take place via ships? Or do you have other plans for expanding on first person exploration?
Chris Roberts: Definitely ships exploring space but now that we’ve got greater funding, I will also invest some money into building some sort of on-site exploration stuff. If you follow some of the stuff we had with the character Tonya Oriel but she’s like sort of a historian/archaeologist/space explorer. I kind of want situations like that where you go to a planet and then there’s this deserted alien city and you can go down and exploring or spelunking and find some artefacts. So we’re definitely going to build a few things like that. I think it will be a lot of fun.
Ravi Sinha: Can you talk us through the ships? What kind of customization is the game going to offer the players? Is the customization for each ship going differ as each one of them caters to a particular need?
Chris Roberts: There’s a huge amount… Any one of our ships has somewhere from 20 to 40 different components that you can plug in, switch and mix-and-match. A lot of it has to do with how you want to fly. You can set up your ship to be just a and then you sort of focus on weapons and generating power through those weapons and shields. You can focus on maneuverability which is much more about equipment that increases your ability to go faster and maneuver well and make turns quickly.
And then you can focus on something like cargo holding where you’re focusing on something like space-to-ship cargo, engines to be able to drive that cargo…you can have a ship with stealth stuff where it’s got a very low radar signature. So there’s just a huge amount of ways beyond just the individual ship types to be able to customize your ship beyond that with different roles. What I’m hoping is that when you go up against certain people, certain enemies or whatever, you kind of know how they equip their ships and you sort of put a sort of countering load-out on your ship.
Ravi Sinha: Given that Star Citizen releases in 2015 – with the entire full game out by that time – there are procedurally generated space sims like X: Rebirth and No Man’s Sky doing the rounds (the latter releasing in 2014). What are your thoughts on these and how would they prepare people when Star Citizen actually does come around?
Chris Roberts: I’m quite excited since there was a long time where no one seemed to care about space games and now all of sudden there’s a whole bunch of them so…I don’t know whether I had anything to do with it or just that I sort of had the same feeling that a bunch of other people had. But it’s cool, I like it. I backed Elite Dangerous a while ago. I actually bought X: Rebirth when it came out and No Man’s Sky looks really interesting.
I don’t think that it’s going to be like…there’s always sort of these fan boy arguments like “This is better than that” like the classic PS vs. Xbox argument. For me, it’s more the merrier. I mean I had my biggest successes in the 90’s where Lucas Arts was going head to head with me with Tie Fighter and X Wing and my sales didn’t decrease. I thought it was good because honestly seeing other people do stuff spurs you on and makes you want to make your own game better.
So I think friendly competition is good. I’m pretty interested in trying some of the games which use procedurally generated content like No Man’s Sky. People don’t think of Star Citizen as being a procedural game but we are using that hand-crafted approach and the procedural approach. It’s the only way we can fill the universe. We’ve got R&D going on for procedural stuff and it’s pretty exciting and interesting.
Ravi Sinha: One of the things most difficult about developing space sims, even in the days of Independence War, was accurately modelling space physics. What is it like these days, especially when you have something as big as Star Citizen? Is it more or less the same or is the task more difficult than before?
Chris Roberts: Space physics are actually much simpler than atmospheric physics because all the elements you to figure out in the atmosphere, you don’t have to figure out in space. Physics is not that difficult. Sometimes I think people sort of misunderstand a level of intelligent flight controls that defines space physics, who actually try and fly your ship around and fires all these weapons, and it’s not very easy for any of us to do. Specifically in most games it distracts them a lot.
In Star Citizen, we’re simulating everything correctly. We basically have a flight computer which is sitting between your input on the joystick and control services and to the thrusters and engines of the ship.
Ravi Sinha: I read somewhere that the Galaxy Server is currently supporting up to 45,000 players can be scaled in millions if the need be. This is a number that has been previously unheard of. The question that I want to ask is what are you doing to make sure to avoid outages and other issues? There will be literally millions who will try and log in at the same time.
Chris Roberts: The Galaxy Server isn’t really a Galaxy Sever. It’s a cluster actually and it works in conjunction with what we call instance servers. It’s the instance servers are the ones that are actually talking to the clients of the people playing and taking care of the load in space when people see other or fight with other. You’re not going to be able to have 25,000 or a million in the same place in the same area but the universe cluster tracks everyone and can track millions of people if necessary, and it’s doing some dynamic match-making on the fly.
It’s basically seeing that “this person and this person have entered the same area” so let me spin up an instance server in that instance and hand off these people to that server. The instance server will then take care of the job of communicating between the two people so they can get up to each other’s position and see other. That’s kind of what the universe cluster does.
The universe cluster is really more of a match-making system what player you’ve got, what ships you’ve got, where you are in the galaxy and then it’s constantly match-making people based on where they are, what their allegiances are, what their ping is…all those kind of things and then puts them together with other players which are in the same area. It then hands them off to a local instance server.
Ravi Sinha: The PS4 and Xbox One released this past November. What are your thoughts on the architecture of both and how they differ from the previous generation?
Chris Roberts: I’m much happier with what they’ve done with the PS4 and Xbox One than what they did in the past. Basically it’s a PC and it’s actually going to make it a better situation for PC gamers. Now a game built on next gen consoles is really just a PC game and so, I’m going to get half these ports for powerful games that are some dynamic technology of seven years that still work on PC. But I think it’s good…certainly not as powerful as a really high end gaming rig but for the money, you can’t really beat that value.
Basically Sony and Microsoft are selling at a cost. I sort of like systems as they are focused on gaming and you don’t have a whole lot of other things in the way. The power up is decent and while it’s not a high end game machine, you can do some interesting stuff on it. But I’m much happier with the newer consoles than the old ones. I’d say the Xbox One can be used for sort of kind of TV or non-gamey stuff and the PS4 is the better gaming machine.
Ravi Sinha: When No Man’s Sky was first revealed, both Sony and Microsoft were enthusiastic about bringing the game to their platforms. You stated before that there are no plans to bring Star Citizen to either next gen console or any other platform besides PC. But considering there is some enthusiasm for these kinds of games, and especially for Star Citizen which is going to be a much bigger game than No Man’s Sky, will these plans change somewhere down the line with enough enthusiasm?
Chris Roberts: Oh yeah. I basically said that I’m not against having a wider audience since I’d be really pleased. To me it’s just the same as…having Star Citizen on, say, PS4 isn’t really any different from putting it on Steam Box. So really it’s about the ability to update our community…People playing in the same universe to all the things traditionally that the platform holders in the console space haven’t let people do. If they change that, then we’ll definitely consider being on the next gen consoles as well.
Ravi Sinha: Star Citizen is following a modular approach to game’s development. How has this made the development process easier?
Chris Roberts: I’m not sure it’s made the development process easier. I do think it’s made the development process more focused because instead of having big goals, we’ve got this massive overwhelming game that we’ve been developing for two and a half year and everyone’s been focused on shorter term goals and less amount of functionality and when you add them all together, it’s still the same as Star Citizen. People got to focus on the hangar, we got to focus on the gun fights, focusing on the first person combat, focusing on single-player stuff because it’s a huge exciting story…so I think the modular fashion has allowed us to split it up among teams and distribute it.
We’re essentially the equivalent of a modern day computer which gains efficiency by having multiple processors. Here we’re having multiple teams staying small, nimble and efficient focusing on different aspects of the game and not getting too overwhelmed. There are people who back the game and get to give us feedback and they get to be involved in the process, which I think is really big. You don’t ever get to be involved in the behind-the-scenes working of a game like this. I think it’s kind of fun, and everyone on my side is having a lot of fun. It’s a lot of fun to talk to people who are excited about what we’re doing. Who doesn’t want to talk to people who are excited about what you’re doing?
Ravi Sinha: When would the next module be coming out? [Already announced as pushed back]
Chris Roberts: This upcoming module is being pushed back because I want to have the proper back-end system so originally when I was first speaking about this modular approach. But to do this I knew I’d need to use the existing CryEngine net code because proper back-end is a big engineering job and we’ve been working on it for quite a while. So I thought that would be alright but the further we got, the bigger audience we got, we’ve thought we’ve got 200,000 people now who are eligible for the alpha.
Just supporting that online requires the net code that’s not built for that sort of volume of people would require a lot of work and we’d have to do a lot of maintenance. Of course this work will never be used in the final game, so that sort of deal is like a bad, bad use of resourcing and so I’d rather delay this so I could polish it a little bit and we can get a proper back end with the systems and instance servers and all the rest of the stuff working. Then we’ll be able to test the game in a proper set-up and hopefully that will help reduce the…down the road but also it’s going to allow better activity between the modules.
The way it’s going to work and should work is that in the back end multiplayer, you’re allowed to chat with your friends and the organization and say “hey, let’s go”. To do all that, really requires a proper back end and the systems, so it’s definitely worth doing rather that than putting out something tacky out first or the other option which is to do a single player shoots some ships down but that doesn’t seem very appealing to me. I’m more interested in testing the game with all the pilots flying around.
Ravi Sinha: We know that Star Citizen will also have a single-player component. Could you provide a few details on the same?
Chris Roberts: It’s Squadron 42, the single-player component and it’s essentially Wing Commander, a next-generation Wing Commander set in our universe and it has a really great story which I think people are really going to like. But it’s that style of gameplay, you’re out in a military carrier, running military missions…it’s more linear than the open Star Citizen sandbox with the Universe but it’s got characters and narrative than a sandbox game.
Right now, we don’t want to talk much about the story since that would kind of ruin it. But I think it’s pretty cool and the key to building that is our newest studio in Manchester, UK and it’s my brother and all his team members and leaders that were at Traveller’s Tales, making LEGO games the last seven years and before that they built Star Lancer for me and Digital Anvil and they also built Privateer 2. They’re incredibly knowledgeable and they’re really great and they’re taking the lead on it. So far they’re doing really great work.