Star Citizen Live: Mission Features

Well, for me, I think principals are people who specialize in the subject matters that they focus on. And they really know that; they're also very dependable. Hear that, Luke? No, but they can also see complex features through and also guide more junior designers and even senior designers, and show the juniors the ropes and show them the systems and talk them through the whole process of developing Star Citizen and stuff like that. Maybe Elliot has something he'd like to add.

Yeah, we're kind of like an information well. We need to know a lot about all of the systems we own in case anyone needs to know something about, say, "Oh, how do I set up jurisdictions?" And then they know to come to MFT and they can talk to the principals, even though we might not work every day with them. We're like a sponge. We have to remember that information and not let any of it go, so we can go, "Oh, right, okay, this is how that worked." But for a lot of the time, we've been writing a lot of documentation, so we don't have to answer.

Documentation, it's like you have so many designers and they come to you asking questions, and at some point it's like, I can't answer the same question 20 times. I need to start writing this down. And then a lot of it's just pouring all of that information onto pages so that more and more people can just digest it in their own time.

But the worst thing that comes from that is that obviously we're still building the game. It takes you a full day to write this huge system up and you're like, "I've done it." And then the next day, one of the engineers will go, "I changed how this works." And you're like, "Oh."

About three years ago, I remember trying to write a document about how to put multiplayer logic into missions. And I started writing it, and then I was like, this is going to take ages. And I just knew on the horizon it was all going to change. So I just went, "Luke, I'm not going to write that now. I'm just not going to bother. It'll be out of date by next week." But we've written them now, because things have solidified a lot more, but at the time they were not super solidified.

I've seen both sides of the coin as well. When I first joined, I found a couple of little systems where I'd go to Elliot and be like, "Please help me with this, I don't understand it," and then he'd be like, "Yeah, that part is tribal knowledge, and we're going to do better and we're going to write that documentation down." And then Ed and Elliot have gone away and spent, like Elliot said, a day writing this. And it saved us so much time further down the line when me and Lars or whoever's coming back to these things, we can just read it. There's GIFs, there's all sorts to help us out.

So we always try to make the first mission. When it comes to building the feature, the best way to build the feature is to build it in a mission to be able to test it, so we know it works. So when we come up with the time trial system, James will design it, and they'll also build the first mission. Overall, everyone looks at it as the new Kareah update, but internally we go, well, that's the location under attack module, that is essentially for telling if there are bad guys at a location and sending out a mission to retake it. Then there's the timed item dispenser, which is what Jumptown runs on as well, and Kareah runs on. It just spits out boxes for us every X amount of minutes. So we build all those modules and then build something to use them, because then we can go to the content teams: "We've built this. This is how we used it as an example. And here's all the documentation. Knock yourselves out."

Right, and everybody from the EU locations teams to the Montreal teams to anybody else, you've now given them a working example, and once you get around to writing the documentation, they now have the documentation of how it breaks apart and what modules are being used for what, and now they can go and make their own mission content. We've seen that with the Montreal team who are making all those new crash and derelict locations and the shanty towns; they're using the mission features that you guys have put together and assembling all their own missions.

It's good for them because they can focus on content expansion a lot more than bug fixing. I work with, I'm going to say Guillaume, Guillaume Theorette. So what he'll do is he'll set up the eliminate, what we call eliminate all, which is just kill a bunch of people at a location. He'll set it up, and then if there's a bug with it, because we own the base system, there'd be a bug with every single one all over the system, because they're all built modular. So then he can just go, "Right, there's a bug here, off you go, Elliot," and then carry on expanding, and then we can investigate if it's on our side or maybe it's something going wrong with setup or something like that. So it just gives them more time to do content expansion rather than focusing on these little bugs that we can quickly do.

Modules. Verbs and modules. Verbs are the gameplay type, and then the module is what they plug in to achieve that verb. It could be multiple, could be one.

Gotcha, so the idea would then be that with all these features, all the modules and everything, they can now go into Pyro and just start adding missions all over the place.

It's just like Lego, they're just putting the bricks on top of each other and figuring out what they want to build. I have made some hideous Lego, though.

I think one of the things that stands out as well from MFT is the fact that we do make those modules for everyone else to use, but we also think of the developer experience as well: how do they want to use this? How can we speed up the mechanic for them to even just place it in the game and whatnot?

Yeah, and modules aren't done once they're made. Elliot's probably adapted some of those modules, added new features as content teams want to express missions in ways that current modules just don't quite serve. And then he goes back in and adds those inputs and adds that functionality. Even more possibilities are possible.

And back to documentation for all those new additions.

It's documentation. They open documentation. It tells them in which tools they can do what. When we talk about having that developer experience, for anyone else who opens up stuff like UE5, that is a highly polished engine for wide use. For any other games company, it is not wildly polished. Make it work, and then the devs can usually figure it out. And sometimes it is the most horrible or very difficult sort of pipeline. However, we make a big effort to try and go, "Right, well, if we put this here and this here and that here, that's going to make it more obvious." And we'll document it like this, because they need to read the documentation, because it's not always 100% clear. So there are little bits and bobs where we'll change how something works, purely so on the front end of it another developer can go, "Ah, that's what that does, that makes more sense." And we do a lot of heavy lifting on the back end to save them having to do that pain on the front end.

Some plain English inputs and stuff like that. In the background, it's all cables.

Chances are, over the longer course of development, there will eventually be some kind of interface.

Absolutely. It does improve. The tools guys are working all the time on making it better, and we feed that back to them, and they make it better.

Well, as you mentioned, unlocking those areas for those teams was one of the major goals there. And we thought, when it comes to racing, how can we make an implementation that can be used in so many different ways? Because obviously the PU is like a sandbox; you want to be able to just plunk this down wherever you want. You want to make interesting locations. So when it came to racing in general, we looked at what came before. We had scramble racing in the past and we wanted something new. And we looked at what people enjoy doing, and we looked at the community in general as well. There's so many amazing racing communities out there. And what they'd done is they had these leaderboards on a website that people raced around. Now, you had to trust the person had done it in that time, or there was no way of them actually having their times tracked officially. So we ended up coming together, me, Ed, Luke, working on this design together and thinking of the best ways of basically giving them those capabilities of being able to do that in the game. Time trials, akin to rallying, was kind of the first thing that came to my head. I'm a big fan of rallying, like Colin McRae and whatnot. And it also benefited in a way that you didn't need a million different people in that one location to do the race. If no one else came to that location, you could still have a fun time on your own. That led us to the current implementation, really. We do have some future goals, and I think Ed will speak more on that. But it was super fun to make.

The Orison missions. Go to an Orison island and kill people. I saw a couple of people playing it. Citizen Kate did a video on it that was very fun to watch, because there was this odd thing where she was playing it and the server performance was really good, so the AI were acting how they were meant to, and they were snapping, and it got to the point where in the video I think she said, "I need to put a silencer on because I keep just getting annihilated from miles away," which was very nice. It's the first step to branching out some of those missions into new locations.

The prison stuff and the Jumptown changes were both done by our other designer, Max. So we made the prison mission; he's made it so there's AI wandering about, you kill them, you can loot them, you can then sell the loot to the contraband kiosk. We actually recently made a change where we were having a problem where people had to go to this one little area to sell the contraband and sell all the mineables. And then you had some of the bad players we talked about stood there waiting, just hammering people, just killing people. So what we did was, we were like, "Right, let's move the mining kiosk to three different locations to give those people a bit of a chance and spread it out a bit more." So we implemented that. And then the other thing Max did was the Jumptown changes. We added two item dispensers into Jumptown, so there are now more boxes that you can get. And we also snazzed up the location with the help of the art team and the locations team. We added a little bit more so it was less of a one-entry sort of thing, because that's a choke point. We wanted to add a few more ways of getting in and out. And you've seen some amazing videos on that as well. Love watching all of them. Citizen Kate again; Bed Bananas did a good one.

I've been working on XenoThreat, trying to get that up to speed with 3.18. It's a lot in 3.18. Persistence is the big one, the big fundamental change. Making a dynamic event work in those situations, it's a lot to do.

I imagine just little things like, now it tracks mission progress. I've seen people on PTU say, "I did a mission a week ago and I came back and it was still listed there."

Well, mission persistence isn't something that we have properly in 3.18. It's more that so much has changed since XenoThreat last ran. It touches every system. Every system has persistence in it, or not in it, or whatever amounts of it.

Wasn't there a bug previously because of persistence, where if the Javelin was destroyed its records stayed there and then the other Javelin was trying to come in? We had that with the AI reinforcements that land: if you blow that ship up, the other one's just like, "Well, I'm still going to try to land there," even though you still exist. It's a fun thing to battle.

We will be seeing the changes mandated by persistence for the next two years. Everything up to this point was made for the game in a certain state. And now with persistence, there's changes to the AI, like, "Hey, if there's already a crashed ship there, maybe we don't land there."

It's also quite challenging to work on a game that has persistence turned on by default. Because what we do is we run servers locally and test our stuff. But if you shut that server down, it persists. So when you run your server again, it's still there, and all the stuff is exactly as you left it. Sometimes that's annoying because you're looking for a fresh state, but it isn't. That's persistence. And so you're like, this is like making games again. I've never made a game where I have to worry about persisting things from my previous run. You really do have to think in a new way to work in that situation. When you're fixing bugs, you have to be really mindful of that, because you're thinking, okay, is this bug still here because it's persisted? Because I had an issue with that. I was setting up some elevators, and the bug was there. I fixed it, and then I ran the server again, and the entity had changed. I fixed it, except it persisted in the broken state. So I was like, okay, fine, this is only a bug in this moment. So if I purge the lot and go again, it really is fixed. But it's challenging, but it's kind of fun.

It also messes with our debug tools. These are all good things, by the way; this is just the war stories. When we're playing, say I'm testing a UGF mission, or I'm waiting for the shuttle because I'm testing something, and then Liza's like, "Hey, you got a second?" or Jane's like, "You got a second?" and I go over to them. And then I come back and I haven't turned god mode on, which will stop me from dying. And all of a sudden, my food and my hydration levels have gone to zero. So my guy is having the worst time of his life. He can't see anything, I can't hear anything. And then I'm like, ah, I put on god mode and then I do my bug. But then every time afterwards when I log in, my guy is still dying of hydration permanently until I actually purge the server and restart.

I think the funniest thing that happened to me was during time trials prototyping. Because persistence was coming in and I didn't realize what it was going to do, I'd logged out mid-race, went away for my lunch, and came back to record a video for an internal review. I thought I was doing really well in this race; I was on time for the Platinum. And then all of a sudden, my wreckage was just at the end of the race, and I just smashed right into myself, which ended my race prematurely.

On Kareah, we were actually debugging it and all of a sudden we fly in and there's three of the chief of security. They're super bosses, they're super tanky. So I come in there with a pistol just to test something, and there's three of them, like, "There he is." And I just got lit up instantly. That's a problem.

Salvage missions at the moment. We wanted a way to show the new salvage mechanic. So I'm currently working on some different legalities and versions. There's a legal version where we have a new company that's being made, I'm not sure if I can call it that, it's called Adagio, and they're literally just salvage contracts. They have a bunch of salvage and they sell you the rights to a piece of salvage. You pay X amount of credits or UEC, and then, "Here's your lot number." You can go get it there, and then you fly over and it's yours to keep. And it's kind of in the safe area at this point, like a Lagrange point where they kind of towed it there. Then there's the illegal version, where basically another illegal corporation, a new one called the Tar Pits, hear about salvage and they're like, "Well, you can go there, but we know that security forces are going there to investigate what happened. You have an hour, good luck." And then you basically have to race there. And once you're there, you have an hour; you have a timer on screen. And if you're staying there too long, well, then all of a sudden fighters come in, or other investigative forces. If you're in a Vulture, for example, well, that is not really good at combat. And we're actually trying to get a mission giver, which I won't spoil who it is, so have fun trying to discover that. But a new mission giver, we're going to give him or her some love. And they're going to actually do a kind of mini rush where it's like, "Well, there's been a massive battle somewhere in the middle of space. There's a lot of money to be made, but it's an open contract. So here you go, but you might find people." And then we're really trying to have emergent gameplay, community focus. So it could be three Vultures all happily salvaging away, and then an org is like, "Yeah, this is our salvage now," and they come in with a Reclaimer and a whole combat suite. It's like, "Okay, kids, move aside. This is ours now."

Oh yeah. It's not really missions, it's system. It's probably what everyone's been clamoring for. It's not a mission, it's the right to defend your ship, which is probably what everyone really wants, right? They want to be able to stand their ground in their ship and not get a crime stamp.

It's not a mission; it goes into us building other stuff. We don't just make missions. Me, Ed, and Luke are also looking at a system and trying to improve the law aspect, where someone might not have the right to defend their ship from an intruder or someone who's on board. It's like, "Well, I can't legally kill him because he's not got a crime stamp." But it's like, you've not given permission to be there. So it's systems like that, and we work on a lot of those as well.

We're always reading Spectrum. I'm on it every night, which is terrible, really bad. But I read all the threads on there, and Reddit and the other ones. I'm always zeroing in on the things we own, hostility and law, and then we're always just picking them up and sending them in, and the next day our embedded QA are just going through everything we found, and we're always trying to improve those systems and make them better.

Well, I started off working with Chris. We were just finishing school and just playing around with graphics on the BBC Micro. And he was doing his games, Stryker's Run and Match Day, and kind of said to Phil and I, because we all liked doing art, we all did art at school, "Come draw some pixels, let's do some stuff on here." So we did that, and all of a sudden he's driving around in a really flashy car, and Phil and I are looking at each other going, "What's happened here?" And he's like, "Sold this game," and he's minted. So then we cottoned on to this, and Chris hung around for a bit, and then he went to Origin, and obviously the rest is history.

So you've been with Chris for a long time then.

Yeah, well, I went into training. I was on the tech side and I was training companies. I was one of the only authorized Autodesk trainers of 3D Studio. So I was going around to games companies in the area, training their staff how to use 3D Studio. One of them was a company called Rowan Software, which did a lot of flight simulator games, so I was teaching them how to build planes. And I kind of got, "Well, I can do this. I can do this for a living." So Erin came back to the UK, set up a studio for EA and pulled a load of us in, and that was me in the games industry from there on.

We went to primary school together. It goes back a long way. But yeah, I'm glad you gave me a heads up on that, because being put on the spot with "Tell me a story about Chris" I think could have ended with career suicide for me.

Tell me a story about Chris.

One that hopefully won't get me fired, and I think anyone who's seen Chris will probably attest it's fairly accurate. Go back to when we were like 15 or 16 and we just played D&D. Chris kind of got bored with the wooden delivery from the DMs. Somebody would be droning on, and he's like, "This needs more emotion, needs spicing up." So he got this idea to bring his friends in to do voices for the creatures, writing a script, getting sounds like rain sounds or wind sounds in the background. And he's having us all doing these crazy orc voices or dragon voices and things like that to play on a tape recorder while he's DMing. Probably about 40 years back, nearly.

What kind of DM [was he]?

It was fantastic. But the thing is, I think it only went for about three sessions, because it was so work intensive. Like I said before, we were all quite good artists at school, so we were even drawing concept art for environments, just to have the maximum amount of immersion for this game. But I think we got about three sessions in and everyone was like, "Jesus, this is too much work. It's like a job. We're only 15." But yeah, it was an experience. I think everyone who played those campaigns, even for three sessions, would remember it. Certainly Erin and my younger brother Simon, who's the financial director.