Star Citizen: Reverse the Verse LIVE - Chris Roberts on Development

Well, obviously mining, that's a huge one. And I think that sort of creates a gameplay loop that is what people... I mean, there's a certain amount of, you can run missions, you can take cargo and sell it and trade it, but we don't have all the other stuff that we want to do. Mining is another pretty big chunk of that. So I've seen that people are having fun with the first, what we call tier zero implementation of mining. There's still a lot more stuff to go and you'll be able to mine out in asteroid fields and all the rest of stuff. But yeah, so it's really nice to see that. Rolling these out is very important to the overall sort of PU plan, because essentially we want you to be able to... it's a big sandbox and it also should be up to you to do what you want to do in it, and we don't want you to have one way to play it. If you want to be sort of a miner or you want to be a trader or you want to be a pirate, you want to be a mercenary, you want to be a bounty hunter, all those, or if you want to be a farmer or something, all those things are possible. So you can be someone that's more focused on non-combat or someone that's focused on combat. Basically, it should be like the real world. You should have the options to do all the kind of things you can do in the real world and it's up to you to decide what you consider fun and what you consider success.

Now, you said, of course, mining is the big tentpole feature. And it's not the only feature. There's quite a lot of cool stuff.

Well, the grouping stuff is important. That's, again, tier zero. So there's still a level of implementation that will be continuing on in the subsequent patches. But it's very important. I think everyone voted on it when we had the survey. And it was actually one of the things that when we got 3.1 out there, we were like, okay, looking at quality of life, we said there's some things we'd like to improve right now because we want people to be playing right now and experience... because in some ways we look at the game as a live product that's happening right now. And yes, it's still got bugs and crashes and it's still missing a lot of content, obviously, and missing features, but it should be a whole game cycle and everything. But all of those aspects, it's important for us to make the experience of getting in, getting in with your friends. And in some ways, some of the funner content that I see is sort of all player driven, right? So like there's that Daymar rally that they were recently doing. And so having tools for people to be able to get their friends in easily, communicate with them and do it all is really important. And also some of it's foundational for the VOIP slash FOIP stuff that's going to come. So that's another big win. And of course, there's the new ships and there's some new weapons and items you can have. But those are good. And then we've been working on things like, I'm not sure if we've cracked it properly yet, but we've been trying to make the turrets better because they need a lot of help. We were fixing some issues on the IFCS that were just broken from before. I think it's not all done yet. And just some other sort of quality of life stuff and laying some of the groundwork for 3.3, because 3.3 is, of what we're doing this year, the big one, because that's the one that's bringing a whole bunch of critical tech that will allow us to sort of go beyond where we were.

Well, yes, for what we have the time to do. So for instance, there's quite a bit more stuff that we want to do, have more variety, some more nuance, and obviously allow you to mine not just having the Prospector. But for what we've done, yeah, I think we've done a nice combination of... visually quite cool. And there is, I don't know whether you want to call it a mini game, but there is a certain amount of management you have to do as you're mining that makes it more than just holding a space bar for a certain period of time and uptick the resources. And I think as we sort of fill out, like we've got plans for refineries and stuff in 3.3 and beyond. So as we fill out the sort of whole supply chain that will happen with that and have more roles, I think there'll be some really cool stuff where people can be focusing on mining and their friends can be taking what they've mined off to the refinery to get it refined into the various materials that then could be sold on somewhere else. So I think it's a good foundation, but yeah, there's a lot more that we want to do, because we sort of want to build out pretty much all these professions we've talked about to a level of sophistication that perhaps you don't normally see.

For me, when I was jumping in, one of my favorite things to see, and maybe I'm just a weirdo, is flying over the surface and seeing the surface scattered with Prospectors all doing their thing. It's very, very frontier-like.

Well, I think there's some UI. The overall UI needs a better unification. We've just had all these different, like here's the mining UI, here's the scanning UI, here's the HUD stuff. So that's a longer term thing that we'll be fixing up and moving more to sort of 3D UI, which I kind of prefer. But our biggest problem is just pulling all these things together, even with the amount of people we have. It always seems to take longer and it's more difficult than anyone ever conceived or thought. I think the Agile method that we've been using is definitely helping out, where people are sort of having groups that feel like their own little team that can achieve stuff. We're still getting used to that. You can definitely see the benefit, like going from 3.0 to 3.1. And having a cadence is also good. I think 3.2, when we first went to PTU, it was in a lot better shape than, for instance, when 3.0 first went to PTU. And that was partly because we're not taking as long between releases. And so it has to be kept in a better state. And people are, like I said, getting used to the cadence. So I think as we go forward, you'll continue to sort of see benefits from that.

It's hard. Every single feature that we... basically we have a hundred things to do all the time and the time and the resources to do 10 of them. And so you just have to make calls on which one at that particular time you felt was more important. And so the sort of grouping and friends was less important earlier. I think we've increased the player count. We went to 50 with 3.0, and hopefully when we get some of the new stuff that will be coming in, the object container streaming and the bind culling, we'll be able to go beyond 50. We also have some physics optimizations that we're working on longer term that should bring up the ability to also have more players on one server and decent frame rate. So yeah, when we did it, it kind of felt like we'd gotten to the point where there was enough people playing or stability and we should sort of focus on that and also sort of focus on the experience now. So I sort of feel like until we got to that 3.0 level of stuff, there wasn't enough of a game or world to go and explore and do with your friends. Now it sort of feels like there is, and you're seeing it with the mining, and with 3.3 we'll open up a lot more stuff. And so then it really sort of feels like, yeah, okay, let's get a device or a mechanism, a system in place for people to get together to talk, have their various chat in the game. And you'll just see that getting improved as we go forward. Like I said, for 3.3, we'll bring in the VOIP and the FOIP, but also you have more of the Spectrum side of it. So you can also be doing the chatting in Spectrum and you'll be seeing it in the game and vice versa and stuff like that. So it felt like the right time. With all this stuff, there's a whole list of things. I see it on the forums all the time where people go, why don't they do this? Oh my God, they're crazy or they're idiots or they don't know what they're doing. And it's like, we're aware of pretty much all of it. And honestly, it's a matter of priority. And it's always a hard decision to make. So for instance, there are quality of life improvements, like some of the things I've just mentioned, like making the UI or the HUD better or whatever. That takes time to do. But then it also takes time to do a new feature like mining. And then it takes time to do the debugging, making sure the game's running. Like when we're going to PTU, there's a whole bunch of bugs that come up that people have to fix. And so it's really just a matter of allocating the resources and time to that. And sometimes you have to make difficult or hard choices. I love to get new content. I love to get new features in. But if you're always doing that, then the other stuff sort of drops by the wayside. And so that was a conscious decision we made. That's why we did this for 3.2. Okay, going forward... and partly the quarterly release stuff forces us to do that. But we said going forward, we need to spend time improving the quality of life, improving the stability, improving the performance, so the experience playing now is fun and can be something that people are interested in, versus always thinking about what's going to be tomorrow, always about, oh, well, when it's finished, it'll be all this great stuff, but right now it's all a bit of a mess because we haven't finished everything. So we made a conscious choice to that. We're still doing that. It doesn't mean that there aren't some hard choices. We have our roadmap, which we have this whole thing that you have to agree to read with the caveats about how it moves and all the rest of stuff, specifically because we said we're going to be more date-driven than feature-driven or content-driven. And then the moment we move something from it, there's a whole bunch of people going, oh, I can't believe it. They're idiots. They don't know what they're doing. They're incompetent managers or whatever. And that's just the choice you have to make, because you'll come to these things and someone will say, well, do you want to get rid of these bugs? Or do you want to get the performance better? Or do you want to have the experience playing it now better? Or do you want this feature? And sometimes there will also be features that will come along where you go, okay, well, we really need this other feature to make this feature work properly. That one's not going to be ready for this release, so we're going to have to move both of them out. And we said that in our caveats. So the time we move it, it still does. But that's just the nature of the beast. I would guess that the majority of our community sort of get it. I kind of view the modern internet as the sort of Tale of Two Cities, the best of worlds and the worst of worlds at the same time. So the best of worlds allowed us to make this game and have all these people support it and push us forward and do something at a level no one can do. The worst part of it is that, of course, everyone can say whatever they want, and the filter of civility isn't the same as it would be in person. And you tend to focus on the loud people, and the loud people aren't always everybody. And it's just the nature of it. So you have to know how to filter all those. But ultimately on our side, we say, okay, this is what we want to do. Long term is where we're going. And we'll make some of these choices, some of these decisions I know some people are going to get upset about. But it's all in service of trying to make this game as good and fun and meet the vision of myself and everyone working on the game and also the community itself. But it's a journey. It's not something that like tomorrow is magically going to be brilliant. And so, yeah, you just got to learn to put a heat shield on sometimes when you go to the forums.

Well, I think what's really useful, the stuff I appreciate when I read, because I look through the stuff, is when people are constructive. They're like, okay, I have the problem here. Here's what my solutions are. And in a lot of ways, I will respond to that much better than I'll respond to someone shouting over everyone else, telling them, no, this is what everyone thinks. You're wrong. The majority thinks this sucks. It needs to be this way. We're an online game. We know our metrics, right? So we take a look at it and we know very well when someone's saying, oh, yeah, everyone's like this. And we're like, well, that's not actually what the numbers show us. But it doesn't help you, right? Much better if you're saying, look... And I've seen quite a few examples, and we've changed some stuff where people have put forward good arguments. And actually sometimes there'll be good arguments out there, and sometimes people may think, oh, well, they're not listening to me. That's not true. It means that that's actually affected maybe some of the stuff that we're thinking about, but we can't change it right away because, of course, everything's planned and all the issues I just talked about. With the resources and what they're working on, because the worst thing that can happen in development is when you keep switching context. Do this, no, do this, do this, do this. And that drives developers crazy because they're like, I just want to finish one thing. So there's a lot of times where people will be making suggestions about stuff where we'll go, okay, yeah, that's actually something we want to do and we will get to. It just may not happen for six months or a year. But that is the nature. But we do value the feedback. And if it's nice and civil and constructive, it's even better.

I saw a good conversation in Spectrum a couple days ago about the virtue of one very well-constructed and thought-out post versus a thread, a petition with 'we just want this' and a bunch of people voting. It's the power of one well-thought-out and constructed post. If you can make your point, any one backer can affect...

Oh yeah, look, everyone that's on the team, they're all human beings, so to reach someone, to change someone, or to give them an idea to do something, it's much nicer if, instead of calling them an idiot or incompetent, you make suggestions and go from there, because they're more receptive to it. It's just sort of person management 101, and the same applies in forums and the internet in general. I know that there's a certain amount you just can never get away from. I kind of liken it a little to sports teams, right? I grew up in England. I'm a big football fan. And hopefully England will... I don't know, we'll see what happens in the World Cup. But it doesn't matter, after a match there'll always be: the tactics are wrong, that player was an idiot, he didn't know how to do his... That's just the nature of the beast. People are kind of venting because they have passion, and that's okay. But I am saying, if you want to affect design changes, it works much better when there's: here's the issues, here's what I think the solutions are, put forward. And even if it doesn't necessarily end up changing stuff, it does a lot of times make people think about it and consider it. And I know all our designers and everyone else that's involved in design and creative decisions, they do read our forums and other various feedback venues, and it does affect their thinking. So it's good.

I think the quarterly releases are really good. They're great for discipline and cadence on the production side. Sometimes it's tough because you have to move things out and you know that people are going to be upset. Every time things move, the pitchforks are out: I'm not playing. Well, screw this. I'm not playing till 3.3 or 3.4 or wherever they think the next feature is going to come in. But no, I think it's good. And it is actually something consciously that we started last year after sort of taking so long with 3.0, and we just said we don't want to do this again. So we said, this is not a short distance race, it's a long distance race, and we've got a lot of stuff still to do, and we're building a universe at a level of detail I never thought that we would ever build in my wildest dreams when we first started this project. Things have happened to enable us, including the support we've had from everyone and some of the technological breakthroughs that we've had and just the general sort of power of PCs that are allowing us to do that. And so I think it's really important to focus on giving gameplay and content regularly to the community to play, and we sort of view Star Citizen that you play right now as a live product. Yes, it's missing loads of stuff and it's going to get all that stuff, but we need that experience to be good, and there's a discipline that comes with that that I think ultimately will make a much better final product. I think it's really good. It's good for the teams. Again, it's a cadence to get used to. The date doesn't move, so they have to make it, and it causes people to be realistic about what they can and can't do, and then we have to make tough decisions if it looks like the team's trajectory isn't going to deliver the feature that we said, and then we'll move it. I think it works really well for Star Citizen. And also, even if we said, okay, this is what we consider the full final game, it's the kind of cadence we would want afterwards anyway. I think you don't want to go a year or two years between updates like you can see on something like World of Warcraft. I think you want to have these quarterly content and feature updates where the universe, the game is evolving all the time.

It is a must have. So we will not be able to deliver Hurston and the content without object container streaming, because unless everybody's got 32 gigs on their machines, we don't have enough physical memory on the smaller memory machines to have all that data there. So it is a must have. It's something that we've made a massive focus, which means that some other things may move out of 3.3 so we can deliver object container streaming. But for us, we won't be able to deliver, because 3.3 is going to have Hurston and Lorville and some truck stops, space stations, there's a whole bunch of extra stuff that goes into it. So it's a big content... and four moons. So it's a big content expansion for us and it needs the object container streaming. So we're making a big focus. It's the biggest priority on the tech side. It has our top people on it, so to speak. So it is our most senior of the engine folks. This is what we're focusing on. So there's quite a lot of stuff that's still out there to be done, but we're making very good progress. But the nice thing about that is that once it's all working, hopefully that kind of stall when new ships load in won't be there, because it'll all be asynchronous, happening in the background. And that's actually the biggest issue with the streaming: we have to make all the code thread safe, because we have to asynchronously load everything in the background and then just turn it on when it's there. And so that means that there's years and years and years of code that wasn't written to be on other threads that need to be converted across. So it's as big of an endeavor, which we've been working on for quite a while since last year, as moving to 64 bit. We have to change vast amounts of game systems to work in the new sort of component thread safe manner that the old CryEngine stuff didn't used to.

I'd have to go and... we discussed the... basically it's like a scratch disk. So as you load things in, if you have more memory, it may not discard them as quickly. And then, of course, you just reactivate it. So that would be usually the benefit. I don't know if that's going to be in the tier zero object container implementation, but that is what we've talked about on a longer term. So if you have more memory, it basically keeps things around longer. Because what happens with object container streaming, it can only keep so much in a container. Think of it as a box, and you put this stuff in and then, oh, it's too full, I have to get this other stuff out. Well, I'm not close to this other stuff, so I'll get rid of it. So basically, if you add more memory, you just have a bigger sort of memory allocation for that.

It's a must have. Everyone is aware of it. We could deliver some of the 3.3 stuff, but we won't be able to deliver the expanded content because there just won't be enough memory. That's one of the reasons why we don't already have some more moons or Hurston and stuff, because we're already tight on the memory, so we can't just preload everything and just have it sit there.

And not to bury the lede, but it's not just Hurston, as you mentioned, it's four additional moons with Hurston.

And a whole bunch of truck stops. Some refineries. It's a lot. It's massive. It's a much bigger content drop than 3.0 was. That does mean that some of it may not all make it in there because it's so big, but it's a massive step. It is our biggest drop for the year.

Because 3.3 is so big, when we talk about priorities, when we talk about decisions that have to be made and some of the toughest decisions, I think there's probably a couple of those in the future.

Yeah, definitely. Well, because basically if we're ensuring the object container streaming, that may mean some engineering resources that would have been doing something else may not have time to do that. That's an example of... you always have to make hard resource decisions and you have to say, okay, what's my priority. So that sometimes will mean, okay, sorry, we won't be able to deliver this when we thought we would, because we need to make sure this happens.

Well, capital ships in the PU is possible. Object container streaming makes it easier. The Idris still has a few things to finish off before it goes in the PU. I think we're missing the missile room, and it sort of needs to be made non-Squadron-focused. So much of the work on it is Squadron-focused, right? So that is on our roadmap, but we're not planning on doing that for 3.3 or 3.4. So, sorry guys. But I do want to get the bigger ships in there. We have some other tech things that we're working on that will enable... because there's some other optimizations we need to do besides object container streaming in terms of the bigger ships, making sure that they're LODing correctly and we're not updating too many of the attachments if we don't need to. So some part of that will be in 3.3 and some part of that will be in 3.4.

So yes, that's part of the plan. One of the big pushes we have, because obviously, if you look how long it's taking us to do Stanton and you figure out how many star systems we have... I want to be alive by the time we get them done. And I think most of our other backers do too. So no, it's kind of a tool issue and it's a bit of a content issue. So the more sets that we build, and we're fleshing some of them out in Stanton, but we've still got more to do, and the better the tools get... And that's one of our issues, is that we just don't have enough tools people. We don't have enough engineers. And even though I told you we have 500 people, everyone wants more people and we don't have enough to get all this stuff done. But we're making a sort of long term focus and investment, not just on the tech that enables this, but also then the creation and the tools. Because we need to be in a position where we can turn around a moon in a week, not in two months or three months. And the same on the planet. So longer term, we are looking at tools that would be much more like kind of city building or RTS building kind of stuff, where you can sort of paint and do stuff very quickly and build prototypes. And then once we have a whole set of biomes... so, for instance, the four moons that we're doing around Hurston utilize some of the biomes that we've done for our current three moons, right? And we've created some new stuff as well. But the more of that you have, the bigger palette you have, the quicker the turnaround. So we actually on 3.3 have said ourselves... I'd have to check with Ian, but I think he was trying to get all the four moons done in like a month with his team. So that would be way quicker than we did the previous stuff. And so we're working towards it. So we're very aware that the trajectory needs to be much faster on this. But it'll take a while. So I think that sort of stuff, you'll start to see the impact of it next year, not so much this year.

I think we've demonstrated a facet that exists in all game development, and that's exponential growth. If you produce this much in the first year, that allows you to produce this much in the second year, that allows you to produce this much in the third year, and so forth. And you see that borne out in ATV. Remember what ATV used to be in the early days, and then as we hit content phase.

Well, when it was Wingman's Hangar, we basically showed nothing in the game, and it was just Eric goofing around. So it's very different. I totally agree. And also we've grown as an organization and we've got more of the processes now. So yeah, we have a plan to get there. I'm not going to lie to you and say, oh, it's going to be done in a year's time or two years' time. But the first thing is you've got to get the tech, and then you start working on the tools and you start working on the kind of pipeline and workflow, and then you can kind of get it down to something that works quite efficiently. So our ship production pipeline's gotten pretty efficient compared to what it used to be in the early days, especially on the size and the scale of all the stuff they're doing. And longer term, we need to get the planets and the moons and the environments all there. And also don't forget, it's all changed, because it used to be, oh, well, you would just ask for landing and then we'd have a little cinematic and then you land down in a very sort of constrained area where you could go and talk to a bartender or buy something. But now it's a whole planet, and it's a pretty big planet or a pretty big moon. And so now it's like, okay, what are we going to do to make that? So it's not just, well, there's going to be one landing area. It's like, oh, well, how are we going to get little towns around here? Or how are we going to get fauna and an ecosystem and resources you can do? So we're starting to lay all that stuff in. That's all in our minds. So yeah, look, as a player, you should be able to just hang out on one of the worlds and do that forever if you wanted to. So we are going for that level of detail that we never were going for before. And obviously that requires a lot of tech, a lot of tools, and obviously more content.

Well, we're making a big push to get it into 3.3. We were looking at it in 3.2. I just personally felt like we weren't there. The AI side of things in general is also an issue on Squadron; it's gone slower than we'd want it to. It's partly because we've sort of re-engineered this big system, because we need something that's dynamic on a level that you wouldn't normally... because normally on a lot of games, your AI is pretty focused. It can be fairly scripted around sort of the trajectory you're taking through that story or those levels. But for us, we need it much more systemic. And we've got this dual kind of master: we've got the PU, which really needs to be systemic and has all these different things. And then we've got Squadron, where we want an incredible level of detail in what they're doing, both combat and the social passive side, so they really feel alive, and you're with this crew really feeling them as real characters and people. And so there's a complexity that comes with that, but getting it all trickling down to the final functioning, with fluid transitions and animations and all that sort of stuff, has taken longer. And also, even though we have a lot of people, it doesn't mean that we have 100 AI programmers. We have three AI engineers. We'd like to have more, so if there's any great AI engineers out there... We're actually looking at taking our gameplay guys and saying, you know what, you're working on AI. And there's a certain amount of, you've got to get the system and the infrastructure in place to get it done. So that's just sort of been moving a little slower than we wanted, both for the ship stuff... I think we may have fixed some of the stuff, there was some broken recently in the PTU, but I know I've seen it on an internal build that we've fixed them, to be more challenging anyway, but that's a longer term thing as well. And so, yeah, it just felt like the experience... we didn't want to put in an experience that was a bit too rough around the edges. So we want to spend a bit more time to make it a bit more fluid and a bit more polished. Ultimately, like everything on Star Citizen or Squadron, I want that to be as good as any... you look at The Last of Us Part II and you go, wow, they did an awesome job on their animation and the fluidity of it. And that's the kind of stuff we look at and say, yes, we need to be doing that in our game. And we will. But yeah, it's just taken a bit longer than we thought. We've put together a pretty focused strike team to make sure that we get it for 3.3, because it's also quite critical for the stuff that we need in Squadron. Optimistic about having it be in 3.3, but...

That's super important to us. It's fundamentally part of what Squadron has, because it's just you and everyone's AI, and their crew members. So that's dependent on us getting the Squadron stuff to the point that we're happy, and then we sort of bring it across to the PU, which is kind of the same thing that happens to FPS, because it's another level, right? You have the multiplayer aspect where you have to make sure it's synced over all the clients and the server, which you don't need to worry about on the single player. But yeah, we have design, we're starting to build a structure for that. I couldn't tell you where you're going to get your first AI companion on the roadmap. I would suspect next year you're going to see that. Won't be this year, but that is definitely important, because we need to, just on the PU, fill out that world. I talked about this 90-10 sort of ratio, where you want to have about 90% of the world be NPCs, AI, and about 10% be players. And that sort of allows you as a player, if you want to play single, solo, even with your groups, but you want to be PvE or whatever, you don't have to... It's not all about fighting another group of players to fight against or do things with. You can interact with the world at large, whether it's NPCs hanging around Lorville, or flying on a ship, and you're a pirate and you decide to liberate some goods from them, or some AI pirates trying to do the same to you. Because ultimately, as a game, there is a certain amount of wish fulfillment that comes in the game, where no one wants to be the shittest pilot in the world or the shittest dogfighter. And so, therefore... when it's player versus player, there will always be one player that's a loser and one player that's a winner. When it's player versus AI, then there's a lot more chance for the player always to be a winner. And so one thing that AI does is sort of provide... yeah, I get my ass kicked by other players, but I can go whip up some AI pirates myself and feel kind of good about it. And so that is quite important to fleshing out the world, and having also crew that would flesh out some of the bigger ships for you, especially if you don't have a whole bunch of friends to play with or they are not available at that particular time. So it is important, but it's sort of dependent on some of the Squadron stuff we're doing, which I would say roadmap-wise won't be this year. So it'll be next year at the earliest.

Okay, so, well, we're working on it. Production at the moment does not feel comfortable sharing what we have. I'm not going to surprise anyone and say Squadron 42 isn't this year. But we still have some aspects to work on. The biggest issue we have with the Squadron roadmap, besides the fact that we've got to figure out some ways not to sort of give things away... but that's solvable. The biggest issue is, when we have the quarterly roadmap on Star Citizen, from the beginning we're like, okay, well, if this feature doesn't make this, then it'll go to the next quarter. And it's sort of this revolving thing. So it's not quite the same as, bam, here's the final schedule and roadmap for it, with the release at the end. And so, yeah, on the production side, without feeling like every single thing's buttoned down... trust me, I'm on their case too. They're very reluctant to release something that may change because they feel like they're going to get totally pilloried for it. And I know I've been criticized for promising stuff. So essentially it is something that's a work in progress. We've also had to adapt to the new Agile process, which is very different than the old kind of waterfall scheduling process. And we're using a new thing called Portfolio, which is part of the JIRA thing, which is sort of a long-term planning and scheduling thing. We're trying to roll that out for Squadron and the live stuff, which helps give you sort of the longer term visibility, because the Agile development process is quite good for shorter term iterative stuff like we do with these quarterly releases. It works very well for that. But if you're doing a longer term thing that has a whole bunch more interdependencies and features, and there's a vast amount... Squadron as a game is insanely big and ambitious and has a level of fidelity and content that probably, if we started a while ago, I would have said maybe we shouldn't have done this much, but this is what we've kind of committed to. And so, yeah, that's really the reason. It's not necessarily that we're like, oh, we've got it, but we're not going to share it with you. It's more just that, from a confidence standpoint, no one wants to put something out there that they feel like will change again and they'll get criticized, because all our production people get very sensitive when they read all the things about how we can't manage our way out of a wet paper bag and all this stuff. Even though they've worked at, whatever, Blizzard or Rockstar or Activision or EA... whatever big project you've played, we've got people here that have worked on all those at pretty senior levels. But it's just the nature of the beast. I thought it was quite interesting at E3, Cyberpunk 2077. They sort of came out of hiding after all these years and everyone was really excited, and the journalists got to see it behind the scenes and were excited by what they saw, but that thing has been around for as long or longer than us. And it sort of went dark, and the head of CD Projekt Red was talking about how they stopped here, they refactored this, they redid this, they iterated, they did that. That is game development. It's also game development when you're reaching really high and you want a very high quality bar and you're not sort of locked into, oh, we've got to release for the Christmas quarter because that's what our shareholders need. And it is part of the process to make something really good. And you can see it in Red Dead Redemption 2, which was going to come out, and now it's going to be later because they need a bit more time. Rockstar does this a lot. Blizzard does it a lot. We want to make a great game. So it takes time, and that is kind of one of the reasons why we haven't shared the roadmap yet, because some aspects are very confident and some aspects are still up in the air; they need a little better breakdown and planning on some of the longer term stuff. Because again, like I said, the Agile process is very good for the short term planning. But when you're asking someone to sort of plan something that's six months out, that becomes more difficult. And then people tend to be like, well, I don't know what my tasks are. And then if you plan it properly, you're literally taking your whole team and they're doing a proper breakdown, and they're doing that for days, versus creating things or fixing bugs. So then you have this other issue: do you take some of your team away from what they're delivering right now to do the planning, to get you a proper thing for later on, or do they fix that bug so we can get 3.2 out there? So it never happens as quickly as you want. So it is in process. When there is something that production feels comfortable with, Erin feels comfortable with, then we will share it. I'm over being blamed for being optimistic and over-promising, and people feeling like I lied to them, because that's totally not my intention. I am an optimist by nature. You can't be a game designer trying to do things at this level without being totally optimistic, because otherwise you wouldn't start the journey. And I would say that most creators tend to be optimists. You generally won't find a pessimist that's a creative type, because you have to be kind of optimistic to do that. So I'm not the best person estimating my time or other people's time, because I always feel like I could do it quicker than it ultimately would be. So there you go. So I have solved myself with that, and let the other people who are maybe a bit more realistic or pessimistic do that.

Okay, so we're not quite ready to announce it, so you sort of preempted it, but I think we've been sort of looking at our format and feeling like we want to shake things up. First of all, I think we have probably three types of people that we need to have content for. So one type is super diehard, completely knows everything about Star Citizen, obsesses about it, is on Spectrum, on Reddit, watches everything. Then we have the sort of more casual, check in occasionally, what's happening, but it's not a 24-7 obsession. And then we have sort of, hey, what's this Star Citizen thing? It's kind of cool. Tell me a little bit about it. I'm interested. And I think some of the ATV problems that we've had is that we just mash too much stuff together, because in today's world of YouTube, I know myself when I'm watching, it's a different format. It's not the TV format where it's a half hour show or an hour show. It's like, oh, I've got a few minutes, yeah, I'll watch this video. And the moment that video becomes 10 minutes or 20 minutes or 30 minutes, then it becomes a bit of a viewing commitment and it hurts, and so you don't get as many people. So part of what I feel, and I think it's not just me, it's everyone that's part of the team, is that we would be better served to sort of break our content up more appropriately. Because it's really frustrating that you work on a 30 or 40 minute show, everyone jams to get it out there, and then there's people going, I just want to see the dopamine cut. Show me the pretty pictures for five minutes. I don't want anything more. What's the summary of the show? I don't have time to watch it. And it's like, okay. So I think what we're thinking about doing is taking the featurette stuff and making those bigger, more in depth, better, higher production quality, but not try to force one every week. Like, let's do a featurette, let's do a deep dive. We finished mining, let's go into mining, let's go into how we did it, what all the challenges were. Like a sort of really slick behind the scenes, like a little mini doc or something. And that's for your subscribers, your diehards, and the people that really want to know everything about it. But that's a different thing than your sort of weekly update, which is like, okay, here's a weekly update, here's kind of a quick snapshot of what's been happening in Star Citizen, and here's a studio update in a sort of brief format where you see some really cool stuff, but it's not really going into the details, but you get a sense of what's happening, what the progression is, and it's digestible in a nice, easy, non-massive amount of time required way, and very visual and fun. And then we'll also want to be doing some content that's more applicable to people that are like, hey, I want to know about Star Citizen. And so we're sort of thinking about, instead of trying to jam a whole bunch of content into one show, why don't we, not do less content, but break it up. Break it up and be more targeted, and spend a bit more time to increase the production quality levels. We're actually actively sort of hiring up on the video side, what we call GVP, general video production. So we just brought in a new senior editor. We're looking for a showrunner. And so we're trying to up... And we'll probably set it up in such a way that we'll have the capacity to do things and do them to a little higher quality. Because the other big problem that we have with the content generation is getting the developers, getting them to interview, getting the B-roll from them; it can be a heart attack every single week. It's sometimes why things were late in the past, just because they're busy, they're trying to ship something, or they've got one of their sprints that's due and they've got to get it done. And so with a bit more time, a bit more planning, a bit more staff and focus on the video side, and not being beholden to, oh, we need one featurette or two featurettes a week and we need the studio update and all this sort of stuff all crammed in, I think we'll be able to deliver it more consistently in quality and do a better job with the development side, so they feel like they're not just getting harassed on the Tuesday or Wednesday for some B-roll for the show on Thursday. It's an ongoing challenge, but like in everything we try to do, we want to improve every aspect, whether it's the game, how we interact with the community, how we do our video. So we're always looking to make it better. And so this is kind of what we think would be better. So like I said, it won't be less content. It will just be more targeted and different sort of kind of content for the appropriate people. And then if you want to see it all, of course you can see it all. Because otherwise, what's the point? Everyone slaves over their ATV and now there's someone who made a five minute cut of it because they don't want to watch. It's like, well, we should be doing that five minute cut. The community doesn't need to do that for us. We should do it. And so people like that could have that, and the people that really want the in-depth piece, that really want to get behind the scenes and see what's happening there, could have that. And I think that would be better, because sometimes we've had features that we haven't finished the feature on, right? And so everyone's sort of talking about what they're going to do. And I find that kind of frustrating, because we don't have the final imagery. We don't have the final graphics. People are still talking about what they're going to implement. So for me, I'd much rather we do a feature like, okay, we've done this thing and now we can do a nice making-of and stuff. Not dissimilar to how we shifted around on the commercial side. With the quarterly cycle, we've staffed up on the video production. Instead of having to grab cinematic artists or other people to try to capture gameplay or do some of this stuff, we hired some, I think some from the fan community.

Yeah, we brought in some community video makers.

Bring them in to help us make some of this stuff, so we can create some of the Warlock commercial that we just had, or the Eclipse one. But anyway, we're trying to improve on all these. Like everything, we just may be a bit slower than you guys want us to be about it.

Now, we're not announcing any specific plans right now. We're going to take the next couple of weeks, the next couple of months, to experiment with a couple of things. But the idea here is to basically make our content smaller, easier to digest, easier to share with your friends, less of a continuous bombardment. You remember the days we used to do a YouTube video every day of the week, Monday through Friday, and it starts to become kind of... we never want Star Citizen to be a chore.

Well, yeah, and you don't want to have filler, right? Because I know everyone always has rose-colored glasses about the past. Last year it was quite nice. We cycled between the studios and then we had the featurettes and stuff. And it was a bit easier because we hadn't had any releases, so there was a bunch of stuff everyone was dying to know about. But even then it would be like, this Austin one, they don't have anything, screw that, that's boring. Most boring episode ever. And so I think it's not so much about quantity, it's about making the quality higher and then making it sort of targeted appropriately. So that's what we've been thinking. We've been working towards that. We've still got some infrastructure to put in place, because we're trying to solve the issue of how to keep everyone totally up to date on what we're developing so you feel like there's progress, not show everything, because otherwise when you go to CitizenCon there's nothing to know, or you won't play Squadron, you'll be like, well, I already know what's going on, and also not drive our developers crazy because they've got multiple deadlines on them, because that's not fun for them either. So we're trying to find a good cadence for that. But ultimately, I think our goal long term is to give you higher quality content, more targeted.

So just to assuage any concerns right now, we're still doing ATV, we're still doing Calling All Devs. Ship Shape's not going anywhere. We're still going to do Ship Shapes, although it might break out into its own thing, so it's easier to just have it and share.

Well, again, that's what I think. I think at that point you should make the Ship Shape... why would it be with ATV? It should be about, okay, we're ready to talk about a ship, and that should be its show. It should be the focus, because otherwise you've got people that skip... we can see it on the YouTube metrics. Not everybody watches the whole show. There's a much smaller percentage of the people that watch the whole show; a lot of them watch just a portion of it, usually the studio update or whatever, or maybe if there's a ship update they'll go and skip to that one. So I think it's better to be: this is about the Hurricane, this is about mining, this is what the studios are working on this week. I think that would work better.