10 for the Producers: Episode 02 (2014.12.29)

We're always constantly hiring new people. We're interviewing a lot of fresh talent - we've got a lot of candidates that we're looking at - so we are trying to expand our studios globally. We've just had a new round of employees in our Austin office, in our UK office. We've got a few people who just started on here. So we are actively looking to grow. Travis is also kind of spearheading a lot of the recruiting, so have you any insight on that?

Well yeah, no, it's-... I guess I don't know where the statistic comes from that our team's not growing in size because it is, every day. We are constantly growing - it's a huge game to make. As to the point of what particular positions we're really looking to hire for; I can't speak for every studio. Here in the Santa Monica office we're really looking for gameplay engineers - we need a build engineer, we need somebody else on the Flash programming side to help out Zane, I need ship artists, tech designers - a lot of which we're in the process of hiring right now so...

And producers...

And producers, yeah! Yeah it could be "10 for the 5 Producers" at some point.

On top of that I've only been here what, 3 and a half months now, and even within just the first week I was no longer the rookie person. We've already hired quite a few people, so to kinda echo Travis, yeah our count has gone up, recently quite a bit.

Yeah, when I started here there was three people, so we're at like forty something now - that's a high percentage of growth whatever that is. I am not good at math.

If you would like to join the Star Citizen team, if you go to the Cloud Imperium Games website and check out the job openings we are always looking for talented individuals so if you feel you have the experience and the skill-set to join us, please visit the site and apply!

Yeah, absolutely!

That is totally your question.

Yeah it's every single day we come up with something that might be a better idea. Whether we can do it in time for, say, 1.0, which we're in the middle of finalizing right now, or if we're going to have to put it off 'til a later date, and what that later date is and what the time involved in researching that technology is... but yeah...In the time that-... I took over the ship pipeline in late August, beginning of September, and since then we've added a ship paint which involves a second UV set to all the ships, which was a whole big thing that we had to go back and redo all these damage states. The UK team right now at Foundry 42 is working on adding a new procedural damage system for ships so you basically can eliminate the 25's, maybe 50's, maybe the 75% damage states and handle it all through procedural damage - and that all sounds really promising and cool but that will, again, impact the way that we go about making our ships.So it's a constantly evolving thing, and then on the design side, I mean if you saw the initial Cutlass release there was a lot of-... the Cutlass was done, I think it was put in the hangar... December of 2013 - just in time for Christmas around this time last year. And that ship originally, it's design wasn't fully fleshed out, the variants weren't fully fleshed out. So one of the things we've done is Foundry 42's going through and doing the white box for all these and trying to figure out more of the logistical things, like at the time we didn't have a lot of answers on, "what is a med bay? How does a med bay work? What is the bounty hunting? How does that work? How do you capture people? How do you reclaim them?" - Not reclaim them for resources but reclaim them as, like turn them in. Those things weren't answered, the designs weren't in place, and now that we have answers to those questions, yeah we have to come back to the ships and look at them again with a critical eye and go, "okay, now that we've got what we think is a solid gameplay loop, does the ship actually support that mechanically?"So yeah, it's a constantly evolving thing, but all game development is like that, right? It's a constantly evolving, iterative process, so that's really I think part of the fun of being in game development is it's always changing, it's always new there's always an openness and a quest for new ideas, and better ways to do things... so yeah, it's part of what's great about the job.

Even just as producers our day-to-day stuff is evaluated on a regular basis such as: "How do we handle certain types of emails? Certain email groups? How do we send our weekly newsletter or even weekly production report?" So we try to find ways to streamline every single thing we do, so if we find a process that is inefficient or at least could be iteratively improved, we are going to explore it. That's pretty much what we do.

Yeah, if you don't like change this would not be the gig for you.

So this is actually a pretty good question - "what resource other than time is bottlenecking us" - and I would say it's... to answer it in a certain way... it's the gestation period. So there's always that analogy in production of-... especially on things like engineering tasks, right? Where you have so many people working in the same area it becomes less and less efficient, or they can't work in the same area physically, at the same time, on the same code base. We're seeing this a lot in our large world conversion. There's ways that you can break it up, but there are definitely ways in which one guy has to go through it all the way and it's more efficient to do it that way, so you always ask the question, like, is it one of those things where if you, say, have a woman who's pregnant and she will give birth in nine months, if you add nine more women is she going to have a baby in one month? And the answer is no. So a lot of these things, very similar to normal game development, just take a certain amount of time for somebody to go through it and do.In the nature of how SC is being funded being different, it actually.. it interestingly doesn't impact us that much. I would say the openness and the iterative nature with the community impacts us much more. The funding itself, I mean we approach this like building a game, right? So the community aspect and the community involvement part has a lot bigger effect than how we're funded - even though that is through the community. It is nice though to not have visits from... you'd call them the "corporate suits" or whatever... but the publisher, or notes from the publisher, or especially notes from the licensee who is giving you information on how you should respect their property - cause it's our property! So that makes it a lot more fun.

And it's one of those things that... and you mentioned the pregnancy... where there's a misperception that if we're running out of time, just hire more programmers, hire more artists to make it go faster - to kind of give a real world example on your 9 women pregnant...

We're probably the worst two people to talk about pregnant-... anyway...[laughter]

I'm expecting! [pats tummy] ... But yeah, there are some things where throwing more people or more money at it is just not the answer. And that's something that I think every game studio comes to that crossroad at some point where, "do we need to hire more people, or is it just a matter of making sure people remain focused on the task at hand to just streamline the efficiency a little more?"

Yeah, I mean actually you even see that in the time, if you've been following the project for a long time, with the initial studio in Austin being very small. It's very common in game development that at the early stages, having a bunch of people with a bunch of ideas - you know, the "too many cooks" thing is not going to help. So you kind of start small and then you grow and then once you get, "okay, this is our plan for 'X'"... Now, with our plan for 'X' we can have four parallel teams working on 'X' in these, like, very discreet swim lanes - now we can expand four-fold on 'X'. So it's an organic growth process.

You kind of asked the right two people about that! How it works is, we have our QA testers, whether they're testing in game or we get bugs reported from you in the community, the bugs are then recorded within our own bug tracking software. From there the producers take a look at which developers have the bandwidth and what the nature of the bug is. So if it's going to be, for example a HUD or UI bug then chances are it would go to Zane. So we have producers look at everybody's task loads and assign those bugs out to each individual person, and we try to get rough estimates of how long each one would take to resolve from those developers. And that kind of helps us build our work load for particular week, or even month.

Yeah, no, it's gonna be kind of a bad answer to this question 'cause it's contextual, at least in regards to the part of "how long does it take to fix something?", right? So, a very important part of the bug process is, especially when we're in the finalling process right now, is triaging, right? So we get in issues reported by QA, and maybe something's really important to one guy, but it's not important for the product overall - so he marks it as "critical", we lower it to "moderate" or "trivial" in JIRA. (JIRA is the bug tracking software that we use). And essentially something, depending on the priority and how long it's going to take to fix, really informs the speed at which we're going to fix it. So for example in AC 1.0 we're focusing very heavily on these new ships that we're introducing that you'll see very soon (or have already seen depending on the air date of this episode), and what's cool about is a lot of the kind of more legacy ship issues, or things that we know we're going to go back and rework later, we're not going to end up spending time on them. So it's about prioritization and managing people's work loads against the milestones and releases that they have coming up. So to answer the length question, it might take five minutes to fix a bug bug it might not get fixed for a week, depending on somebody else's priorities, right?

Especially if we have something that's blocking us releasing the build, so you know, if you've got into a ship and the client crashes, even if that was a five minute fix, obviously that's going to be a higher priority than something like a gun's appearing slightly off. So it's just constant managing of priorities - what can we push back to the next build, or next version, or next update, given a limited time frame? Like right now we're so close to the 1.0 release, that we have to look at what bugs are not going to be important enough to push back.

Yeah, and if you want to impress your friends with, like, certain vernacular, that may or may not be commonly known... "WNF" - "will not fix", means if you get a bug and you look at it and you go, "meh, not high enough priority to fix for this release - WNF. Kick it off to the next version." If you get something that's tough to reproduce and you can see somebody spending sixteen hours on it because it's like, "oh, we had eight players, and 20% of the time when I think I was kinda doing this it maybe crashed" ... "CNR, low priority," so "can not reproduce, low priority. Give me better repro steps and we'll look into it, otherwise we're not going to waste the time on it for now."

Well, the simple answer is in the long run that's going to be informed by the design, but in the near term, I think we've talked about on some of the pledge goals, the ships like the MISC Reliant, the Aegis Bulldog - so there are a fair number of individual, or two-seater, or three-seater ships coming along. And once we start filling out the Persistent Universe and filling it out with all the different roles that we want to have, and the ships to service those roles, and the paths of scale in those roles, then yeah, we may find gaps we need to fill, we may find new-... I mean it's kind of similar to what Darian said, I think previously, where once you start playing the game more and more, and you have more and more of the systems fleshed out, you start to see where your weaknesses are and where you need to augment it with more ships, and... you know, the ships are in service of the design of the game. So it's really heavily dependent on the design.

This is actually a good thing you asked this question. There may-... it sounds like from the question there may be a little bit of a misconception about the modular system. If you remember back to say, the Cutlass and the... what did they call it? Did they call it "Cutlass-gate" or something like that? Anyways, the idea behind the modular system is you can already swap out coolers, avionics, motherboards, I mean that's something the system already supports. So the additional functionality of the modularity system is to swap out whole internal blocks, right - so the innards of the ship.Without saying too much about the variants for upcoming ships, what this allows you to do is, say if you have something that's a cargo bay you can... maybe on a variant it's a medical bay, or maybe on another variant... well, use the Cutlass example - maybe on another variant it's for transporting prisoners. With the modular system what it allows you to do is basically, if the ship's designed right and the internal architecture is designed right, you can remove out the cargo bay and swap it for the prisoner transport part. And then if you have larger and larger ships, like say up to an Idris, or a Javelin, or a Hull-C or something like that... or a Carrack for example had it's little modular bits... you can start removing whole rooms and replacing them with other rooms, so you can further customize the role and capability that your ship can perform. So it's not so much about items it's about whole parts of the-... not parts of the ship, but ROOMS of the ship - parts of the ships that can support, you know, multiple items inside them, or gameplay inside them. So that's the idea.

Another facet of that is, going back to the motherboard and CPU of a ship's computer, there's going to be modularity to a certain degree on those, such as a motherboard that can handle greater data throughput, or certain apps that you can get for the computer that can lock on to more targets, or has faster locking speed, or maybe have better infrared spectrum signature detection.

Yeah, and what he's saying is completely true, and one of the things that we actually have set up on all the ships right now are a separate targeting computer, a separate avionics computer, a separate CPU. And the CPU can best be thought of as your power plant for your ship, but for the data pipe. So the three pipes we have on the ship - it's the data pipe, which is powered by the CPU, the power pipe that's powered by the power plant, and the heat pipe which is not really powered by anything but is provided into by the weapons and anything else you have attached to the ship - engines, etc. So, yes, for example in 1.0 you'll see the Willis Ops custom targeting computer for example, allows you to dock more simultaneous targets in your... oh, Zane's going to kill me if I don't get this right... "target overview context pane". So, you know, decreased missile lock times... it's all these different kinds of behaviors that fall out of that system. And that's something that we have not-... we on the item manufacturing and design side, have not done a great job of showing off that system very well. You'll start to see it in 1.0, but as we start adding more and more options, and items and components, it'll become much more a part of everyday gameplay.

So this is where the skills of a producer come in. If we don't know the answer, we go and find it. So we actually sought the answers from a few of our Crytek guys - Sean Tracey, Paul Reindell, and James Wright. I've got that answer for you guys here, so forgive me for reading from it, but I think you guys'll find it pretty informative.So Sean actually wanted to say, he actually thought it was a really good question, and an astute observation that changing many of the base implementations of data types like "Quad T", Matrix 4434, and even double precision integers, make integration significantly harder and time consuming, no doubt. So in some cases it can actually be automated which can help mitigate it, but the toughest times will be when and if Crytek introduces a new feature that specifically uses data types, and we have to update that feature completely. Sean also goes on, and James goes on, to say that our 64-bit change is actually a very targeted change of all world space positions and transforms. It is not a blanket and blatant 64-bit change, so while it does raise the difficulty of taking integration from Crytek, it is not impossible. In fact before we complete our first integration from our large-world branch to our main branch, every integration the large-world team took was like taking an integration from Crytek in many ways. So we have had a lot of practice at this point.So on the Crytek front, it's basically quite far away from the base engine at this point and once the 64-bit switch is done it will be even further. But this is not uncommon with games that start with an engine license as middleware. The point of most of this middleware is that you will end up customizing to your games needs, and let's just say that the technical needs for Star Citizen are beyond those of any game that's been done using this engine.

So that information comes from Paul Reindell, who's our lead engine programmer, Sean Tracey who's our CryEngine tech director, and also from James Wright who is also the guy who's actually heading up the 64-bit conversion for a lot of the world space transforms.

You know, I actually have not heard of this. That's pretty cool. I'll have to go look that up!

Ooh! Ooh-ooh-ooh! [banging fists on knees in an excitable manner]

Yeah? You looked it up?

Yes! Well actually, no-no-no, not that I looked it up. Being an avid, amateur cosmologist..

Oh, different from cosmetologist?

Yes!

Okay. Okay. [both laugh]

Every astral body emits sound. So when we landed the Rosetta on the comet recently, do you remember hearing that-... did you see that thing where they were recording sound from the actual comet itself?

No.

You can actually look up online for astral body sounds - things like Jupiter, Saturn - they all emit certain sound patterns. You can actually listen to that. Some of it, I think, I want to say some of it's sped up, or slowed down, I can't remember...

Is it like-... Is he right? It is like radio frequencies?

Yes. If I'm-... I don't want to misrepresent it, or misstate it, but I want to say yes.

Okay! Ah, that's interesting, yeah...

On top of that... well you have-... That's why radio-based astronomy is so big. Radio telescopes. So we can listen to sound from, like, pulsars. Everything gives off, if I remember correctly, some form of radio.

Like somewhere on a spectrum there's little waves and amplitudes happening? Frequencies and stuff?

Yes.

Ah, that's cool. I did not know that.

It's how we listen to some of the most distant objects in our universe is basically picking up what those sound patterns are, and they're really cool to listen to. So as far as functionality in the game? It would be a cool vanity thing. I don't see...

Yeh. I was picturing a ship with, like a giant radar dish here listening for a jump point, or something.

How about in the shape of a giant ear, just for like-...

Yeah, exactly! I was thinking like an old man with one of those ear cones.

Actually I just thought about something. Did you ever read the book, or see the movie "Contact"?

Yeah!

So, did you read the book?

No.

So, the movie actually had almost nothing to do with the book. Where the movie was about religion versus science, the novel itself (which is why it's my favorite Carl Sagan novel) was about transcendental numbers - numbers like pi. About how there's no end - it's a perpetual number that never has the same pattern repeating once. But in the novel, if you went far down enough into those numbers, they actually started seeing patterns emerge, which is where they got the plans for that giant structure. So how do you have something that's so fundamental to the laws of the universe have a pattern for designs of a space tunnel, and a spaceship? While that was a lot of creative license and there's no actual pattern found in those numbers, it would be interesting, as a hidden Easter egg in the game. Like, if you listen to a certain sound patterns long enough your computer will start processing it and you find sort of hidden messages... like "you left the light on in your oven"

Yeah I can already see like we'd go to Citizen Con though, and we'd have some guy with a big beard and uncut fingernails, and he's, "I figured it out!" [both laugh]

"I've been sitting here for four years!"

"My wife has left me, but... it has finally all made sense!"