We now have a response from someone inside Cloud Imperium addressing the recent departures and major concerns with the game’s development which have been fuelled by the community and Derek Smart’s recent blog post.
Today a lengthy explanation from Cloud Imperium’s Community Manager, Ben Lesnick attempts to answer any concerns point by point. He covers items such as the Star Marine stalling, feature creep, concept ship sales, drop in recent sales numbers, staff leaving and more.
It’s a lengthy update which may or may not appease all members of the community who have become increasingly frustrated by the lack of progress and delays.
Putting on my community hat for a moment, and based on reading many, many damage limitation posts over the years from developers, some of this makes sense while some of it still doesn’t answer delay concerns.
The lack of progress concerns are not something that’s going away overnight, there are members of the community that want to see some real action taken and perhaps even a slight change and scale-back in Star Citizen’s development.
In light of the fact we’ve been covering this issue closely for the past week, you can read the full update from Ben below.
Update: You may also want to watch/listen to PC Invasion Podcast #6 – Star Citizen Special with Derek Smart which was recorded at the same time as the this Cloud Imperium post appeared.
Hey guys! I know you have quite a few concerns this week, and I’d like to take a little time to settle some issues as best I can, informally. This is all in the spirit of improving communication, and I’ll do my best to keep on this thread to answer things. Let’s try and keep the discussion polite… we have enough toxicity already for a group of people who all want the same thing. 🙂 If there’s anything I’ve missed in this long post, let me know and I’ll say what I can. (I’d originally been going to reply to one of the big threads with this, but it got so long that I figured I would just put it out here for everyone to see.)
‘Star Marine is delayed indefinitely.’
First of all: the phrase ‘delayed indefinitely’ being bandied around is incorrect. We do not have a release date to announce yet, which is not the same thing. I’m aware that there is an ongoing sub-debate in which one poster will insist it is delayed indefinitely accompanied by the dictionary definition of indefinitely, and then the counter argument is that the phrase has a different meaning beyond the individual words. This counter argument is correct. ‘Delayed indefinitely’ is a games industry PR term for ‘cancelled.’ Anyone (and apparently this is a great many people) reading clickbait headlines will believe we’ve cancelled Star Marine. This is not the case, to the point that it implies the absolute opposite of what’s actually happening.
What is happening? The weekly updates from the team will give you a better idea, but the short story is that Star Marine was not ready for launch when we had hoped (and planned.) We spent several weeks expecting that resolving a then-current crop of blockers would allow a PTU publish. When this didn’t happen, we conducted a full review of the module lead by our top technical folks from around the company. What they determined was what you read in Chris’ letter two weeks ago: we need to rebuild several ‘boring’ backend pieces and we need to fix serious animation issues before there would be any benefit to a release.
It would not be my place to give you a timetable, but with the number of people I’m seeing who genuinely believe that we somehow now aren’t doing the FPS module I will say that we are talking about a delay of weeks and not months/years/decades. Certainly, I hope not to be pushing folks for weekly updates very soon. 🙂 (And to answer the inevitable question: why don’t we give an internal target date right now? Because the worry is that we might find ourselves in exactly the same position when the current set of issues are resolved. We’re at a point in the process where we believe we know exactly what to do. We’ve already let you know what that is, we’re going to continue sharing the progress… and when that work pays off (or if it doesn’t) you’ll hear about it.)
Ben is correct here. In Chris Roberts’ update he says it has been set back by a couple of months at least. Some publications appeared to have thought it was cancelled.
‘Star Marine is Call of Duty in space!’
First of all, let me speak out against reducing any concept to such broad strokes. Yes, Call of Duty and Star Marine are both first person shooters. That is just about where the resemblance ends. (I guess that’s more of a pet peeve of mine, though. It’s something publishers require for internal pitches… you don’t come up with an original idea, you come up with an idea and explain to your boss how much like a popular game it is. To wit, I once gritted my teeth working with EA on a Privateer reboot that was to be pitched as ‘Battlefield meets Grand Theft Auto.’ Shudder.)
But I digress! The important thing that I want to point out here is that Star Marine isn’t an aside in any way… it’s an essential part of Star Citizen, something the rest of the game must have. We aren’t making a giant first person shooter, but we’re making a game that needs that technology in order to work. Star Marine is the blood and sinew of the game, the connective tissue that plugs planetside into boarding into space combat and so on. One of the least sexy but most important aspects of game development is building the behind-the-screens modules that make up the finished form. For everything you see, there’s dozens of pieces working together: audio systems, streaming managers, graphics renderers, physics layers and so on. Star Marine is that on a macro level… it gets plugged into Star Citizen to build the whole we’ve dreamed of.
(How long have we dreamed about this? Let me tell you: when I was eighteen years old I was lucky enough to visit Digital Anvil, Chris’ previous game studio. The Wingman himself gave me a tour of the office and demo’d the then-upcoming StarLancer in their little theater. StarLancer was great, he explained, but StarLancer 2 was going to be even better: they were already working on a design that would let you get out of your ship to fight boarding actions. I thought it was the coolest idea I’d ever heard. I’m thirty-four now, and I still think it’s up there. 🙂
I will end this one noting that we HAVE built a little in-world fiction to make the module a game-within-a-game… but that’s purely gloss, something we can do without taking developer time in order to make the experience more immersive. It’s intended to show you our dedication to the world of Star Citizen, not our belief that Star Citizen needs to be a CoD-style gunfight.
This is correct in the fact that FPS elements are required for aspects of the space sim. However, creating a full-blown multiplayer FPS experience has pushed the game’s scope further and away from the initial game design notes released to backers.
‘You are spending too much time polishing the game.’
This is absolutely inaccurate. We are grappling with blockers, not a polish (for those unfamiliar, polish is typically what a game does at the very end of the development process… you make the art nice, the particle effects fancy, make sure there are no ‘replace mes,’ etc.) I do not know how folks are remembering Arena Commander as some sort of polished experience. When we shipped it, it had a single ship, physics that didn’t work and multiplayer that didn’t multiplay. The community helped us make all of these things better, and there’s still massive amounts of work being done (and to be done.) No one on the team believes that Arena Commander represents the finished form of Star Citizen, and if anyone outside tells you this is the case then they are just wrong. (Now: Arena Commander looks great because we have amazing artists, but even then none of it was polished. Every single piece of art you saw last year has been revamped since then, every single ship has either been reworked or is scheduled to be. You’re seeing very impressive work, which is what Chris gets out of the team… but you aren’t seeing a polished game.) (And I am aware that the common reply to this is: but Chris used the /word/ polish in his letter! That is not the same thing as doing a polish pass, and I’m hard pressed to believe that that’s genuinely confusing folks.)
I don’t think there is a problem with too much polishing. Arena Commander is broken and there is nothing else of the game to see right now.
I don’t have much to say to this, beyond that it’s not accurate. At this point, we are not adding additional features to the plan, we’re building out the ones we’ve already scheduled. I’ve seen some recent posts about how Chris’ “first person universe” is at odds with the original Kickstarter-era plan… and that’s again not the case. It’s a more recent way of describing what he wants to accomplish, but everything we’re working on is still what was pitched back then: Privateer-style persistent universe, Squadron 42 single player game, first person boarding and so on. (A desire to avoid feature creep is exactly why we stopped doing stretch goals, despite being aware that they drive revenue.)
The removal of stretch goals was a good move., Coming up with new ideas for every million raised must have proved difficult. The FPS part of the game has expanded so there has been elements of feature creep.
‘You’re spending all your time on concept sales!’
We aren’t! Concept sales are something of a slow burn that uses mostly outsource talent who would not otherwise be working on the game. Early in the process, they require a fair amount of design work. Luckily, that’s work we need for the broader game: how will bounty hunters work, how passengers will work, how will repair work and so on. Once that’s done, they’re given to a concept artist (almost always an outside contractor) who works with high level folks on the form and function. When the ship’s design is finished, it gets assigned to a technical designer who figures out how the specific ship will integrate into the game (How big is it, how do the internals lay out, etc.) The fact that we can have regular concept sales is because we have the pipeline working properly – it should be a good sign for outsiders reviewing our production process, not a bad one! (Although the reason it work so smoothly is that it’s infinitely easier to predict a timeline than when you’re taking into account creating new technology and solving game issues. When a producer is trying to do that, he has to base the timeline on something much more vague… whereas you can pretty much know exactly how many hours it’ll take Ryan Church to make a spaceship!)
Concept sales have been an important part of fund raising for the project. With $400+ ships being put up for sale it’s been lucrative for CIG. With stretch goals being removed, there needed to be something else to continue raising funds. Creating these do take time and effort and perhaps the resources would be better spent on releasing finished content.
‘My ship isn’t flyable yet!’
Despite rumors (and jokes, I promise!) to the contrary, we have not forgotten about the Caterpillar, Banu Merchantman, Xi’An Scout, Constellation, Retaliator, Orion, Herald, Vanguard or any of the others. 🙂 Every ship we’ve sold (and quite a few we haven’t) are on the block schedule and every dog will get his day. Why are some ships prioritized before others? There’s a couple reasons for that.
The thing to remember is that we work with limited development resources. That’s not to say we don’t have enough people or we need more money or we need to do so-and-so. I know that everyone imagines Star Citizen as kind of a straight line: you’ll do this ship then this feature and then this ship. But try imagining it like a producer does, as a giant puzzle they need to solve, a huge schedule grid and a list of personnel they can assign. So while it’s tempting to say: let’s get the Vanguard out today and sell it! the actual way it’s done is: let’s make sure the Vanguard is ready by the time we plan to finish the game. Imagine (and these numbers are made up) you have twenty ships and five artists. Even if you kick off five ships at once (Instead of one, or equivalent parts of ten of them which often makes more sense) there’s still going to naturally have to be a priority system to get to the final goal rather than an immediate benchmark. And with that, I can share a little bit of the logic behind the process, how we assign out our resources to make the ships we’ve promised on the schedule:
* Ships that will be used in Squadron 42 are a priority. (I know this is taken many different ways, and I even have a point to address for it below… but for me, it shows our dedication to the game above increasing revenue. The artists who are building UEE battleships and Vanduul dreadnaughts could instead churn out ships and variants that we could sell… and we don’t do that.)
* Ships for Arena Commander 1.0 are the second priority. These are the single seat ships that make sense to include in the current dogfighting alpha. We are very, very close to the end for these, with the Merlin and the Herald actively being worked on now. There are a few more that were added to the schedule later that we’ll see down the line, such as the Archimedes… but they’re coming!
* The bigger ships aren’t going to be flyable until Arena Commander 2.0. If you recall the Arena Commander 1.0 launch way back when, we ended up publishing with just the Hornet and then we methodically finished and introduced all of the other ships. We will do the same thing for the multicrew release: perfect one or two example ships and then use the processes we’ve created to make the others flyable.
* Some ships just wouldn’t make sense to focus on yet. We have a dogfighting arena right now, for building out that very important part of the game… but quite a few ships AREN’T oriented for fighting whatsoever. Things like cargo ships, tugs, science ships, mining ships and the like will be prioritized in line to go with the tech required to make them interesting for you. So the Orion can come when we debut mining, the Hull when we start the cargolympics and so on.
This does go some way to explaining the ship omissions and it makes sense to push for Squadron 42 ships as a priority.
‘They’re only making assets for Squadron 42!’
This is also not the case. Many of our artists are working on Squadron 42 and many are not. Here in Santa Monica, the only artist we have on a Squadron asset is the one assigned to the Herald, which is a ship that will appear in Arena Commander 1.0 well before it’s needed for SQ42. The good news is that as Squadron 42 ships are finished, we free up extremely talented artists who can then focus on the ships in the queue.
This is good news, it at least indicates major progress on the single player Sqaudron 42.
‘Sales are down, we’re doomed!’
We are not concerned, and you shouldn’t be either. We saw the same trend last year, and are keenly aware that interest in Star Citizen is based on our ability to deliver fresh content. Star Marine has delayed that, and the influx of new players suffers as a result. But every single person on the team is confident in what we’re doing and that we’re going to deliver things that deserve real attention.
(I know how frustrating it is when we say that something looks cool but we can’t share it, so please forgive me this one. I shared it with the mod team, and they encouraged me to tell you all. About two weeks back, when Erin was in town, he called everyone together to show some of the first selects from the Squadron 42 motion capture shoot. It was incredibly early temp footage… lips sometimes didn’t move, objects were missing texture, crowd scenes were empty… but I teared up like a baby seeing it. It was the reminder we all needed that incredible things are coming, and it was a personal assurance to me that the project I’d given so much to was going to be a reality… that I’d helped bring back the spirit of Wing Commander. It was a big emotional moment for me.)
Ben is right, sales will fluctuate at different times of the year although it would probably be best they cooled off on the concept ship sales for a while. There should be enough money in the pot to start actually releasing more of a game.
‘Where are the previous web features you’ve talked about?’
This is a good question, and reviewing our past six months of communication I believe it’s probably the most serious concern here. The short answer is, of course, that they’re still coming. Organizations 2.0 and the star map are still in progress, although like every other aspect of the game there are blockers and there are resource allocation requirements. As with Star Marine, it’s the decidedly unsexy back-end stuff that takes the most time. Turbulent has had a lot added to their plate that they didn’t originally intend to handle, from actually working on Arena Commander (leaderboards, matchmaking, etc.) to the massive tax changes earlier this year mandated by the EU.
In the case of the star map, it’s coming along very well. The most significant blocker is getting the data right at this point, which means locking down areas of the PU that we were (until recently) free to change and move around at will. Organizations 2.0 is taking longer, although they’re looking now at how to bring some of the features out earlier. The biggest blocker there is that we’d rather have it tie into the persistent universe than be a sort of fake web-only experience. Expect to hear more about it later this year (that’s all I could get out of ‘em! :)).
‘Chris Roberts is wasting his time directing the performance capture shoot.’
Positively untrue. I am not sure where the belief that Chris fell into a black hole came from (save simply the fact that he can’t do 10ftC while on set) but it is not true. Our production process already works with teams around the world and we’ve done the same here – Chris is still intimately involved with everything from the intricacies of ship design to what kind of boots particular characters will wear. And frankly, he’s the best man for the job. Chris is an accomplished interactive director, arguably the world’s finest for this genre given his experience with Wing Commander III and IV (two projects he both ran and directed.) The plan was to have him direct the shoot from day one, and that wasn’t ever going to change. We made sure our company structure and our internal communications would support it well in advance, and… well, they did.
Performance capture needs to be done. Chris has experience in directing so it’s not unusual to see him spring into action.
‘X employee is leaving, we’re doomed!’
As I said in a recent post, turnover sucks… but it’s a constant in this or pretty much any other industry. The sky is not falling. From the inside, it’s always interesting to see how the world reacts. Because it comes off as so specific – the guys who decide to let you know they’re leaving like Travis or Eric are the heart and the soul of the game… because you know them above anything else. In the past two years, we’ve had some amazing talent that has moved on for plenty of reasons (other opportunities, personal issues, etc.) and it’s always sort of a shame to see they’re never appreciated. I’ll also say that the averages work out pretty well: we are hiring extremely talented people many times faster than we lose them.
I also believe there’s some confusion as to just what a producer does. It’s nowhere near as glamorous as it sounds: they aren’t designing the game or telling anyone how it’s going to work… they’re responsible for very methodical processes. A producer is essentially a scheduler, someone who works with giant spreadsheets and charts to make sure the components needed to realize Chris’ vision can happen and that personnel are tasked with individual processes. And then they’re responsible for holding the whip: you said the Freelancer rework would take seven days and you’re on day eight, what’s going on and how do we fix it? and so on.
To speak personally: I’m very sorry to see Travis go, he was a friend and I will miss his company (we actually first met when we were teenagers, when his dad was a producer for Chris!) I didn’t know Alex very well, but I certainly have a high degree of respect for him. And for my honest money, the biggest loss was Chelsea. She really helped set the kind and dedicated tone for the CS department early on, and we’re going to be extra careful to make sure the care she put into helping backers carries over as we get bigger.
And let me end this one by adding: if there’s anything that genuinely makes me unhappy, it’s the speculation about actual people and their lives. It’s almost always bizarrely wrong and it’s just so stupid. Did anyone see the Reddit thread about how Chelsea’s ‘body language’ during her goodbye on RtV proved there was something sinister going on? I mean… seriously? The drama folks are imagining doesn’t exist. Like most people, we’re more like a family than a TV soap opera; there’s certainly tense moments on occasion while we’re arguing about big things (I will confess being ready to murder Travis once or twice… twice), but at the end of the day we truly all get along. The reaction on RtV was real because we’re all genuinely sorry to lose the friends we’d see every day, nothing more.
While Travis has moved on to pastures new, the bigger concern is the departure of Alex Mayberry which was a big catch for CIG.
‘It’s not open development!’
I think this is an easy one to attack because it’s a vague concept; there’s no dictionary definition for ‘open development’ and if there were we’d still argue about it. We see open development as sharing our progress with you every step of the way, and I believe we do a fairly good job of that. We’ll continue to get better, we’ll continue to work from feedback… but we make the entire team available to interact with you, we tell you what we’re doing on a weekly and monthly basis… we think it’s pretty open.
But make no mistake: open development does not mean you get every single build, or that you get to play with everything we do the moment we start on it. Backers help test the game when it makes sense for them to do so – when we’re at a point where the feedback is valuable instead of obvious. Working it any other way would be a very expensive mess (pushing a patch costs money; patching three times a day for the sake of showing you we’re not lying when we say the game doesn’t work yet would use up that $80 million pretty darn quick!)
(A concern I’ve seen relating to this: there’s so much stuff in the leak that we haven’t released yet. That’s absolutely true, but it is NOT a case of holding material for some fancy reveal. Much of it is Squadron 42 content, a fair amount is part of the upcoming social module… but none of it is Arena Commander content that should have gone out for review yet. I think the telling thing that speaks to our development, though, is that pretty much anything folks have dug out is something you already knew exists and not some crazy surprise. We don’t share everything right away because it’s counter-productive to the development process; we’re proud of how cool Star Citizen looks… but we also know that our job is to make Star Citizen work, not make sure every alpha build is polished and finished at any given moment. That means coming up with a balance that preserves some sense of discovery for things like Squadron 42. It’s not a perfect science, but it’s something we put a lot of thought into.)
What is open development? How much should we know and see? Ben is correct in that there is no real definition of this but the concerns have been over the lack of progress this year.
‘CIG is not communicating with us!’
I firmly believe that this is absolutely incorrect, but I also believe that it’s a criticism that will never, ever go away. I will continue to push my folks to their limits to communicate with you and we will always try to improve… but if you’re someone who honestly believes development is behind an impassable wall, you’re incorrect. Between AtV, RtV, the monthly report, weekly Star Marine reports, Jump Point articles, Meet the Devs, Bug Smashers, 10 for… we’re putting an insane amount of content out there. And we’ll keep doing more, to the best of our abilities!
The ensuing argument for this one is of course that we’re telling but not showing. To that, I have to say… that’s correct. In so much as we’re anything, we are reporters and not coders, designers or marketeers (probably not a word.) The bottom line is that it’s hard enough to take ten minutes from leads to do the weekly Spectrum Dispatch videos where they tell you what they’re doing… but the shortest marketing video to go with such a video is at least four hours taken away from an artist or a technical designer… and often quite a bit more time. Showing today is a battle between taking time away from gamedev to reassure folks that the game exists, and that’s a losing proposition. We’re working to hire an internal marketing artist to do some of this, but it’s never going to be the way you’re likely imagining. Showing you the game is a lot more complicated than running StarCitizen.exe and capturing some footage for ourselves. We’re here to tell you what’s happening, and show you when it doesn’t interrupt development (like with the recent video of Randy working on the Starliner; we want to do a lot more of that! But it’s a matter of being in the right place when the work is happening, which can be hard to do with studios around the world.
CIG have communicated with the community but not always on the questions the community want answered which is why there has been some backlash in recent weeks. This goes back to the previous point of open development and what it actually means.