Why I’m Leaving CIG
Based on the quality bar that has been set for this project, a Star Citizen character takes me anywhere from 3-6 weeks to get in game. I have been working at CIG for 17 months. In that time I have completed exactly 5 characters. That’s 24 weeks at most.
So why is this?
There are three main ingredients that are needed to produce quality characters for a project. These are concept, budget, and time.
To say it simply, a character is usually only as good as its concept, since it is the concept that defines the parameters of the execution of the character. There can certainly be bad models created from good concepts, but rarely do bad concepts produce good models. Since the concept is the biggest influence of the model, then it is of utmost importance that the concept doesn’t change once the model has been started. Rob and Megan are two very talented artists that have produced great concepts for us since I have been here. It’s when either these concepts are changed during model creation, or after model completion, that causes problems. This is when direction wants changes to the design of the character rather than the implementation of the concept.
There is an important distinction to be made about asset feedback. Different stages in the process require different types of feedback. If a concept gets approved, then the only feedback for the completed model should be about how well the concept was realized and the quality of the model (sculpt, textures). When the feedback on the model is about the design (“I don’t like the leg straps”) then the problem is with the concept, not the model. I have been doing this long enough to know that sometimes problems don’t come to the surface until after the model is created, sometimes things just don’t translate to 3D the way you expected. But this should be the exception, not the rule. This was a perpetual problem in my time here, a concept gets approved, and then unapproved after the model is created- concept level feedback on models.
In looking over the models of the FPS Marines and Pirates, it should be clear that some are better than others. The difference in quality? Concept. The Heavy Marine and Heavy Pirate had completed concepts that did not change through the building of the model. They are a balanced and polished piece of art as intended by the concept artist. The others? Their concepts changed repeatedly during the model building. They crawled to the finish line as Frankenstein assets, cobbled together by different artists. The next important factor is budget. Anyone who has worked in game art knows that quality is not measured in a vacuum. Quality only exists relative to the available budget- tris and pixels. You plan and build your character based on the budget you expect to have. When I started here, I was astounded to learn that no one was able to tell me the budget for character assets. People seemed to be operating under the mantra “It’s CryEngine, the budget is irrelevant” This attitude for game art production is suicide in a bottle. This apparently is still a difficult question to answer. In the absence of budget, Roberts judges all game assets against his own imagination or an asset in another game. Time is the last ingredient. When an artist is moved off and on an asset, the asset suffers. Staying in the “zone” on a character keeps the motivation high, the excitement stays constant and keeps the momentum constant for moving through the difficulties encountered. Your energy tends not to stall when you can go at an asset full bore through the life of the process. Very rarely did anything I worked on enjoy unbroken attention from me. I know there are always times when the project demands you to hop on something, put out a fire, but this has been a chronic problem during my time here. The Sataball Suit, the last asset we made, was met with satisfaction and praise. What was different? We started with a clear, approved concept, a clear budget, and were given the time to build the model. The methods for creating this model were completely traditional, no new pipeline. The pipeline was never the problem. The problem was not getting the three required ingredients. When we get what we need, we can shine.
- Completion and Unapproval
The simple feeling of completing a task is something that we all take for granted. Whether it’s graduating with a degree or emptying the sink of dirty dishes, the feeling that you have actually done something (especially if it is difficult) gives your day, your week, or your life a sense of progression. You are moving forward and hopefully bettering your situation. We all thrive on the satisfaction of completing a task.
This is why redoing a task over and over again is so draining to the psyche. Now, to be clear- I expect to have to redo things at times. Sometimes the circumstances change, the asset becomes problematic, or the bar has been greatly raised by adjacent assets within the context of the game.
Redoing something more than once? Repeatedly? Every asset? Repeatedly? It is clearly not about the asset or the artist. Several times since I have been here, I have had an asset approved by CR only to learn weeks or months later that he had decided that it wasn’t good enough.
One production phenomenon that has become familiar to anyone working under Roberts is ‘Unapproval.’ That is, when something that was previously approved becomes unacceptable later on in production for reasons known to Roberts only. It is usually based on whim or a nebulous quality bar that has shifted.
When you get approval only to have it revoked later on, repeatedly, approval becomes meaningless. It is no longer a metric of progression. It does not energize or motivate you. It is met with apathy or cynicism.
Redoing the same asset over and over again kills the spirit, and I suspect this was largely the reason the UK character team collapsed.
It is essential that artists take ownership of their assets. It is what drives an artist, draws forth the best of their abilities, and makes them stick through the difficult points in the process of creation. There is nothing more satisfying for an artist than to look at a completed work and saying “I made that”. Ownership also makes it easy to know who is accountable for problems with the asset and forces artists to work in a clean and efficient manner that minimizes future problems because they know that they alone will be held responsible if their asset breaks something or looks bad. When assets are perpetually passed from artist to artist, especially if they are “ducktaping” work done by other artists, there is zero sense of ownership. You are the clean-up crew instead of the creator and this very quickly saps your motivation to do your best. Why? Well because you can’t really claim ownership. It wasn’t your asset. Authority and Responsibility
When someone is hired for a position there is a direct ratio of the authority that they can expect to have, and the amount of responsibility that they will be expected to have. A junior level artist doesn’t expect to be able to make any real decisions on how things are done, but no one would hold them responsible if the pipeline or character art as a whole is sub-par. Conversely, a Lead should expect to have most of the control on pipeline and asset review, and for that privilege and trust they trade accountability for the pipeline efficiency and asset quality as a whole. Because they make larger scale decisions on the system, they a held accountable for what that system produces.
It became clear within a matter of weeks of working at CIG, that all the decisions for the character pipeline and approach had been made- by Roberts. It became clear that this was a company-wide pattern- CR dictates all. Instead of articulating the standard for approval and allowing the team to develop the best methods to meet this bar, Roberts dictates what the method is, usually with a fraction of the knowledge that the employee has over their particular field. Then, when the plan or method fails to produce the results CR wants, the employee inevitable takes the blame, after all they are responsible for their corner of the game.
When you have someone at the top who wants to make every decision but is accountable for no decisions he makes and is keen on publicly blaming those beneath him for those bad decisions, it creates an environment of people desperate to avoid that blame. Since no one can hold CR accountable, and they certainly don’t want to be made out at fault, they point fingers at anyone else. This breeds distrust and resentment among coworkers. I have been a victim of undue blame at times and am sure that I have thrown others under the bus as well.
- Forrest Stephan (Remark: CG Supervisour at CIG, was before a Lead Technical Artist)
Forrest became involved in the character pipeline when it was decided to redo the FPS characters. I like Forrest on a personal level. I think he is a good guy at heart and is doing the best job he can with the experience and personality he has. His mandate from CR was to improve the look of the characters. So what is the problem? Experience and attitude. Forrest came into the character pipeline full force. He had already decided what was lacking for characters. Namely- the ship art pipeline and techniques. He was not concerned with budget and memory. He was not concerned with time. He wanted what CR wanted- great looking screenshots. He dismissed my concerns about the time it will take to do characters like ships, tri count, and memory. He told me that I didn’t know how to model characters (after eight years of doing this). Forrest is very green, but more importantly it is obvious that he does not know how to deal with conflict or even disagreement. Putting someone with so little experience in games and no experience with characters in charge of the character team was frankly insulting. Billy and I spent a month undoing many of the ship techniques that Forrest had insisted on- mostly multiple materials and how the UVs were laid out. I wouldn’t expect someone at Forrest’s experience level to know what is common sense to anyone who has shipped a title. That is-plan for the game, not for screenshots and know that you will have less memory than you think. Forrest made every rookie mistake in the book in his charge, but what was worse is that he mowed down anyone who challenged his naive assumptions with insults and dismissal. With CR at his back, he stomped around like a child wearing his Father’s boots. Convincing Forrest that he might be wrong about something is a campaign in itself. Forrest might have value from his contributions to other areas of the art, but his involvement with characters was wasteful in time and effort, and absolutely corrosive to moral. He is simply the wrong man for the job and is one of my biggest reasons for leaving.
Visions are cheap. Ideas are cheap. A good leader is not simply someone with a vision or a great idea. A good leader not only has the vision, but they can communicate that visionto the team, and more importantly they inspire and energize the team members with that vision. Chris Roberts might have a vision but he can’t communicate it. And therefore, no one on the team knows what it is. This is known to every team member, certainly of the art team. Roberts is not an artist and it is clear he is not a visual communicator. The basic understanding of macro vs micro, what is essential to the piece and what is not, completely escapes him. Everything is of equal importance- the laces on the boot are just as important as the overall value pallet and silhouette, in many cases more. This is indicative of Robert’s extreme lack of understanding of the most basic of artistic principals. That level of ignorance and lack of visual depth for an artist would be problematic, but for someone at a director level, it is absolutely crippling to a project.
Robert’s deficit wouldn’t be much of a problem if he trusted the vision of the art directors, people who are actually artist and have directed other artists. But he doesn’t, insisting that he is the only one who can direct the artists. I suspect this is an issue of ego, a man intent on appearing like a visionary. But regardless, the results so far have been disastrous, rife with perpetual rework, wasted time, and mass frustration. No one can buy into CR’s artistic vision because no one, including CR, seems to know what it is.
So the one thing that no one discusses is the biggest problem. Roberts is someone who on a company- wide level is always feared, but never respected. His direction is met with nervous compliance to his face, and rolled-eyed resentment behind his back. When his orders are articulated later to the rest of the team, and basic questions of logic and practicality are inevitable asked, they are met not with an explanation of why CR’s idea is a good one, but the importance of his happiness. The explanation is always the same- “I know it makes no sense, but that’s what CR wants”. This team is filled with people who have experience publishing other titles. Lots. We all know how it is “supposed” to be done. But everyone is faced with the same repeated dilemma, a choice- make CR happy or do what works for the game? Short term survival vs long term wins. And unfortunately it’s the survival option that wins out, mainly because turning away from a directive of CR is a recipe for unemployment.
I am only speaking from one corner of this project, but I know that the micro managerial frustration experience is an epidemic at CIG. Everyone seems to be unhappy for the exact same reason. I don’t foresee anything changing at CIG if Roberts doesn’t change himself. And this is a shame because the company has all the ingredients to do something truly great, if only they would be allowed to do it.