2

Understanding Preproduction

The typical preproduction period for developing a game is very similar to that of the film industry. It means defining the product! For most studios, going into preproduction means that a great, new game concept is on the table and now it must be pitched and sold to a game publisher. This usually means creating a formidable game design document and either a prototype of the game or a vertical slice of what the game will be like—this is usually a single level of game play that will later be used within the finished title.

Key decisions that determine the quality and cost of the game must be made in preproduction—these choices include the number of personnel that will be hired, the length of the production schedule, the features of the game, and honing the final versions of the game’s technical and creative design documentation.

The first step is determining a crystal-clear concept for the game. Although different studios/publishers have different formats for this (see the Sample Game Concept in Appendix A: Extras), there are a few required elements: genre, platform(s), basic elements, basic story, and main characters. Usually, a general blurb or mission statement is also included to sum up the overall intention of the game, a risk/competitive analysis is constructed, and a prototype is made.

Basically, the concept includes all the information you need when pitching a game to a publisher. Sometimes, this is best approached by thinking of it as a presentation. Think high-concept and creatively, roll it up into a great pitch, and you have a great representation of what kind of game you are proposing. Once you have finished the basic concept of the game, you can then get into the details of the actual game, or the requirements.

The concept phase of development is usually marked by numerous meetings to discuss the various features that will be included within the game. These include the artistic elements, the levels and designs that will be used, and the particular game props (weapons, maps, and so on) that will influence the programming and engineering of the game.

A typical way to approach these meetings is to form a list of priorities: high-priority items are the things that set the game apart from others of the genre and that definitely need to be included, and low-priority items are things that basically just gild the lily—in other words, they are items that would be nice in the game, but are not necessarily needed. No matter what priority an item is given, though, all ideas presented should be noted and considered.

It is also during this period that a rough schedule is outlined based on the needs of the game. Because most project management systems rely on time and labor constraints, the team will plan the amount of development that will be involved with implementing each of the items listed in the requirements. Before moving on to honing the final game plan, it must be determined whether the title can be brought in on schedule based on implementing the listed requirements. If it is determined that it can, the team can now focus on developing the final game plan.

The game plan represents the final overall map for developing the title. In addition to the elements already included in the concept and feature list, a game plan also includes a specific budget, schedule, and staffing plan (examples are included in Appendix A: Extras). This is the bible by which your game will be created! Once you have pitched your project to a publisher and they have accepted your game plan, it will become abundantly clear how important staying on schedule and budget will be. Great skills at overall project management always make the difference between successful studios and unreliable studios.

In the film world, there are several elements of preproduction that are similar to those of the game industry. The hard lessons learned by a hundred years of filmmaking (such as working on script development prior to production) can be applied to not only streamline the preproduction process (especially when forming the concept), but to help create a more cinematic product. These will be discussed in Part 2.

2.1  Script Development

There is currently no recognized general format for creating a script in the game industry. Though there are many writers in the field of game writing, there are dozens of different methods by which these writers create a script for a game. Some resemble basic narrative writing (such as a short story), some approach the script much like a graphic novel (this usually means describing short scenes or panels, then listing the dialogue that will occur there), and some have even gone to great lengths to design their own methods for writing the script that can involve detailed hyperlinks and bookmarks that outline the various paths the gamer can take.

In many cases, game developers have relied upon cut-scenes to relay any kind of story to the gamer. These are usually short animated clips that convey a necessary piece of information or relationship within the game to the viewer. Though in some cases these are quite necessary (especially if the story is rather complex), it is the issue of taking players, even momentarily, out of the game play that has caused the game industry to take a second look at cut-scenes. Another factor that must be considered when producing in-game cut-scenes is the need for specific producers and personnel who are skilled in this area. This can contribute to inflating the budget—especially if you are using celebrity talent or directors!

At any rate, whether you use cut-scenes or not, some type of script will be developed to let the production team know exactly what kind of story they are creating. This is important when defining the locations, characters, and style that will be represented within the game. Typically, once a first draft has been developed and read through by the development team, revisions will be asked for. The notes that you take during the initial read-through will be valuable in helping you to craft revisions according to the wants/needs of the team producing the game. The process of writing the script, reviewing the script, and crafting rewrites may repeat itself many times before the final script is approved.

Do not confuse a game writer, though, with the person who creates the game instruction manual! This should be a completely different person (usually a technical writer), and he/she will not become involved with the project until close to postproduction.

2.2  Hiring the Crew

Depending on the location of the studio, hiring the personnel needed to complete production of a game can either be the easiest of steps or the most frustrating. If you are located in one of the game development meccas (such as Austin, Los Angeles, Vancouver, New York, or San Francisco), crewing up can be as simple as posting the job openings on your Web site and letting the resumes roll in. If you are not in one of the major cities for game development, getting interested (and competent) parties may be a bit more involved.

Typically, a human resources (HR) department takes care of all the hiring for a studio, but if your studio is small (or new), you may not have an HR department in place. That said, there are several different strategies for finding the skilled talent you need: getting a recruiter, posting jobs on popular game job sites, going to the major conferences, and contacting schools.

Hiring a recruiter may be the easiest solution, as they basically do the other three things listed for finding talent. A good recruiter does other things as well, such as screening the applicants, fielding the first round of interviews, and organizing the lists of potential candidates for each position to be reviewed by the production team. Once the recruiter has gathered a number of portfolios, resumes, and work samples, the top candidates can then be scheduled for a more formalized interview process.

If you decide to do without a recruiter, you can post the positions on some of the major game development Web sites and begin contacting schools. Most schools that offer degrees in the various fields of game development have a counselor or administrator in charge of helping find positions for students. Keeping in mind that most people you hire from a school will be completely inexperienced, you can definitely get some great people from the bigger schools for a reasonable price. As payroll can be one of the biggest factors when determining the budget, you can maximize the most by utilizing as many entry-level personnel as possible. Also keep in mind, if you are functioning as your own HR department, you need to set up some kind of formalized review and interview process. The sooner you do this, the faster you will get crewed up.

As far as attracting experienced talent goes, setting up a booth at one of the major game developer conferences is a great strategy (Game Developer’s Conference, E3, Austin Game Developer’s Conference, to list a few). Keep in mind, that the more experience and talent a person possesses, the bigger the salary he or she will want. These are the people, though, that you will hire to be department leads, and they will be the ones that take you to a successful final product.

Either way, once a good-sized stack of perspective applicants has been decided upon (usually by screening out people using phone interviews), it’s time to set up some formal interviews. Some of the important topics to cover in the interview are: gaps in employment, multiple jobs in a short period of time, references, responsibilities at the last position, and future goals of the applicant. Also, it’s a good idea to include as many members of the department that the person is interviewing for in the interview. Once the interview is over, the department can then discuss how they feel the person would fit in with them (sometimes called the “culture” of the workplace) and whether the producer should extend a job offer to the applicant.

The hiring process can be a long one—especially if you need a lot of specialized and experienced personnel. When you are all crewed up, the last thing you need is fast turnover (people quitting/getting fired). This problem can be prevented by implementing some great programs at the studio designed to retain personnel. In addition to the obvious perks (benefits and great environment), some other things you can institute around the studio include the following: on-site gym, kitchen with coffee bar/snacks, game room (useful for competitive research and enjoying downtime), and cross-training (engineers always enjoy learning about the art side of the house and vice versa). These things help you keep a solid and happy studio functioning.

Once the task of hiring the development team is finished and leads for each department have been identified, it’s time for a chain of communication to be established.

2.3  Learning to Scrum

“Scrum” is the basic terminology used for Agile Software Development practices. There are tons of books about the topic— as well as some great Web sites—but the one I recommend is Agile Software Development with SCRUM by Ken Schwaber and Mike Beedle (Prentice Hall, 2001). Basically, the fundamental practice of a scrum is to develop a game bit by bit, reviewing the current component in a meeting environment, then moving on to the next step in development through team consensus. It is the regular meetings of directors, producers, and leads that constitute scrumming (“scrum” is a rugby term for when all the players huddle up on the field to continue play).

One of the most valuable things to learn when beginning the practice of Agile development is to not go into the realm of meeting overload! Regular production meetings are extremely important (especially in the beginning of development) and should be of the highest priority on the schedule, but many times a producer can schedule too many meetings and leads find themselves constantly off the floor (where the younger, inexperienced developers need them) and in meetings. A great way to avoid this is to assign senior team members who are not the leads to attend the scrums and provide input.

The basic scrum session is made up of a scrum master and representatives of each department engaging in planning and brainstorming. This team typically works on small sets of tasks designed to help the development process for everyone in achievable, but small periods of time called “sprints”. Each scrum/sprint builds upon the one completed before it. Soon, these small, manageable tasks get the team on track to turn in another deliverable or reach a scheduled milestone.

The use of Agile development and scrums can create a more focused development cycle, build a team with a higher morale and focus, and implement a simple way for tracking progress. Usually, the scrum is used in conjunction with a typical project management system (such as Microsoft Project) and is one of the fundamental practices of the gaming industry. There are aspects of Agile development that can be especially helpful when initially planning the game, too (in the concept/game plan phase), such as discussing the methodologies by which the development team will work together and support each other. This method is preferable to the old model of “code-like-hell, fix-like-hell”.

Other viable approaches to development (there are many) include the practices known as the Iterative (or Incremental) method and the Waterfall (or Cascade) method.

2.4  Iterative and Waterfall Development Models

Sometimes called the “ iterate-until-you-drop” method, the Iterative development model is based around the idea of developing a game in small increments, then taking advantage of everything learned/gained when developing subsequent steps. The approach revolves around an initialization step that creates a base version of what is being developed, then an iteration step that involves experimenting with and creating a product based upon a control list of features and tasks that need to be performed. This approach usually involves cross-discipline work that focuses on functionality. The flexibility of this approach makes it quite appealing to development teams, and the use of analysis and measurement help ensure a quality product.

By contrast, the Waterfall development method is very step-driven and methodical. It focuses on the idea of working sequentially through the various stages of development (analysis, design, implementation, testing, integration, and maintenance), then delivering the final product. There are various spin-offs of this model—and most are criticized for the belief that a particular development step can be made perfect before continuing to the next step. The Waterfall method does not allow for implementing later changes to earlier ideas—or at least not doing so with any ease. For these reasons, the Waterfall/Cascade method of development is usually avoided.

Whichever method you determine to suit your production needs will include a great deal of project management.

2.5  Project Management

Most producers in the game industry have a background in some sort of project management. This can be as formal as having obtained an MBA or certification through the Project Management Institute (or through Microsoft) or as informal as possessing experience from working on past projects (in or out of the game industry). Either way, chances are good that they will be working with Microsoft Project.

Without getting into the nitty-gritty of Project, the simplest way to understand how it works is to envision the development of your game in what is called a “critical path”. This concept is a sort of step-by-step way of looking at creating your game. Because certain steps have to be completed before other steps (called “dependencies”), a chain of tasks must be accomplished and finished before moving on to the next step.

Each step in the critical path can be plotted in the program to incorporate a schedule, keep you on budget, and to allocate resources for each task that is being assigned. In a nutshell, it’s an all-in-one program for keeping the project on track—exactly what the program was designed for! Usually, the producers track the progress of development using a project management program such as Project as their primary tool for determining how and when a project is getting off track.

Alternatively, more and more producers are starting to use Microsoft Excel for tracking. Though it does not have the ability to create reports about schedule, resources, and so on, Excel does allow you to keep detailed flowcharts annotated with milestone schedules and dates that deliverables should be finished. Excel is also easier to read for those not acquainted with Project. This ease helps immensely when making reports regarding daily progress (sometimes called “daily delta reports”) or discussing the status of production in a studio meeting.

Another way that progress is tracked throughout the development process is the use of online documentation. Most productions use an internal Web site, maintained by the individual departments, that lists the status of current tasks and the progress that is being made (sometimes this is a wiki page, sometimes it is a blog template). This site becomes an important tool by allowing the producer to see what is being accomplished on a daily basis throughout the team. The information extracted from the site usually affects what is going into the daily reporting to the studio and publisher.

The primary goal and concern for every project manager—a producer or director—is to keep the production of the game on schedule and on budget.

2.6  Budgeting

When creating the budget for a project, one primary concern must be kept in mind: the game needs to make a profit! The lower the bottom line, the more likely a publisher is to pick up the title. You probably need to have a schedule in place before finalizing the budget, but if you have the rough schedule, you can usually begin forming the specifics of the budget.

The budget must include every aspect of creating the game, including labor (including outside contract work and any benefits involved with employment), the overhead for maintaining the studio during production, any equipment that will be needed by the development team (computers, other hardware, specialized software, and so on), all fees for the title (licensing), and morale. More often than not, the expense involved with keeping a team in high spirits is one of the areas that bloom out of control—along with feature creep and overly ambitious schedules.

Any time the production team gets into a crunch period (usually right before a milestone is reached and the end of production), it is necessary to foot the bill for many meals, morale trips, and so on to keep the team motivated. As personnel will be working ungodly hours (far more than usual, at any rate), they will be eating a majority of their meals on the job—and on your tab! Creating a more realistic schedule helps alleviate this issue, and this contingency must be factored into the budget.

Sitting down with the directors and producers to form an itemized budget is a tedious but necessary evil. Once a department-by-department budget is created (See the Sample Budget in Appendix A: Extras), the next task is to stay on budget! Again, whether you’re using a project management system or a simple spreadsheet to achieve this goal, the closer the team sticks to the budget, the better your relationship with the publisher will be. Ideally, the original budget would be the one to take the team through the entire production phase; realistically, though, there will be contingencies that cannot be completely anticipated. These usually revolve around labor, travel (including attending various conferences), and feature creep. It is an unfortunate reality that, no matter what development model the team utilizes, the budget will have to be revised a few times curing production. Again, to keep in the good graces of the publisher, it is a good idea to stay as close to the original projected budget and schedule as possible.

In addition to affecting the budget, the schedule will also be determined by the budget. Knowing in advance how much you will have to spend on your game will give you a great idea of how much time you will be able to afford in production.

2.7  Scheduling

Creating a detailed schedule is perhaps the most challenging part of getting the project underway. There will always be tasks that have slipped through the cracks, tasks that take much longer than planned, and dependencies that seem to never get finished so that the team can move on. These must be dealt with on a daily basis. Feature creep is the stuff producers have nightmares about and the bane of any production.

There are several ways to approach the creation of a schedule including the top-down approach, the bottom-up approach, and working with various constraints (time, resources, and money to name a few). Most project management personnel have their own favorite method. One of the most common ways to approach a schedule is to think of the critical path for developing the game. Once you have your chain of tasks and have assigned a duration/ period for accomplishing each of these tasks, you will have a basic schedule. Though the production may run for years, it is still necessary to have some kind of idea regarding when each major milestone should be reached. When determining these, get all of the department leads involved. Their experience will ensure that the schedule is realistic and achievable. It will also keep them from casting blame when a milestone suddenly seems unreachable and the team careens into crunch time.

Planning for contingencies will be another necessary task when preparing the final schedule. As the development of a game can occur over a very long period of time, many changes will occur throughout the studio. There will be holidays, personnel changes, vacations, conferences, crunch time (overtime), and training of new employees. There should be a contingency fund set aside for just these occurrences.

Creating a realistic schedule takes time and many meetings, so don’t forget to allow for this in the budget. Taking steps during the concept phase to document the scope of work involved with producing the proposed game will help with this. Rely on the experience of the senior team members to accurately gauge the amounts of time involved with producing each of the particular features of the game. A balance between concept and reality will eventually be reached.

At the minimum, an early schedule should have the basics listed with a set date: production, QA, and the Gold Master. Once these major dates have been determined, the production phase can be broken down into the various milestones that must be reached and the lists of criteria that need to be fulfilled before moving on to the next phase.

Once the schedule is finished, the game plan for development is wrapped up and preproduction of the game is officially finished. With the budget, schedule, and concept package firmly in hand, you can now move on into production.

Interview: Warren Spector, Founder, Junction Point Studios

Warren Spector has worked in the game industry for more than twenty years. After six years at Steve Jackson Games and TSR creating pen-and-paper games, Spector spent seven years at Origin Systems producing several addictive games, including Ultima Underworld: The Stygian Abyss, Ultima Underworld II: Labyrinth of Worlds, System Shock, Ultima VII Part Two: Serpent Isle, Wings of Glory, Bad Blood, Ultima Worlds of Adventure 2: Martian Dreams, CyberMage, and many more. A brief stint with Looking Glass Technologies was followed by a seven-year association with Ion Storm. After founding Ion Storm’s Austin studio in 1997, he directed the development of its genre-bending, award-winning game Deus Ex. He later oversaw development of Ion Storm’s Deus Ex: Invisible War, released in December 2003, and Thief: Deadly Shadows, released in June 2004. He left Ion Storm in November 2004 to found Junction Point Studios, Inc., which was acquired by Disney Interactive Studios in July 2007.

Though now a fixture in the electronic gaming world, Spector’s gaming roots are in the pen-and-paper game business, where he developed Toon: The Cartoon Roleplaying Game (among others) for Steve Jackson Games and at TSR, where he worked on the Top Secret/SI Espionage role-playing game, The Bullwinkle & Rocky Party Roleplaying Game, and the Buck Rogers Battle for the 25th Century board game, to name a few.

In addition to making games, Warren is a published novelist (The Hollow Earth Affair, with Richard Merwin, released in 1988), a film reviewer for the the Austin Chronicle, an assistant instructor for film and television studies at the University of Texas at Austin, and the author of numerous magazine and newspaper articles. From 2000 to 2002, he served on the Board of Directors of the International Game Developers Association (IDGA) and served as chairman of the IGDA’s education committee, forging ties between the game business and academic institutions around the world.

Warren was born and raised in New York City. He graduated from Northwestern University in Evanston, Illinois with a BS in Speech. He received his Master of Arts in Radio-Television-Film from the University of Texas at Austin and remained there to pursue a PhD in communications until the video game business lured him away from academia just a dissertation short of a degree. He is a booka-holic, board game fanatic, lover of basketball, and rhythm guitarist for the band “Two-Headed Baby”. Warren lives in Austin, Texas with his wife Caroline and far too many animals.

Newman: What aspects of modern/current game development trends have you noticed recently that are taking the industry in the right direction?

Spector: There seems to be a growing market for smaller, less expensive games that at least have a chance of breaking new ground, design-wise. When games cost $10 million, $20 million, and more, it’s hard to take a lot of chances. With all the major hardware manufacturers and PC folks offering downloadable games, there are more opportunities for commercially viable, “indie-style” games than ever before. That’s kinda cool!

Newman: What, in your opinion, constitutes a “cinematic” game?

Spector: My sarcastic answer is: “A cinematic game is one that doesn’t know what it wants to be and therefore shouldn’t be made.” Basically, while games and movies share many characteristics, it’s important—maybe critically important—to recognize, celebrate and exploit the things that make them different. Sure, movies and games can both tell stories structured in superficially similar ways; both can feature actors reciting lines of dialogue; both place a premium on appropriate camera positioning; and so on. But introduce real interactivity into the mix—which only games can do, and therefore must do—and the similarities start to look very superficial. There’s a gaming truism that the more cinematic you make a game, the less game like it becomes (and vice versa). That’s still true today and even if we can change it, I’m not sure we should. Games need to borrow the cinematic elements that contribute to a game-like experience, but going too far does violence to both media, I think.

Newman: How has your background in role-playing games helped you with game design?

Spector: Well, the truth is, I had to unlearn a lot of things, moving from tabletop role-playing games to electronic games. I mean, you don’t appreciate the power of a human game master and players who can improvise freely until you don’t have them! But what I did bring with me from the tabletop, face-to-face game world was an appreciation for the intelligence and improvisational abilities of players. They really do want to direct their own, unique experiences, and electronic games that “fake” interactivity (keeping players on rails, forcing them down a linear path)— those games are inevitably going to be inferior to games that truly empower players to make their own, unique choices—and deal with the consequences of those choices.

Newman: Another aspect of gaming that has improved is the storytelling. Has the game writing standard improved in your opinion and how has it affected the gamer?

Spector: I guess game writing has gotten somewhat better over the years. But, fundamentally, players play games to, you know, play. They don’t play games to read a bunch of text or wait around while a virtual character pontificates for hours. If game developers had any idea how to make conversation as dynamic and interactive as we make the more active, visceral aspects of our games, we’d really be on to something. So far, I’m not real impressed with the progress we’ve made on that front. Frankly, if I knew how to solve what I call “the conversation problem,” I’d do it! Given the limits of our conversation tools, I think the best we can hope for is that game writers will learn a lesson from screenwriters—those guys are masters, absolute masters, of brevity. Read a movie script and marvel at how much emotion and character can be crammed into a word or two; a gesture; a look. It’s incredible. Game writers have to write a whole lot less, if I can over-generalize a bit!

Newman: Over the years, you’ve been involved with quite a few successful games—the Ultima series and Deus Ex come quickly to mind. What characteristics mark a great franchise title?

Spector: I wish I knew! There’s a fair amount of bottled lightning involved! Having said that, I think it probably has to do with subject matter—either on the surface or as subtext—that people really care about. Your game has to be about something bigger than the minute-to-minute actions of the player. I also think you need a compelling central character. And you need a world that’s so well thought out, so well realized, that people can believe they’re really there. And, of course, the game play has to be rock-solid and innovative—we’re still a novelty-driven medium, more than a content-driven one. If you’re not offering players something they haven’t seen before, you have a real uphill battle.

Newman: Since consoles have added the ability for online gaming, the social aspects of being a gamer have improved considerably. How much impact does online content have upon the success of a game today?

Spector: No idea. I’ve only worked on one game that had any online aspect—the Deus Ex Game of the Year Edition shipped with a competitive multiplayer mode that we did as more of an experiment than anything else. I wanted to see if our unique avatar/ unique experience concept would translate to a different multiplayer experience than people were used to. I think it worked, for what it’s worth, though I don’t think many people tried it! Anyway, it’s certainly true that online aspects are an important bullet point on the back of a game box, and there are clearly lots of people who want a social experience as they play. I just hope that doesn’t supplant the single-player experience, which has its own pleasures and (still untapped) potential.

Newman: When Origin closed up shop in Austin, it seemed that the talent you had gathered there moved on to form the core for quite a few new studios. What “lessons learned” did you consider when you formed your new studio, Junction Point?

Spector: I learned a ton from the Origin and Ion Storm experiences—a book’s worth! I think I came to appreciate how important team fit was to success. And it’s critical to have team members and publishing partners who believe in your mission. Without a positive culture and committed collaborators, you can’t succeed. That’s probably the most important thing I tried to bear in mind when I started Junction Point.

Newman: Now that J.P. is affiliated with Disney—most known, of course, for films—has the game development style of the studio become more cinematic? What great practices has the studio added from Disney?

Spector: I don’t know that we’ve become “more cinematic” in our thinking. I mean, if you check out our Web site you’ll find an abridged version of our mission statement. (The longer version can be found on my blog site: http://junctionpoint.wordpress.com/) That mission hasn’t changed at all since we became part of Disney. In fact, I’ve always made a point of reminding people I work with—including Disney, now—that if they don’t want games like the ones I talk about in the mission statement, we probably shouldn’t work together! Having said that, we’re all psyched about how open the guys at Disney, on the film side, have been about sharing their expertise with us. I mean, there are ways in which games and movies are similar, and collaboration can make both media stronger. Working with the Disney film guys has already been an education for me and I don’t see that changing—gotta keep learning!

Newman: What advice would you give a young producer/director just getting into the game development industry?

Spector: Well, get it out of your head that you’re going to start as a producer or director, for starters! Figure out where you can really make an impact in the trenches of game development. You need to prove yourself as a designer or artist or programmer or audio person or tester or something before you even think about moving into a larger creative or management role. Remember, always, that ideas are the easy part of game development—execution is insanely hard. Until you’ve been through it a few times, you can’t even imagine! So learn as much as you can about how games are made; figure out what aspect of game development you love— what you’re really, really good at (it’s an ultra-competitive business); find developers or publishers who make the kind of game you love to play, and don’t give up until you get your foot in the door. On a somewhat related note, if you’re in school, stay there. A broad-based education will serve you well in this business.