Hollywood Sound: Part Three

The spoken word has always been a fundamental part of the communication of story, information and emotion in video games. Over the past twenty or so years the degree to which the word, voice, and more recently the ‘actor', are involved with games has dramatically increased and shifted towards a Hollywood model. Most big production games now include huge voice-over budgets, utilizing Hollywood's talent pool. In this third and final part of our look at Hollywood sound integration into video games, we examine the way games are written, cast and how voice is integrated into gameplay.

Early Text Story Telling

Traditionally games have been designed around a combination of simple gameplay elements, from early platform games such as Donkey Kong and 3rd person combat games like Street Fighter all the way through to driving and flight simulation. These early games, even though they produced sound effects and music, could be thought of as ‘silent film' in that they had no voice, only text-based ‘intertitles'. The stories and characters in these early games are very simple and therefore very quick to pick up and play. You are introduced to a character, more often than not in arcade games by the art work on the side of the arcade machine. The early artwork for the Space Invaders video game, for example, was all that was needed to communicate the story and the action a player could expect. This carried through into the first home computer game systems, the cover art on the cassette tapes and cartridges communicating the majority of the action and story information. Basic story began to be elaborated upon by some introductory text on the inside of cassette covers. In Software Project's 1984 game Jet Set Willy, the story runs as follows:

Based on a hallucinogenic hangover, the objective of Jet Set Willy, which turns out to be an almost incompletable and long game, is very simple: to collect all the bottles and glasses from the various rooms in the mansion. You are vaguely introduced to back characters through the text, of which only Maria the housekeeper is actually in the game. This kind of information is designed to be read as the game loads up, an activity that would often take up to four minutes. Even these earlier narrative forms communicating game story, though, connect to cinematic traditions. This style of intro story is comparable to the opening text we see in Star Wars for example. In the case of Jet Set Willy, the player is posited in third person, controlling the character of Miner Willy. Second person examples of introductory narrational text also exist, as in Saboteur from Durrell in 1984 which follows a similar quick injection of back story:

Here, story, then objectives, are revealed. A second person narrative is established where ‘you' are directly addressed in the form of a mission briefing. Both of these second and third person positions are still used in intro movies today.

The notion of establishing story very quickly so that the player can pick up, play and be immersed in the game is something that still very much exists. The evolution of the ‘intro movie' or intro ‘Full Motion Video' (FMV) is born out of this need, and, as hardware and software power has increased, the sinews connecting games to Hollywood film are becoming stronger. An intro FMV (or Non Interactive Sequence (NIS) using the game engine to render the movie), today consists of a 1-2 minute prologue in which the player is immersed into the action through cinematic storytelling techniques; sound, dialogue and music are all used in exactly the same way that they are in any Hollywood film, so that when the player actually gets into gameplay, they understand the motivations of the character and of any back story. This is all done for the purposes of immersion; story and motivation first, then objectives later. Objectives of actual gameplay are left out of the establishing intro movie and are contained in either a more informational tutorial mode, or are often revealed as you play through the game.

Writing the early back story used to be simple, and anyone could do it as it required little more than creating some very basic motivational information that didn't even need to relate too closely to the characters or experiences of gameplay, as seen in Jet Set Willy. However, with the arrival of the spoken word to video games, the professional skills of the writer and the actor have become critical to production.  

The Arrival of the Spoken Word

The technical limitations of the early home PC systems such as the ZX Spectrum and the Commodore 64, and early consoles such as the NES, resulted in some creative ways to employ voice in order to both market and bring a cinematic edge to early games. Usually these attempts used the B-side of the tape, or a supplemental tape that contained music or a theme song, or even full blown, slickly produced ‘interactive' content.

AutomataUK's ambitious 1984 release Deus Ex Machina, for example, featured a tape that you played along with the game. This tape consisted of a full soundtrack with music, sound effects and voice narration by John Pertwee, Ian Dury, Donna Bailey, Frankie Howerd, E P Thompson and Mel Croucher. This is more along the lines of a progressive rock concept album than a video game of the time, and represented a glimpse into the future; albeit rather more abstract in its nature than a Hollywood action movie. It basically introduced the notion that the story will move forward beyond the introductory plot line and be developed with equal production value as you play through the game, and will conclude in a suitable narrative manner. The interactive nature of this content was by today's standards simplistic and along the lines of ‘press stop now and resume the tape when you are given on screen instructions'. However, this is one of the only examples in early video games of syncing of several different media. This bears a striking technical and aesthetic relationship to the way that early ‘silent' film attempted to create sync sound by either piano accompaniment or sound effects created live to the picture. (3)

Stereo sound, and, in fact, the human voice, were not featured in games until the arcade game Dragon's Lair in 1982, voiced by none other than the very Disney animators who made the game. Actual in-game, interactive voice samples were not fully developed until the arrival of the NES and Amiga, and eventually, movies streamed from CD-ROM discs enabled a more cinematic production value. The appearance of Mark Hammil and Malcolm McDowell et al, in 1994's Wing Commander III and the Wing Commander series helped to give birth to the FMV movie and the notion of cinematics sequences in video games. These movies did tend to leave somewhat of an obvious gap between the live action movies and the gameplay experienced by the player; however they pushed the bar further than anything heard previously in terms of immersing the player into a tangible universe populated with believable characters. In 2002, Rockstar Games pushed the use of star talent even further into the AAA range with their title Grand Theft Auto: Vice City, in which Dennis Hopper, Ray Liota, Burt Reynolds, and Deborah Harry among others, appeared as character voices.

Star Acting Talent

The sudden influx of Hollywood talent from all corners of the production environment; dialogue, animation, visual effects, and music composition, implies a large increase in production value. However, there are still some areas where the skill-sets haven't quite ported over as successfully as others. The music composition side is working out well, with new structural languages being defined to accommodate already existing composer's models and work flow. The dialogue side, on the other hand, seems to have been in a transitional phase despite the occasional caliber of GTA voice talent; however, over the last five or so years this has started to change. The changes are very much down to how well the writer understands the needs of gameplay, and how much money is invested in actors who can make the lines come to life while incorporating a degree of improvisation. Currently an audience seeing the cinematic trailers flaunted by next generation titles are expecting nothing less than a motion picture experience from a video game that cost around $60. The notion of getting properly trained and paid actors onto an interactive project is still a relatively new thing for most developers to deal with. Around ten years ago game developers tended to get their friends in to do the voices for game characters, or to employ cheap local student talent, both of which, while cost effective for production, would invariably undermine the believability of the finished product, and more often than not result in the long-term failure of the game. This attitude, to some extent, still exists in some dark corners of game development, with those who don't understand audio and its relationship to immersion and all other aspects of gameplay. In terms of the actors themselves and their representation within the games industry, the Screen Actors Guild (SAG) now represents a large portion of the acting talent in the USA and is creating representation ensuring scale deals that, although not equal to the deals they get in motion pictures, are nonetheless big figure deals.

The importance of having star actors is beginning to be widely recognized; as well as adding quality to the finished product, their name association allows a great deal of additional marketing to occur, which in turn enables the creation of a certain amount of excitement around a project. The marketing happens in exactly the same way that a movie uses a star name and the model is being refined and based on movie marketing even more as development moves into next generation consoles.

Production: Story, Character and Gameplay

Emotional Objectives

The different types of story and character requirements vary greatly based on the genre of game being created. Linear, story based games are far easier to make cinematic than an open world game, as they are gated, closed world, mission by mission based games. Much of EA's non-sport output, for example, relies on this linear progression in order control the cinematic flow of the story. There is currently a move to a gameplay where both open world and linear story concerns need to be met. In the GTA series of games, we see the co-existence of a two-fold story mode: a linear story which can be picked up at any stage is surrounded by open world gameplay and ancillary missions.

There are clearly greater concerns with connecting gameplay with story than there have ever been: a story and its characters must makes sense at all times, the player must be aware of motivations, and also of game objectives. There is certainly a cinematic trick for doing this, and that is to pair both the informational with the emotional. If a game objective must be met the player will be given an emotional, character-driven motivation for doing this. If done well enough, characters and story will be strong enough to give the player an emotional motivation to action.

Production Timeline

Understanding what needs doing and when is crucial in developing voice content in any game, whether it is a 50,000 line epic or a 1,000 line platformer. Each project will represent a different challenge in terms of dialogue, both structurally and also technically. Platform-specific issues such as RAM and media type will determine the amount and quality (sample rate) of dialogue files the developer will be able to use. Not only this, but the style of game will also have expectations which will more than likely push those limits with each successive project. Generally speaking, an era of greater content and greater interaction is expected to occur with next generation consoles.

With game play, story and characters needing to be so closely interlinked, there is evolving the need for a closer link between writer and game designer. It is one thing having a game with no story, and quite a different thing when you have a story but need to design a game around it. But, more often than not, the development cycle of a project places demands on both story and game play. Going back to the model of development first mentioned in Part One of this feature with regards to music (figure 1), we can see how the story and game play development needs occur in parallel:

Adaptation Writing for Interactive Media

An area now rapidly expanding is the incorporation of big name writers to provide the story line for video games. What the writers currently provide for games is, of course, based on what they would provide for a film or TV script, with much of their language and direction of the former media still intact. This is comparable to a film-only composer delivering a linear score if they did not first understand the interactive structure of the game. However, being able to understand the game content and to craft a script that satisfies all the ancillary character dialogue, cinematics and gameplay demands of an interactive title is something that perhaps writers for traditional linear forms may not yet be prepared for.

In film or TV a screenwriter is usually employed and given the job of adapting the story content into useable and filmable dialogue and stage direction. The 'adaptation writer' for interactive entertainment is a role that quite simply does not yet officially exist. This person would fill in the gap that currently exists between integrating the story and plot into the game design and being able to make changes in both. The person also needs to be able to rework character's dialogue to create lines for them to say that are story and character related. Such typical content would be generic lines that the character says when standing on a street corner, doing nothing in particular, or in a more extreme example, being shot at; these are key areas in games where dialogue often fails, and where it often becomes annoying. Some of the more successful titles in terms of ‘film style' adaptation in dialogue are heavily scripted and linear in that they use dialogue as a scripted ‘game moment'; EA's Lord of the Rings trilogy is one such example. This suggests that it is easier to adapt a pre-existing story into a linear game. However, the challenges involved in creating both a story and an open non-linear world require an extra special writing talent.

The adaptation writer would be heavily involved in facilitating the script between the design team, sound director and the storywriter, making sure that the content that is created by the writer fits into any particular game mechanics that have already been defined. They could also find themselves adapting already existent content, from a novel for example, in which a story and characters are already clearly set out, yet creating the in-game dialogue and cinematic assets from that content to fit into exciting and dynamic game content. This is a job that at the moment is being done by a wide group of stakeholders, usually falling onto the designer's or game director's plate.

Where these personnel will come from remains to be seen. It could equally be conceived of as a job a designer with a particular penchant for dialogue and writing could rise to and specialize in, and equally it could be a role that a film or television screenwriter could adapt to as a freelancer. What is essential, in wherever they come from, is that they both understand games and have experience with the high caliber of talent and performance expected by Hollywood .

Multiple Script Views

In Figure 1, the ideal situation, the overall story arch is figured out before production begins, in the first two stages of pre-production and proof of concept. One of the key production tools throughout the whole process is the script. Character dialogue is not adapted for production until the end of the tertiary phase. The actual script will require several different people to be working on, or from, it. This implies that the information should be viewable in several different modes.

View 1) story writing mode, which may be a simple linear story with character names and action description;

View 2) character mode, which would be a view where each character's dialogue can be viewed and worked on;

View 3) VO Production view mode is a printable view of the sheet that allows the dialogue to be read and recorded by an actor. For this they will need to see the context of their cinematic lines (i.e. be able to see the lines other characters say before or after their line, and also any exclusive lines their character has;

View 4) A final view mode would be for subtitling and translating of the content into different target languages. A flexible system that allows all these different views and crafts to be easily applied to the same content is essential if the game is to feature large amounts of dialogue.

VO Recording Hollywood Style

In terms of actual recording, many film studios now cater for game audio recording sessions as well as film. Typically under used in film, ADR stages are ideally suited to be used as the venue for the production of mammoth amounts of game dialogue content. Game audio is an interesting market to film post-production executives. They see the amount of lines required and the caliber of acting talent involved and they see the revenue. This is another positive move for Hollywood into video game production, as the higher end studios are far more suitable environments to record star voice talent as they already have the high-end client service culture worked out from their work in film. They are also already geographically situated in areas celebrity talent can easily access, such as Soho in London, Burbank and Hollywood in LA. All these things put established film VO companies in the leading edge position for marketing themselves towards ‘Hollywood Game Production.'

Performance Capture

With the advancement of animation streaming technology and hard disk space on next generation consoles, not only will the sheer amount of voice content increase but so too will the ways in which they are produced. A qualitative production element to consider is the fact that voice-over recording is starting to become entrenched as part of motion capture shoots (and vice versa), and is therefore bringing together the artistic disciplines of both physical visual performance and audio performance. The separate disciplines that were once covered by a dialogue director and a motion capture director will eventually become the remit of one director who will direct both the visual physical performance and the dialogue performance. Whether a ‘director of performance' or a ‘character director', this person must have experience, skill sets and understanding of both fields, technically and aesthetically, and cannot neglect either field at the expense of the other. The key is understanding that the actor's performance is being captured as a whole, both sonically and physically in one session. It is easy to imagine the flip-side to the trend for the removal of star actors with poor speaking voices when sync-sound film arrived: those actors who are not able to physically act in motion capture situations may require a stand-in to do redo physical work. It is safe to say, productions will more than likely favor actors who can do both at once.

The voice in video games is entrenched in filmic techniques, aesthetic approaches and production models; however, there will always remain specialization in terms of integrating and conceiving of the content for interactive media. The more the people with the specialist skills can make use of the established talents of Hollywood, the closer the two media will become in terms of ‘production values.' However, in terms of production and implementation itself, the models will always remain distinctly different. It remains to be seen whether this initial relationship with Hollywood will last in the game industry. Provided there is still a strong return in revenue, the sinews will go from strength to strength. The future certainly holds change in both linear and non-linear media production environments; however, one thing is certain: the dominance of the Hollywood mainstream model is set to continue.

--

Notes

1) Jet Set Willy (c) 1984 Software Projects Ltd by Mathew Smith ftp://ftp.worldofspectrum.org/pub/sinclair/games-info/j/JetSetWilly.txt

2) (C) DURELL SOFTWARE 1985, SABOTEUR by Clive Townsend ftp://ftp.worldofspectrum.org/pub/sinclair/games-info/s/Saboteur.txt

3) Richard Abel, Rick Altman (eds) ‘The Sounds of Early Cinema', Indiana University Press, 2001.