Storytelling in games as compared to film
Inspired by cinema, game developers take this understood medium and apply its learnings to a vast, unmapped frontier. They are faced by a new set of challenges: giving players control over narratives, introducing them to systems and ways of play that aren’t immediately intuitive, and letting go of authorial intent. This dissertation will explore and understand the ways the principles of cinema apply to games, and how games are developing those principles to create cinematic experiences for the new age.
The three keywords of this dissertation should be defined so that no confusion may arise from their application. Image will be used to talk about the visuals the player is shown, and the key tenet of cinema: visual grammar. Story, then, will be in relation to the narrative design (a concept unique to games; the way a player is fed narrative through ways other than image) and structure of narrative. Finally, experience, will be used in relation to the emotions invoked through control schemes, decision making, and other aspects of interactivity and gameplay games provide to shape the player’s experience.
Chapter 1 — The Cinematic
Interaction is what sets games apart. This doesn’t, however, mean that games are separate from other artforms. Cutscenes take control away from the player and follow the conventions of cinema. With the progression of technology, cutscenes can be performed by professional actors, motion captured and animated. Developers can choose to use advanced lighting simulations, mimic the imperfections of cameras, and even record motion capture footage with virtual cameras.
This chapter will explore a brief history of cinematography in gaming: how modern developers take inspirations from the medium of cinema and apply it to interactive medium of gaming.
Additional Nomenclature
To talk about image in video games, I first must outline and explain the ways image is presented to the player. Three terms I will use from here onwards are:
Cutscene — “any non-interactive storytelling or scene-setting element of a game” (Hancock, 2002).
Third Person Perspective — Follows the player character during gameplay. Much like a film camera, these can tilt, pan, and zoom.
First Person Perspective — The player assumes control of the character from the character’s perspective.
Influence of Cinema on Games
Hideo Kojima, director of the hugely successful Metal Gear Solid series, doesn’t hide the influence cinema has had on his work. If anything, he embraces it, his Twitter profile proudly stating: “70% of my body is made of movies.” (Kojima, 2010). His 2004 release, Metal Gear Solid 3: Snake Eater opens on a clear pastiche of James Bond — a title sequence directed by Kyle Cooper (Rogue One, mother!, Seven), no less. The main theme? A haunting tribute to Bond’s female vocals, performed by Cynthia Harrell (Hibino, 2004).
Kojima cites Hitchcock’s North by Northwest as a huge influence on Metal Gear Solid (1998), stating “It is Hitchcock’s use of this first-person view that resembles one of the key features in MGS.” (Kojima, 2003). He notes that, due to the player being able to freely switch between first and third-person (bird’s eye view) perspectives “it was difficult to direct certain scenes as [he] would have wished” (Kojima, 2003). Metal Gear Solid employs another perspective, however, dependent on the player’s context. If the character leans against a wall, the camera swings down to the character’s level, providing a clearer view of what lies around the corner. Kojima goes on to say that “…it is by limiting on-screen information with these three cameras that helps create a feel of menace.” (Kojima, 2003).
Filmmakers without Cameras
Dori Arazi, during a talk about his time as Director of Photography on Sony’s 2018 game God of War spoke of the challenges he encountered (Arazi, 2017). Coming from a film background, Arazi says a knowledge of optics was crucial to his success. As all his work was virtual, he and his team first had to teach their software to interpret the data their virtual camera recorded. This required them to understand how each lens distorted the image, and how different lens worked under varying lighting conditions. His team was able to craft a 15–85mm zoom lens with a 1.8f aperture, something not yet possible with any lens on the professional market. While this is any cinematographer’s dream — allowing for huge freedom in framing — Arazi mentioned that the PS4’s computational power is a limiting factor when it comes to lighting. With each extra light added, the processor’s load is multiplied exponentially, and so Arazi had to use each light as effectively as possible.
The benefit of a virtual camera is that heavy equipment isn’t necessary for intricate camera moves. The animation department can take over from the virtual camera under Arazi’s supervision. Dollies, tracks, jibs, and even shots that would require use of a drone can all be easily animated after the fact. Arazi made sure to highlight the fact that a virtual camera is not actually a camera — it is simply a way of recording motion data that can then be manipulated in post-production. Camera shakes could be eliminated, missteps in the actors’ blocking, or even whole actions could be altered and hand animated. One could claim that this is where the uncanny valley comes into play as characters actions become too perfect, or too rehearsed. This, though, is an unfair criticism to make; an incapable or inexperienced animator would commit these mistakes, but an incapable or inexperienced actor would also make the same mistakes.
Games give Directors of Photography (DP) greater control over the finished product. Much as a film DP would need to location scout, Arazi had to collaborate with level designers and concept artists to create the virtual world the camera would inhabit. This does, however, require a wider skillset: Arazi quoted his history in animation and design as a help — knowing how an animator works and communicating his needs was easier as a result. Rather than finding inspiration through location scouts, Arazi had to inspire: creating sets that the Designers were happy with — ones that could accommodate battle segments, puzzles, or whatever other gameplay mechanics they wanted to implement — but also ones that enabled the Director to lead the action in cutscenes.
Coverage (a film term for capturing action from several angles) is never a problem. As motion capture studios capture data from tens of angles, the placement of the camera becomes something the developers can change during production. With the ability to seamlessly merge takes or hand-animate minor details, the team can easily pull out the best performances without having to worry about hiding cuts, or editing in general.
Arazi cites the game’s interactive camera his main challenge: it was difficult for him to give up his authorial intent. Working with designers and animators they decided that, though the player has ultimate control over the game’s third-person perspective, there were animations and combat scenes that the team wanted to make sure players experienced. When Kratos goes in for a brutal killing blow, the camera punches in on the action as seen below in Figure 2, then goes back to its neutral position for the player to keep going.
These techniques combine the interaction granted by the medium as well as cinematographic principles, mixing gameplay with a dynamic camera that reacts to the players inputs, while still producing an image that emphasises the action and has intent behind it. It’s a display of the unique workflow of the games industry: the DoP has control over the image while working closely with the animators and designers and having a grasp of all departments.
Cinematics as Loading Screens
With the advancement of technology, cutscenes can fulfil more than one function. A cutscene is a perfect time to flush the computer’s short-term RAM (random access memory) of the previous level and use it to load the assets needed for the next area. This rids games of loading screens which notoriously grind pacing to a halt. Naughty Dog takes these up a notch though. Uncharted 4 (2016) combines cutscenes and gameplay during moments of tension to slow the pace, mask loading of assets, and trick the player into believing they have the ability to affect the outcome of the situation. Take the scene in Figure 2 (below): following a lengthy car chase during which the player is jumping between cars and tearing through Madagascan countryside, the player’s car crashes and the main character, Nathan Drake, becomes trapped underneath its flaming wreck.
This is, in fact, the midpoint of the sequence. The player controls Drake’s arms as he’s trying to pull himself out of the vehicle (Figure 3 — notice the two bars to his left and right that both have red tape wrung around them — intended to draw the player’s eye). As stakes are high, the player will try to be as quick as possible, but Drake’s animation cycle and the rate at which the car sets aflame is well-timed. Despite the player’s actions, the scene plays out at exactly the same pace every time. The timing not only sets up tension, but also allows enough time for behind-the-scenes processing to load the next scene’s assets and allow the player to keep playing uninterrupted.
It’s a clear aspiration to be more filmic — a short breather in an action-packed sequence, only to be thrust right back into the action.
Why not make films instead?
Criticisms levied towards Kojima and others of his philosophy place them as unsuccessful filmmakers making do with second best. Metal Gear Solid 4 is infamous for its ~8 hours of cutscenes, holding two Guinness World Records: “Longest end sequence in a video game” at 1 hour 9 minutes and “Longest cutscene in a video game” at 27 minutes (Guinness World Records Limited, 2018).
This opens an interesting debate: are purely gameplay-driven games a la Half-Life 2 (2004) which feature no cutscenes more deserving of the “video game” title than ones like Metal Gear Solid 4, which, as mentioned, contain almost 8 hours of cinematics? Davey Wreden’s The Stanley Parable (2013) and The Beginner’s Guide (2015) are derogatorily labelled as “walking simulators”. They feature no cutscenes, nor do they have involved gameplay mechanics: the player simply navigates the environment, taking in the narrative delivered through text, environment, or narration. Without a fail state, can these pieces of work also be called games?
This is an old-fashioned train of thought: avoiding the fail-state was the player’s objective in arcade games like Pac-Man (1980) or Donkey Kong (1981). The maturing of games as a medium has led to new sensibilities: the objective of “walking simulators” is narrative, and Dear Esther (2012) developer Dan Pinchbeck feels critics think that “games have to be defended from innovation” (Pinchbeck, 2016), and that their criticism often “starts coming apart at the seams” (Pinchbeck, 2016).
This is a positive to most: games as a medium are still a young frontier. Developers can freely experiment with structures and storytelling methods.
Procedural Camera Systems
CD Projekt Red’s The Witcher 3: Wild Hunt (2015) is a huge game: its 450,000 word script (Stein, 2015) rivals most novels — Tolkien’s Lord of the Rings trilogy totals 455,000 words — and to animate the game’s dialogue was a gargantuan task. As The Witcher is an open world game — players are free to explore and interact as they wish, and in whichever order they wish — the team decided to automate this process. Cutscenes for the main story were all hand crafted, but side content was mostly created through a ‘procedural’ camera. The team created a template: the player character, Geralt, and the people he talks to have a pool of animations the system would randomly select from, ranging from simple hand gestures to body movements and expressions. Alongside a camera system that would select the best of pre-made angles (with criteria like: is the lens free of obstructions?) the team could quickly animate thousands of conversations between Geralt and other characters. They could then go in and quickly polish the scenes, work out kinks that came up — no automated process is perfect, after all — and save massive amounts of time.
Chapter 2 — Interaction
Interactivity is the key differential between video games and cinema. Through interaction, the player can affect image, story, and their experience.
Half-Life 2’s Invisible Tutorials (Brown, 2015)
Valve Corporation’s Half-Life 2 (2004), a “first-person shooter” game, uses all three keywords to immerse the player in its world.
Analysis by essayist Mark Brown explains a scenario in Ravenholm, an abandoned village, the player comes across. As the player enters a house, their eye is drawn towards the grotesque scene on the back wall (Figure 1) directly opposite the entrance (Brown, 2015). A zombie corpse has been split in two with a saw-blade. There are two conclusions to draw from this: the cause of Ravenholm’s abandonment is made apparent (story), and the player is shown that the most effective way to fight the creatures is through use of saw-blades (experience and gameplay).
The only way forward is through the door blocked by the table on the left of Figure 1. As the player approaches and uses their Gravity Gun to pick up a saw-blade, a zombie is revealed (as shown in Figure 2). The only way out, then, is to use the saw-blade as a projectile weapon. And so, without using dialogue, text, or cutscenes, but instead using cinematic principles like leading lines and colour theory (note the generous swathes of blood in Figure 1 and bright light in Figure 2), Half-Life 2 tells a story and teaches the player a core gameplay mechanic.
This kind of environmental and visual storytelling is an interesting field, venturing even into psychology as Naughty Dog developer Emilia Schatz claims: “You need to figure out what your environment is telling the player, and figure out how you can give the player as much information as possible so they feel very informed — but at the same time influence their decision to be the right one.” (Schatz, 2014). It is not fully unique to gaming, however: this technique is reminiscent of the Kuleshov effect — the idea that two sequential shots can alter the viewer’s perception of emotion portrayed by an actor — through subtle environmental clues the designer is able to influence the conclusion a player will reach.
Empathy through Gameplay
Naughty Dog’s The Last of Us (2013) is a great example of a game that uses its mechanics to create a bond between player and character. The narrative is set 20 years after a global cataclysm. The characters struggle to survive in a hostile world, scavenging for supplies. This translates to gameplay: the player will go into combat armed with scraps; a splintered baseball bat or 3 rounds of ammunition. Faced with large groups of enemies, the player must make real-time, desperate decisions on how to use their limited supplies. Using them now means they won’t be available later — who knows what lies in wait just around the corner? The player, then, is thrust into the same position as the characters, as a single mistake could spell disaster.
Conversely, Red Hook Studio’s Lovecraft-inspired Darkest Dungeon (2016) uses this desperation to make the player the villain of the situation.
The game places the player in the shoes of a baron of an abandoned estate. To rebuild, whatever cosmic horrors lie within must first be purged. The player recruits adventurers, using them only as a means to an end. We are presented with two main statistics of each adventurer: Health and Stress. As the player sends adventurers out on successive missions, their stress value rises and the likelier they are to suffer Afflictions, such as ‘Paranoid’, ‘Hopeless’, and ‘Masochistic’. After suffering an Affliction, characters that suffer more stress have a chance of being struck by a heart attack and dying permanently.
Characters suffering Afflictions might lash out against the player. These subversions are usually harmful: a Masochistic character might hurt one of their fellow party members, while a Hopeless character might run to the back, ruining the party’s tactical composition and disabling certain choices for the player. The game, at this point, takes control from the player. Instead of using cutscenes, though, characters act on their own, pushing the idea that they are independent and rebelling from their commander.
Another essay by Mark Brown calls this “Morality in the Mechanics” (Brown, 2016): the player is encouraged to push adventurers to point of extreme stress, of mental illness, and then discard them once they are no longer of use. The aim of the game, ultimately, is to use others for personal gain — shedding humanity in favour of greed. The game does not explicitly ask you to make a black and white moral decision, but the most efficient way of playing is by throwing away tired adventurers, bringing on new ones, and repeating the cycle. Healing broken characters takes time and money — it’s up to the player if they can spare any.
It’s also important to note the camera perspective of Darkest Dungeon. It is neither first nor third-person, but seen side-on, as illustrated in Figure 3. There’s a further separation from your adventurers — a side view is impersonal, and places them on the same level as enemies they encounter. The player has constant view over their statistics: Health and Stress. It’s dehumanising — the adventurers are just pawns in a game.
Lucas Pope’s indie hit Papers, Please (2013) takes a different approach: the player takes the role of a border crossing immigration officer, wherein the gameplay revolves around granting or denying people access to the fictional country of Arstotzka. The player is given different objectives each day, and more complex ones with each passing day — if a neighbouring country declares war on Arstotzka, the player is not allowed to grant access to the citizens of that country. Deadlines get tighter each day. Bribes, sob stories, and threats all come flying towards the player, and only the player can make the decision on their course of action. At the end of the day the player earns money for their work, spends it on rent and medicine (if needed) and continues on. It is up to the player to decide how corrupt they want to be, and how far they will go to feed their family.
Film grants us the ability to empathise with characters through the actors and their expressions. As social creatures, we react empathically to the feelings of those around us. The filmmakers take us on this journey, however, while the mechanics of the aforementioned games push the player into the role of the arbiter and catalyst rather than a mere observer.
Chapter 3 — Narrative Design in Linear and Non-Linear Worlds
The focus of this chapter will be on comparing the ways narrative can be communicated to players in both linear (wherein the narrative progresses in a set order as the player does, and is most film-like) and non-linear games (wherein the player is granted some degree of freedom and agency in approaching the narrative), as compared to that of films.
Ludonarrative Dissonance
Clint Hocking, former creative director at LucasArts and Ubisoft, coined the phrase “Ludonarrative Dissonance” while discussing Bioshock (2007). He claims that “By throwing the narrative and ludic elements of the work into opposition, the game seems to openly mock the player for having believed in the fiction of the game at all.” (Hocking, 2007). Going back to Uncharted 4, the protagonist Nathan Drake is portrayed as an everyday man: his adventures are punishing and tiring, firefights leave him stressed, and he’ll often joke and quip at his situation. The dissonance, then, comes from the fact that by the end of the first game of the (at time of writing) 4-game series, Drake has shot, maimed, and injured more people than a trained Marine would in their lifetime. Drake’s constant portrayal as a next-door-bloke becomes dissonant when, by the fourth game, Drake is handling full-automatic weaponry and fighting off hordes of paramilitary forces.
This is an important definition: it’s a criticism unique to video games arisen through combining gameplay and interaction. Films don’t experience this kind of problem: theme is shown through narrative and imagery, while the pursuit of “fun” is a key principle in the design of games.
Campo Santo’s Firewatch (2016) is a good example of a game that overcomes this dissonance. The story follows Henry, a man escaping the tragic reality of his ill wife and dissolving marriage as he ventures out into the American wilderness as a fire lookout. The gameplay revolves entirely around exploring the wilderness and interacting with Delilah — a fellow fire lookout whom we never see — over a walkie-talkie Henry carries everywhere with him. The story takes a turn toward thriller as mysterious figures seem to haunt the two over months. This is eventually resolved quite anti-climatically, a commentary on the horrors the mind fabricates when in isolation for extended periods.
Narrative Structures
Tom Abernathy of Riot Games and Richard Rouse of Microsoft Game Studios conducted a talk at the 2014 Game Developers Conference named “Death to the Three Act Structure!”, decrying the application of the classic Artistotelian three act structure (protasis, epitasis, catastrophe — ie. setup, confrontation, resolution). They note that the structure can be almost applied to a cutscene and narrative-heavy game like Uncharted 2, but that the serialised TV drama (à la The Sopranos (1999), Breaking Bad (2008), and Game of Thrones (2011)) narrative structure is a lot more convenient for games: “…showrunners have [to think about] structure on every level of magnification: … structure of scenes, about the sequences that scenes make up, the acts between ad breaks, half or even full season arcs…”.
A good example of this is Telltale Games’ The Walking Dead (2012). The game was released episodically; episodes would come roughly two months apart, with five making a whole season. This meant that designers had to deliver a satisfying narrative within each 2–3 hour episode that players could ruminate upon between episodes, but also push forward the narrative that spans across the whole season. While this is fine for a game with an episodic release window, the two then talked about Naughty Dog’s The Last of Us (2013). The game is quite neatly divided into chapters — each named after the season during which the year-long narrative takes place — and each chapter has self-contained arcs with characters that appear and disappear, as well as arcs that span the whole game.
But how does this apply to open world games?
Josh Sawyer’s talk on narrative design in Fallout: New Vegas (2010) focuses on the arcs of characters within quests and player agency. Players want the feeling of agency upon the world and of their character being unique, and this was more important than structure. For this, the team at Obsidian Entertainment had to make dialogue options according to a set of guidelines they made: providing tactical and strategic choices, forecasting the consequences of their choices, and validating all options to avoid “win” or “loss” choices (Sawyer, 2012). The latter was most important to them — having choices that were clearly the best would punish players that weren’t playing according to that arbitrary standard. However, validating all choices was a huge amount of work, and so the studio created a “Reputation” system that would keep track of the player’s attitude towards all factions. This meant that forecasting the consequences was a lot easier — they could use the player’s reputation to write further encounters.
The conclusion of Abernathy and Rouse’s talk is a direct counterpoint to someone they quoted at the very beginning of their talk: “Screenplays ARE structure.” — William Goldman, and propose that “Game stories are NOT structure” (Abernathy & Rouse, 2014). There is no globally applicable structure for games.
The two quote research conducted by Deborah Henderson of Microsoft, stating that gamers find it a lot harder to retell the narratives of their favourite games, but consistently remember characters and their characterisation, not necessarily their role in plot (Henderson, n.d.). For this reason, they conclude that game narratives should instead focus on character first, aligning character motivations with player motivations. This is another unique challenge for games: artists, writers, and designers can concentrate on providing memorable and interesting companions that join the player on their journey, and can be a vehicle for interaction, exposition and narrative. Combining narrative, level design, and gameplay mechanics, leads to a structure that is custom-made for the game and its requirements. If this leads to a three-act structure, fine, but developers get a lot more freedom in structuring their story.
Diegetic Worldbuilding
Designer Jurie Horneman explores the design of narratives outside of cutscenes. In his 2015 Game Developers Conference talk, he defines two important terms for games: “story” — that which is pre-authored, about something happening to someone, and “fictional” — related to things the game is asking you to pretend are true (Horneman, 2015). He explains that the mechanics of games are abstract: health points, failure states, player character deaths are all abstractions the player accepts to understand the game. However, that what is fictional, then, is also that which is meaningful. Every game design element can be seen from the mechanical and the fictional side.
It is the narrative designer’s job, then, to explain to the player mechanics in a diegetic way, thus diegetic worldbuilding. Most games solve death with a checkpoint system: the player simply resumes from an earlier point and attempts the challenge again. Horneman’s example of Prince of Persia: The Sands of Time (2003), explains this with a framing device: the titular Prince and player character, is retelling the events of what happened, and any mistakes the player makes are framed as a mistake in his story, lapses in memory. Health points are a game staple, and Half-Life 2 (2004) uses the player character’s exo-suit as an explanation for his physical wellbeing being numerically quantified. Removing a staple such as this, though, like Ico (2001) is a statement in itself, and redefines the gameplay in that game. The fail state of Ico isn’t the player character’s death, but the capture of his love interest by shadowy creatures. The creatures are not bothered by the player character, they simply slow him down. With minimal changes Ico recontextualises the game’s objective: do whatever it takes to keep the companion away from danger.
Ultimately, little can be said on the narrative design of games as compared to cinema. The games that feature cinematics heavily take inspiration from cinema and, in majority, tell linear stories that do not need to adapt to the player’s actions. In this, developers have to be wary of ludonarrative dissonance and structuring: it is best to structure their narratives in ways reminiscent of serialised dramas — short narrative arcs that are best suited for shorter play times, with overarching arcs that can engage players for a time much longer than the length of a feature film. In other cases, games venture too far from the realm of cinema to apply the same principles.
This dissertation began on a faulty premise: that games bear likeness to cinema, and that this likeness is sought after by developers. The research undertaken has, however, shown that the most cinematic of games are made only by studios with the highest funding; and that games as a medium have a bevy of challenges and considerations unique to them that are not relevant to cinema. It has been an interesting foray into storytelling nonetheless, and one that has shown the most creative, most novel ideas come from the developers experimenting in this vast frontier. While cinematic games hold stories well told — influenced by cinema — it is the ones that utilise the strength of the medium that make the standout experiences: playing Uncharted and watching Indiana Jones evoke similar feelings, but there is nothing quite like the guilt of sending a party of afflicted, crushed adventurers into the Darkest Dungeon.
Bibliography
Abernathy, T. & Rouse, R., 2014. Death to the Three Act Structure! Toward a Unique Structure for Game Narratives. [Online video]
Available at: https://www.youtube.com/watch?v=m6Hjfu0-oZY
[Accessed 4 April 2018].
Arazi, D., 2017. God of War and Cinematography in Games. Bydgoszcz, Camerimage.
Brown, M., 2015. Half-Life 2’s Invisible Tutorial | Game Maker’s Toolkit. [Online video]
Available at: https://www.youtube.com/watch?v=MMggqenxuZc&
[Accessed 2 April 2018].
Brown, M., 2016. Morality in the Mechanics | Game Maker’s Toolkit. [Online video]
Available at: https://www.youtube.com/watch?v=6RHH7M4siPM
[Accessed 28 March 2018].
Guinness World Records Limited, 2018. Record Application Search. [Online]
Available at: http://www.guinnessworldrecords.com/search/applicationrecordsearch?term=longest+cutscene&contentType=record
[Accessed 2 April 2018].
Hancock, H., 2002. Better Game Design Through Cutscenes. [Online]
Available at: https://www.gamasutra.com/view/feature/131410/better_game_design_through_.php
Henderson, D., n.d. Microsoft User Research Study. s.l., Studios User Research.
Hibino, N., 2004. Snake Eater. s.l.: Kojima Productions.
Hocking, C., 2007. Ludonarrative Dissonance in Bioshock. [Online]
Available at: http://clicknothing.typepad.com/click_nothing/2007/10/ludonarrative-d.html
[Accessed 6 April 2018].
Horneman, J., 2015. The Design in Narrative Design. [Online video]
Available at: https://www.youtube.com/watch?v=f8VIlfTtypg
[Accessed 4 April 2018].
Kojima, H., 2003. Hideo Kojima at the Movies: Hitchcock Films. Official Playstation 2 Magazine, 1 August.
Kojima, H., 2010. HIDEO_KOJIMA (@HIDEO_KOJIMA_EN) | Twitter. [Online]
Available at: https://twitter.com/HIDEO_KOJIMA_EN
[Accessed 2 April 2018].
Konami Digital Entertainment, 2017. Metal Gear in Concert. [Online]
Available at: http://metalgearinconcert.com/
[Accessed 2 April 2018].
Pinchbeck, D., 2016. The problem with ‘walking sims’ [Interview] (28 September 2016).
Sawyer, J., 2012. Choice Architecture, Player Expression, and Narrative Design in Fallout: New Vegas. [Online video]
Available at: https://www.youtube.com/watch?v=LR4OxNfzTvU
[Accessed 7 April 2018].
Schatz, E., 2014. From Uncharted to The Last of Us: An Hour with Emilia Schatz, Pt. I [Interview] (29 July 2014).
Stein, Z., 2015. This is how big the script was for The Witcher 3: Wild Hunt. [Online]
Available at: http://uk.ign.com/articles/2015/05/29/this-is-how-big-the-script-was-for-the-witcher-3-wild-hunt
[Accessed 4 April 2018].
