Part 1 can be read here.
6. The Curse of Voice Acting
I have nothing but admiration for voice actors. The truly talented ones, the ones who aren’t just reading off the sheet, are character actors, able to find just the right spot within their range and create an entire personality around it. There is so much depth and emotion that comes from a good performance by a voice actor that even the most stock dialogue is given personality. It allows for greater immersion in to the game’s universe, and the writer can convey things in a way that pure text can’t. Listing the voice actors whose work I so admire in gaming would be doing a disservice to those I would invariably forget, whose work I do not know, or simply do not have the space to discuss.
So when I say that I don’t want voice acting in an Obsidian-based Fallout game, it’s not because I think there’s a lack of talent out there, or because I don’t enjoy voice acting in gaming. It’s because voice acting has made games much, much more difficult to develop.
RPGs are, by their nature, dependent on dialogue in a way that most games are not. With a larger cast of characters and heavy narrative, it’s only natural: it becomes all the more complex when choice is introduced in to the game, as is common in WRPGs. Lines have to be crafted depending on your response, lines have to be crafted for other characters depending on your response, entire story paths are laid out due to a push of a button.
Unfortunately, these exact points in RPGs have frequently forced writers work around voice acting, rather than use it solely as a tool to better tell the story.
For example, Fallout: New Vegas’s DLC. There were four story packs, Dead Money, Honest Hearts, Old World Blues, and Lonesome Road. Combined, they were budgeted for 10,000 lines of dialogue. That’s 2,500 lines per 4-10 hour experience, compared to the 3,000 lines of dialogue you are likely to hear in an average motion picture, and the 63,000 lines in the main game.
To better understand what that limitation means, let’s look at what you can reasonably extrapolate to be each individual DLC pack’s design objectives:
A: Tell a complete story without requiring you purchase other DLC
B: Interweave with the other DLC
C: Relate to the main campaign so as to not feel too separated from the main game
D: Be playable in any order
E: Be 4-10 hours long
And of course, you won’t hear every line in a single playthrough – like the rest of New Vegas, the choices you make in the story impact NPC reactions and their dialogue.
So, obviously it means these packs have to, in some ways, focus less on character interaction than the main game. It certainly helps that The Courier isn’t given a voice actor, but how else do they accomplish this?
Well, every DLC pack starts by separating you from your companions. Then, you’re removed from the main setting of the game. You’re put in to a new setting which is contained (Old World Blues, Dead Money), separated from much of civilization (Honest Hearts), or simply desolate (Lonesome Road). All three of options lead to a minimalist cast.
The effects of the DLC affecting the main story of the game would require recording more dialogue, so the stories are essentially self-contained. Old World Blues, the most blatant example, outright brainwashes your character so you can never discuss the events of The Big MT outside of the DLC.
A greater emphasis is put on atmosphere and visual storytelling. Dead Money with its terrifying, cramped setting and survival horror elements. Honest Hearts with its vast wilderness. Old World Blues and its strange, broken high tech environment. Lonesome Road with its horrific landscapes and monster design.
And a number of other, very clever devices were used to help tell the story within that budget constraints – such as mute characters, characters who were discussed in the main game but only introduced in the DLC, journal entries from dead characters – to tell a story in a setting that made sense.
Other tricks in dialogue have to be used as well, not just in the New Vegas DLC, but through most RPGs: While your (often mute) player character can pick multiple dialogue options, they might end up getting the exact same response, or the writing is subject to dialogue-splicing. Here’s an example of the latter:
PC: So where can I get some grub in this armpit of a town?
NPC: Elda’s Diner, just down the street.
PC: Any recommendations on where I can get something to eat?
NPC: The same place I eat every night: Elda’s Diner, just down the street.
PC: I’m looking for Tommy Two-Tone, you know where he hangs out?
NPC: That rat? Yeah, I know where he is. Heh, the same place I eat every night: Elda’s Diner.
PC: Where’s Elda’s Diner?
NPC: Just down the street.
This is just sample text I wrote to illustrate the point. You can see the various tools being used in dialogue trees to create fully fleshed out dialogue on a budget.
While dialogue-splicing this does create more of a variety in how you’re able to interact with characters, relative to the amount of dialogue that’s actually recorded, the responses can often feel bland and lacking in characterization due to the (understandably) utilitarian nature of the writing. Another issue in this process is that voice acting requires that the dialogue be finalized earlier in the production process in order to allow for beta testing.
Anyone who isn’t Hunter S. Thompson will tell you that writing is rewriting, so this loss of opportunity for iteration creates a lack of polish to the dialogue that might otherwise be allowed.
And most damningly, reactivity to your actions is drastically reduced. They can’t afford to include as much dialogue in the game, so they don’t write it. The choices you have in conversational dialogue become two, perhaps three, and rarely more than that, because the responses must be limited to reduce studio time. On top of that, there are less ‘triggers’ (consequences occurring in later events due to a specific action within the dialogue) in order to reduce the amount of lines that must be recorded later.
The writing becomes less nuanced as a result, and the game, no matter the talent behind it, loses the opportunity for some creative choices. These issues are not unique to the Fallout: New Vegas DLC, though the implementation of their workarounds might be.
And from the perspective of voice actors, they’re forced to read for a performance that will be used to fit the needs of any number of splices that occur within in a dialogue sequence, which does reduce studio time… and the quality of their performance. As much as I love many different parts of Alpha Protocol, including its excellent audio engineering and performances, once I caught on to the use of dialogue-splicing I could not un-see it.
Now, after all that, I want to clarify my position. Voice acting is not bad, it is restrictive.
Voice acting also can add emotional depth to a scene, integrate gameplay and story, and bring a new dimension to a character. In particular, voice acting lends itself well to humor and the nuances of speech: the subtle anger when speaking a name, a breathless joy upon being reunited with someone thought-long lost, and so much more. And when I think of my favorite vocal performances in gaming, they tend to be from RPGs.
And like I said, not having voice acting is a restriction too: Portal could not have been the same game without a voice actor for GLaDOS.
So why would I suggest a six million dollar budget for Obsidian’s Fallout when that’s not enough to support full voice acting? There are two reasons: The first being that full voice acting isn’t the only option for voice acting at all. Fallout and Fallout 2 used a method where major and memorable characters had voice acting, though not necessarily for every line of dialogue. Minor characters, or less important dialogue sequences, didn’t use voice acting at all. It was used at critical junctures to help emphasize elements within the story and setting, without eating up the entirety of the game’s budget. It was used as a tool, and to great effect.
But the second reason is this. Good voice acting does, as I’ve said, add a lot to the dialogue. But for that potential upside – there is, after all, bad voice acting – it incurs not only a large cost, but the need to sacrifice certain creative choices. Those choices, like the ability to write complex dialogue, reactive dialogue, and reactive quests, are Obsidian’s greatest strengths. Why, of all things, would you ask them to sacrifice that?
Part 3 will be posted tomorrow, discussing some of the Obsidian Entertainment staff that deserve to be involved in a new Fallout.