Thoughts on game voice casting: Part Two

This is the second of two articles in which OM founder Mark Estdale has shared detailed insight and information on casting actors for video games. You can read Part One here.

I spent the last month in Los Angeles, at the production coal-face; casting three games, baptising a new generation of head-mounted camera rigs (HMC) for our face-capture pipeline, overseeing four productions, and signing off on the completion of the OM voice- and face-capture stage in Santa Monica. It's an exciting time, but what sticks with me most is the massive importance of casting. Not just from the developer's perspective but from the actor's too. This article is aimed at both.

Sony Interactive Entertainment's Daniel Birczynski gave a talk at Digital Dragons 2018 on the "the sound of God of War." Casting God of War, he said, involved over 2,600 person hours. This isn't to say that every game needs to spend that kind of time casting, but it does point to its value.

As I stated in Part One, the objective of casting is the timely recruitment of the right team. What defines "the right team" is of course governed by the production methods you adopt to achieve your objective. I'm going to ignore casting for PCAP [performance capture] and MOCAP [motion capture] here, as they are fundamentally classic stage/screen forms and there's a wealth of great guidance to be found elsewhere. My focus here is casting actors for voice- and face-capture in a recording studio. This is the roadmap for casting that we use at OM and the logic behind it.

"If any actor bleats about the way you work or tries to educate you about their process and unmet needs, my advice is run like the wind"

Simplified, the objective in casting is finding talent with the ability to be true to both character and the moment the character is in. The key to acting for games is based on character first rather than script first. Unlike stage and screen, with line learning and small linear scripts from which characters are defined, the dynamic complexity of most game scripts require them to be sight read, with the characters fleshed out in advance by the developer in character briefs.

I frequently hear developers apologise about their scripts and processes -- my advice is don't. Games are a radical performance form that differs from stage and screen. Don't be held back. In casting, your job is to hire actors who embrace and are inspired by what you are doing and the way you work. End of story.

The essential skill that matches your need is flawless sight-read performance -- a skill that is all about instinct. And as life is lived instinctively, and as acting is the art of emulating life, then I suggest that sight-reading for games is the height of the actor's art. Embrace it. It shortcuts the "actor's process" but not the actor's essential skill.

Some developers do try to kowtow to stage and screen methods in a pursuit of quality, but on the whole I think it is an overly costly pursuit that yields debatable results, apart from ego -massaging press opportunities. Instinctive performance can match and surpass rehearsed forms.

If any actor bleats about the way you work or tries to educate you about their process and unmet needs, my advice is run like the wind. They are wrong for you. Don't hire them unless you have the time, money and skill to either re-educate them or do the heartbreaking work of hammering the square peg of their "needs" through the round hole of yours.

The good news is that most good actors do actually have the skills you need. Learning to trust instinct is a foundation of their training. Yet for many actors, once the freedom it brings is found it becomes dormant and unused in practice, as the careful "script-out" crafting, rehearsing for stage and screen may only touch on instinct at the very preliminary stages of production.

When I cast, my main focus is on finding the instinct skillset. The point is that it is vital to know exactly what you are looking for. It doesn't matter what values you have or what methods you use to achieve your ends. The essential is clarity about what you're looking for and why. Clarity is everything. It goes back to good recruiting practice -- see Part One.

What I look for in casting is the character fully alive and behind the character, an actor who embraces teamwork, who is intelligent, self-aware, flexible, at ease in the production environment, playful and, of course, instinctive when sight-reading.

The value of thorough casting cannot be underestimated. The results will be seen and heard.

Being clear about what you want and why is where you start

Time spent on preparing clear casting and character briefs is time well spent.

The casting brief

A good casting brief lays out clearly what will be required from the actor: time commitment, expected schedule, fees and payment terms, recording location(s), details on vocal stress, MOCAP, PCAP, physical requirements, likeness usage.

"If you want something unique, look for the unique. Cast widely and don't go to straight to the usual suspects"

It will include contract terms and the non-disclosure agreement (NDA), plus details about the title, genre and age-rating, and information on controversial, sexual, violent, religious, political or other potentially sensitive topics. And of course it includes details about the casting process, how to submit, what to submit ,and what not to submit.

The more precise you are at the outset, the less email back-and-forth you'll get as a result. Also, precision enables you to clearly evaluate if the applicant or agent is paying attention.

A note to actors on NDAs: these are legally binding commercial agreements. If you break them you can find yourself being fired and sued, and it could mean the end of your acting career. News travels fast. Stick to the conditions without exception and ignore what you see on the internet. Something on Twitter or the Internet Movie Database does not override your obligation to keep to the terms, and it won't stand up in court. If you need written permission to share your involvement, get it.

The character brief

An ideal character brief is concise and includes a character image. Supplying a video with the character's animated range of motion is the ultimate visual reference. If those are lacking, or if there are security concerns, I use a character image or a selection of images from anything that embodies the character.

Follow with a couple of short paragraphs about the character, their age, accent, vocal qualities, their world and their relationships to others, to who they are, what they want, what stands in their way, and how they get what they want. The audition lines are below that. I like to keep it to between 150 and 300 words. If there's nothing in the audition lines that really brings out the character, then I write the character information as short monologue to be performed.

When sending out a brief I will include a tabled summary of all the characters, so actors and agents can, at a glance, see what I'm looking for. I like adding some kind of psychological analysis, like the Jungian Myers Briggs Typology.

The analysis is also very powerful for assessing if the developer has a united view of the character. When decision makers are in conflict, I encourage them to do the test for the characters they aren't united on. The process is fun and the results give clarity.

Our casting process

If you want something unique, look for the unique. Cast widely and don't go to straight to the usual suspects. I call it "embracing the great unwashed" -- it is time consuming, it can be dirty, but it's panning for gold and you will find glorious nuggets in the most unusual places.

The evaluation process

We would never cast from an actor's demo reel alone, even if we know the actor by reputation. Their voice and experience are just a starting point. We want to test them under production conditions. It's the basis of good recruitment practice.

Over and above a performance bringing the character to life, it's helpful to know the ability of the actor to take instruction, to test their confidence, commitment, empathy, playfulness and level of comfort, their speed, resourcefulness, nerve, attention to detail, observation skills, and their momentum both with the character and as a potential team member. Their talent is irrelevant if they don't have a personal positive impact.

"A mind-blowing statistic is that 80% of applicants fail to follow the brief on a wide casting, and it is even higher in Los Angeles"

At every stage I'm looking for the signs. Do they give or do they take? Are they generous and inspiring or do they suck out your soul? Do they energise others or demoralise? If that momentum is negative, their talent, however great, will follow suit. Do you want to be dealing with an egotistical narcissist or a humble, passionate and skilled supporter?

The sightly devious

I want to find intelligent, engaged and committed actors (and agents, as good ones make the casting process easier and a lot less costly). In my briefs, I will occasionally leave out critical information. Those who spot the omission and ask for the missing information get noted and a big team YES for attention to detail.

Selection

Throughout the process we evaluate submissions and use a scoring scale, from A to E -- A = Awesome, E = Exclude. If the brief isn't followed it is an automatic E.

The filtering, some classic examples:

• Accent. If specified as native -- "can do..." = E
• Vocal quality. If singing voice is specified as Soprano -- if not = E
• No headshot requested -- a headshot = E
• No demo reel submission -- a reel = E
• Name, email and contact number in body of email response -- if absent = E

A mind-blowing statistic is that 80% of applicants fail to follow the brief on a wide casting, and it is even higher in Los Angeles. It doesn't matter how talented or famous the applicant is, if they don't follow the brief they are displaying a trait that is not wanted in production. It's brutal, but for the team it makes life easier as 1,000 applicants will be culled to 200 by the applicants themselves at the first hurdle.

Self-submissions

We ask for self-recorded reads to be submitted as mono mp3s. We don't mind the audio quality as long as the performance can be heard clearly. Self-submissions are evaluated blind, so any reference to the actor's identity is removed.

We use a matrix to help with evaluating self-submissions, culling down for live auditions. For the matrix we score A to D, focusing on character fit -- A = nails it, B = good, C = weak, D = bad. For the auditions we call As and Bs only. However, we prioritise As and Bs based on the actor's overall score, so if they submit for multiple roles their average score will determine their call likelihood. We do this as it highlights actors who are self and production aware.

Also, the matrix clearly highlights what an agent is doing, as the average score of all the actors from a given agent gives a score. You can see me talking more about the matrix here.

Live auditions

Below is a visual representation of how we think and score during live auditions. We add an E to the scoring. We regard Ds and Es respectively as pulling teeth and flogging a dead horse. D is only ever worth it if the actor has something special and unique, because the work needed to get them into the right space will cost time, money and emotional effort.

What differentiates As and Bs is flexibility. An A can go anywhere with you; they can play multiple roles, even against themselves without being spotted and switch accents with ease. B is more limited; they may be perfect for just one specific role or accent.

The audition is under production conditions and recorded and possibly filmed, but the evaluation starts as soon as the actor arrives. If they are late they will be scored E, B if they are on time, and A if they are early.

We'll note how actors interact with casting team members and other actors when they arrive for the audition. We may purposely have actors coming for the same role in the waiting area together so we can observe their interactions. We are simply gathering as much information as we can to assess how the actors will fit with the team. It can be very revealing.

Once in the studio for the audition, the actor may be asked to perform the lines from the brief. They may be asked to perform sides given to them as they arrive at the audition, and they will also be given cold lines in the studio. After a short introduction they'll read whatever they have without direction.

My best note to actors is to make firm decisions about the material you have been given. There will be contextual holes. Fill them yourself from what information you have and not by trying to guess what we're looking for. If you're selected to audition we're looking for what YOU bring. If you don't have a clear idea about who your character is, where they are, etc., your performance will be empty.

"My best note to actors is to make firm decisions about the material you have been given. There will be contextual holes"

After the first read we will ask for adjustments. They are to see how well the actor listens and responds. If the actor has the material for some time, they may not adjust well. Adjustment gets better with material received when arriving at the audition, and is best with material read cold -- and with good actors.

Actors familiar with the process are generally relaxed in auditions. Their psychology leans towards enjoying the experience, experimenting and perhaps learning something new. Those unfamiliar tend towards desperation and try hard to impress. That is understandable, but it does get in the way of the actor giving their best. We understand this so we'll normally run through three or four times. We want the actor relaxed, safe and playful. We want their best and we'll try and wheedle it out of them.

One thing I do when directing auditions is focus solely on the character I'm casting and not on the actor. I'm watching for the character coming to life. Mentally, I blank out the actor in the studio. I may seem cold and dismissive -- it's not personal, I'm just focused. Actors who try to engage or blow smoke up the casting director's ass when they come in are noted for lack of awareness. Sensitivity takes the director's lead and engages on invitation only.

We score during the audition, but once the auditions are complete we blind listen through the recordings, discuss and re-score. Then we compile a final score sheet based on the arrival assessment, session score and listen through score.

We then submit coded audio samples to the developer, so they can select their favourites. We submit our ABCs, and the occasional D, without our notes or scores. At this stage we don't want the developer influenced by our thoughts and who the actors are. We want gut reactions. Once we get feedback from the developer we share the names and give our input. From the joint results and final discussions the cast will be confirmed.

Of course, some developers do have their own preferred processes and want things done their way, and we comply. We'd be daft not to. This is the nuts and bolts of what we like doing.