Interviews

Sound Analysis With EA Audio Director Aubrey Hodges
By Louis Bedigian

"I really believe that audio in general is the emotion in a game. You can tell that when you turn off the audio in a game – all of a sudden you don't have that same feeling."

No one plays a game solely because it’s pretty. We may start a game for that reason, but if everything else sucks, the odds of anyone seeing the credits roll are slim to none.

Audio is different. No, I’m not trying to imply that gamers are willing to suffer through garbage gameplay for a mesmerizing score, or endure a lackluster experience for realistic engine revs. But can triple-A sound improve the quality of a game that falls between “excellent” and “slightly above average”? You better believe it. I don’t need to give examples – just think about the games you love. The proof is in their sound.

One of the key people in forming a game’s sound is the audio director. Aubrey Hodges is one such person, working with musicians and sound designers at Electronic Arts to ensure their games are as compelling as possible. In GameZone’s two-part feature with Mr. Hodges, you’ll learn about the grueling process of audio development and uncover a few details regarding the new season of EA Sports.


Sound-wise, Madden 08 wasn’t a huge departure from the previous game.
Could bigger changes be in store for this year’s edition?

Start by telling us what an audio director does. I've interviewed musicians for years and you might very well be my first audio director.

Aubrey Hodges: [Laughs] Well of course I didn't start this way. For many, many years I've been writing game music and scores all the way back to the Sierra days with King's Quest and Leisure Suit Larry and all that stuff. But eventually what happens, when you get a bunch of folks together in a studio like this, is that somebody has to sort of help set the vision for the product. Not just the parts individually but collectively. What are our goals? How do we work?

Since I've been there and gone through most of the stuff myself, I understand the fastest way to get from point A to point B and how not to make common mistakes that get made all the time in games. It doesn't mean always that everyone's going to listen to everything I have to say and do it 100% my way. But it does mean I can certainly say look: this is the best way to work, this is the best way to accomplish our goals, and let's try it. A big part of my job is taking a look at all of our projects and setting a vision for what it is those projects are going to sound like every year. And then I have to take that vision that we get signed off on and I have to figure out a way that can actually be accomplished, via contractors in some cases or internal guys.

Once that happens I literally sit down and map out the development for each project in terms of what the tasks are, how long [each of] the tasks should [take] individually, and how much development support I think we're going to need from the programmers to get the stuff into the game.

Then I work with the programming team to determine, based off what we want, when our schedule should have it – what order would they prefer developing our audio features into the game. It might be an advantage to them to put the tackle sounds in before they put the breathing sounds in due to the way that they work. I work closely with the different development managers on each team to ensure that the schedules that we have are in the right order. That way I can go to the audio folks that are making the sound effects or designing the music and say, "Here's the order I need your tracks in." It's not just a free-for-all, "Here's a list of 50 things. Do it." No, it has to happen in a certain order because they're working on level 16, level 12, or they're working on this part of the game before they're working on the rest of the game so I need these elements to go with it.

Then, basically playing the games, listening to the games, and finding problems with the way that the sound has been implemented. Or casting my ballot for what I do or don't like about decisions made outside of my control. We sometimes don't cast, for example, the voice-over talent. And even though we're the sound guys and we get blamed and the sound has a low score because they don't like this voice-over guy or that, it's not us. It's other people within the organization because it's a typical big company and they have the decision-making power.

We have a vote. We can say, "No, no, no, I don't think this person would work. Can you please try this other person." But it doesn't mean it's gonna happen. But I at least have to put my official, you know, "This is what we think should happen here" in for every feature that's not under our control.

For the things that are under my control, I try to fix whatever can be fixed within the time limit given before we have to ship. We have very limited resources just like every company out there so you can only get to so many things.

I literally go to the alpha of a product and start running down a top 10 list of things that need to be fixed or we just can't ship. An over-my-dead-body type of list. As it gets closer to beta we start making new lists of things that are not the way we want them to be and try to fix those as well.

At the end of the project I'll take a hard look at what we've accomplished and what we didn't accomplish and write a post-mortem about where we succeeded, where our successes were, and where I thought we could've done better and where we could improve next year. Then we start the process all over again. So I'm dealing with that for 26 to 27 products a year.


Need For Speed Carbon

When you're doing an annual game like Madden, how does it differ from a brand-new EA project that comes to you, one that won't be updated every year?

AH: Let's deal with the iterative title first. The iterative title is in some ways easier in the respect of having a base to work from. You're not dealing with complete silence. However, it can be more challenging, technically speaking, in that when you go for new features it means you have to break the existing feature to make the new feature. You go through these really ugly periods where the game sounds like garbage 'cause you're having to break a sub-section to make another thing happen.

In the case of Madden, for example, we will do things that are based off of existing code. Well, if the existing code changes dramatically, i.e., they switch to a completely new physics engine, we're screwed because we're basing our footsteps and our breathing on the physics engine. So if they rip that out, all of a sudden what drove one portion of our game is gone and all of a sudden you hear crickets. Having to play the game and go, "Where'd the footsteps and breathing go?" And then someone says, "Oh, they ripped out the physics systems." [Now] we've gotta figure out how to tie it into something else, to tell it when the feet are moving or when the person is supposed to be breathing.

That's the un-fun part of an iterative development. The fun part is you have longer-term goals. I'll do three- and four-year planning of where I want a title to be in three or four years and then baby steps each year to get where we want to go. So we might one year include footsteps. And then we might include breathing for single-player. And then the next year we might include a feature where you can hear some talking from that player. And then the following year we might include the ability to pick which player you're going to listen to. For example, I want to hear the sound of the wide receiver or the running back.

Same with cars. Maybe this year we'll do a recording of just one type of NASCAR, but the next year we'll do a recording of a restrictor-plated NASCAR, because it's expensive to do these things and we have to leverage the expenditure over a few years or the game's not possible. Not possible games means we won't get to make them anymore. So we figure out the best way we can get a lot of bang for the buck over a period over a few years because they're technically very challenging to make, in addition with trying to come up with unique features. So we plan in broader strokes and accomplish a little bit more each year. Of course we try to re-freshen the music and any sound effects, we try to do that every year.

Any new features that are completely groundbreaking, where you're like – "Wow, where'd that come from?" – they take a lot of effort to make happen because you're dealing with an existing code base sometimes which already works and you have to ask yourself: "Wow, do I really wanna break the game to add something?" That's a bear.

Now when you're dealing with a new title, it's wonderful in the sense that the sky's the limit. You can come up with all kind of new things that you want to do. However, now you're dealing with the reality of having to do everything for this project, usually in a pretty short timeframe. EA is not known for dragging things out. They release games in a pretty timely fashion -- we don't spend five years making games. So we've got to figure out how to make something go from nothing to fantastic in usually 18 months, at best. It's even more daunting when you realize, for the most part, we go out and record most of our sounds. We use very few library sounds. That creates some challenges. We have small audio teams and work our tales off to get it done in the time we have.

In terms of the score, we really wanted, like in the case of Superman, to deliver something that had the Superman flavor that everybody's expecting, the sort of Hollywood approach to it. But we wanted to deliver a lot of music and I believer we achieved well over an hour and a half of music in that product. It's astounding – that's a lot of music.

We're trying the best we can to deliver the best audio experience, because I really believe that audio in general is the emotion in a game. You can tell that when you turn off the audio in a game – all of a sudden you don't have that same feeling. It's not scary, your heart doesn't get pumping, you don't feel like you're really there, like you're really driving. It starts to feel like it's strictly surveillance. It's not a game. That's how I feel about it.

I got into audio many years ago when really there wasn't much audio. Just little beeps out of a PC speaker. It's a lot different now, and we're finally starting to deliver that entertainment people have been going to movies to get. I think that's probably one of the reasons why the video game industry is surpassing the movie industry in terms of sales.


NFL Head Coach 09

 

Medal of Honor is the perfect example. If I turn the sound off, the experience is ruined. Both the music and the sound effects are integral to that series.

AH: They help you understand what's going on in the game and we try to do that even with Madden. When we, for example, set up a play, and you're getting ready to snap that ball, if you listen carefully you can hear the defense yelling out things just as if you were at the real game. Well those a real – we're actually cluing you in as to what that defense thinks might happen. So if the guy's going, "Run, run, run," well they think it's a run so you might want an audible. We're trying to help you with as much audio as we can.

A few years back in NCAA, we even had the crowd meter going so that if you pumped up the crowd with the button, you could make the other team false start. So we're really trying to make the crowd and the sounds of the game and the environment, all of these sounds, not just come alive in terms of their entertainment value – 'cause they are entertaining. But to help you play the game, to be an integral part of playing the game.

In NASCAR, if you're on a surround sound system, a car coming up on the left behind you. And you know if he's going to try and swing to his right 'cause you hear his engine sound as he's passing. You can make your adjustment without even looking at the rearview mirror. That's what it's all about for us. You know if he's going to start drafting too because we not only changed the sound of your engine, we also added all kinds of wind effects to let you know you're slipping into the draft.

We approach [this] two ways: we approach it artistically from the standpoint of what would make a wonderful and compelling presentation. But we also approach sound design and music creation technically. What would help you play the game? What would help integrate into the game and help really sell the suspension of disbelief and actually help you play the game better, and make it so that you enjoy playing the game and are not frustrated. My job is to really take a look at all the audio designs – lead audio guys, sound design guys and musicians – and see into directions that I think are going to be good for the project.

I'm sort of the emergency strike team as well. I will write music and design sound effects as needed on every game, all the time. [The developers] have families, they have lives, they need help sometimes. It's my job to keep an eye out and spot when I need to jump in, and then just jump in and make it happen.

I know it's kind of early in the year but can you give us an idea of what kind of sound changes we can expect from Madden, NHL, FIFA – all the different sports games?

AH: I have to be careful about this. Although I can... We are making a change in the Madden commentators. I can't mention who they are. We are using some of the biggest names in the business. We listened to the fans and we are making a change. And that will be more than one. It will be a fantastic year for Madden fans, I think they're gonna be thrilled with this new direction. I don't think I can say much more than that.

We are also making a change for the direction of Tiger Woods commentary. I'm sorry I can't provide more names right now. We will announce it... We have a lot of legal agreements with the PGA and Tiger Woods and Madden and the NFL so we have to be very careful about what we say and when we say it. But we're very excited about those two changes in particular. We're going to lend some very stellar sounds to the product this year.


New sounds for Tiger Woods 09? Sounds good to me. (Note: Screenshot from Tiger Woods 08)

Do you audition actors for voice-overs, or do you select actors based on demo tapes?

AH: We basically have quite a few casting agents and talent companies for voice-over actors. These are for the non-marquee talent; non-John Maddens. We have an idea of what we want each character to be.

In [NFL] Head Coach, for example, the new version coming out this year, we used over 15 voice actors. It's a monster project – 30,000 lines or more. It's a massive game. If you're into sports, fantasy sports and coaching are your bag, you're gonna love this title. We went out and found the very best [talent], they came out here and recorded, and it worked out really well for us. In some cases we do some remote recording as well, and in some cases people will come to us, agents for example, and say, "Hey, I've got this guy, he's really good." And we'll listen to the reel, and if they are good, we'll think of a way to get 'em in.

With the star talent, in some cases, we simply are obligated to use certain people based on the contracts of the company. For example, when we did Superman we used real actors from the Superman Returns film. In that case we wanted authenticity. But in a lot of cases that's cost-prohibitive, and we just use real solid voice actors. Mostly within the Orlando area, in our case, but also throughout LA and overseas.

You mentioned that there are times when certain sounds have to be created first because a specific level is also being made first. What is the purpose of that? Is it because that specific level may be more difficult to work on or more crucial to the game?

AH: Depending on how they started, it’s that they’ll start a certain core technology that maybe they were borrowing from something else.

Or maybe level 16’s stadium was the first one the built. Maybe the Seattle Seahawks stadium actually got built first, [and because of that], they want all the sounds associated with that. Or maybe the programmers, let’s say we’re talking about something more integral to the way the game’s mechanics work, maybe everyone sort of felt the way the play mechanics were running just really needed to be fixed because it was broken and effected a lot of other things. So push some feature that may have occurred later on to the beginning to fix that because there are so many dependencies that relate to that one thing. In other words, that one task might be breaking 10 others.

In the case of Superman and some of the other non-sports games, those factors play into it as well. Which [level] do they start with first, how many did they originally have versus how many they shipped with. In some cases, level 14, 15 and 16 didn’t happen and 17’s still there. So after 13 it goes to 17. [Laughs] But maybe then they added level 2 and 3 which were originally cut, so you never know how it’s going to go.

Stay with GameZone in the coming days for Part 2 of our interview with EA Audio Director Aubrey Hodges.