|Jesse Harlin's Blog|
An archive of my previous articles for Game Developer Magazine\'s monthly \"Aural Fixation\" column
Hey, GANG -
I went to log in this morning and I found myself staring at the most-listened-to MP3 list on the front page side bar.
I feel like it's been the same list of songs forever. In order to get some fresh blood on there, I just deleted "The Force Unleashed" and "Emo Skater Girl" from my profile's audio player. In fact, I deleted everything from my audio playlist and tossed a couple of new tracks on there.
I'd recommend that some other folks do the same thing just so that we can see some of the content from other talented users.
Perhaps what we need on the front page is more of a Randomly Featured Artists list than a "Most Plays" list. That way we're currently seeing material from new people and it's a good way to spread some exposure around.
Just an idea. What do you think?
(From the February, 2007 issue of Game Developer Magazine)
Few phrases carry more weight in the industry now than the magic words “Next-Gen.” From the Wii to Gears of War’s cover mechanics, developers and consumers alike are trying to figure out what that phrase means. However, before audio professionals can delve into the finer points of surround mixing or user-driven content, there are still unresolved issues from the current generation that need to be addressed.
This list represents features that have all too frequently been brushed under the Current-Gen audio rug. Their industry-wide standardization should represent the first steps towards the definition of Next-Gen Audio.
Disc space is no longer an issue. The only things standing between diverse audio and two-footstep characters are old habits and narrow aesthetics. As games got bigger, Current-Gen audio strove for coverage. The results were often little more than a sketch. Next-Gen audio should now strive to fill in the canvas. The solution is simple: if the player will hear the sound frequently, there should be five to ten randomized versions so as prevent fatigue. Hit impacts, foley, weapons, environmental sounds – real worlds are rich with variation. Ours should be, too.
It’s happened to every gamer. Their character enters a new area. There’s no map. Their radar malfunctions. The power has been cut and surprises lurk in the shadows. Regardless of the visual mood, a lack of occlusion means that they can hear the signature vocalizations and foley sounds for each enemy around every corner and know exactly what to expect next. At that moment, the lack of occlusion has completely destroyed all drama, tension, and anticipation that the level designers and environmental artists have painstakingly worked to create. To lack occlusion is to undercut the efforts of the other game designers.
There’s nothing immersive about a loading bar. While dynamic loading is a battle that must be planned for and waged across all disciplines, it’s the responsibility of the Audio Lead to initially request dynamic loading of audio assets. With the launch titles launched on the new consoles, it’s clear that “Next-Gen gaming” means little more right now than prettier graphics. Eventually this will expand to all disciplines as greater immersion is inevitable throughout the next console cycle. The ability to load and unload sounds only when needed will be vital for audio. Smarter memory management should be one of the primary goals of all audio tech moving forward.
Much like occlusion, robust DSP systems speak directly to the heart of immersion. DSP effects like filtering and reverb place our sounds in the spaces the characters inhabit. Unfortunately, with surprising frequency, Real-Time DSP is a casualty of poor planning, poor tech, or both. Foley in outside spaces should sound different than the same foley files played inside interior spaces. If characters are grunting and groaning inside of a cave, their utterances should be processed with a cavernous reverb (I’m looking at you, Final Fantasy XII).
Abandoning Chipset MIDI
You only have to look as far as GameSpot’s review of The Legend of Zelda: Twilight Princess to see that chipset MIDI scores aren’t going to pass muster with the gaming press in the Next-Gen age – and rightfully so. Enough game scores over the last console cycle ditched the old tech and embraced digital recordings that gamers’ tastes have largely evolved. Consumers have come to expect their epic games to have epic orchestras. It might be excusable were every game using chipset MIDI doing so because they were implementing an extremely interactive music system that digital recordings can’t achieve. They’re not, though, and chipset MIDI should go the way of 2D sprites and text MUDs.
Interactive music made strides over the last round of consoles. Unfortunately, thousands of new games come out each year, only a fraction of which have interactive scores. The majority of games still have “trigger and forget” loops tied to in-game locations. This fatiguing approach leads many players to turn music off in their Options menus. What’s worse, it’s led to Microsoft’s insistence that music in every Xbox 360 game be replaceable with the user’s own music files. This should be seen as a great failing on our parts as game composers. We can make music systems that don’t fatigue users. The goal of all game sound should be to make it as integral to the gameplay experience as maps, NPC objectives, and weapon upgrades – music included.
Story, characterization, and gameplay are rapidly distancing themselves from the text-scrawl days of games’ 1980s Neanderthal forefathers. Like chipset MIDI, voiceless games are already getting dinged in their reviews for their lack of dialogue. Again, disc space limitations are crumbling and localizing dialogue for a global market isn’t only doable, it’s proven to have tremendous benefits in opening up new markets to titles. Words like “toy” and “diversion” used to describe our games in the mass media. Soon, we will be the mass media and already terms like “art” and “cinematic experience” are replacing the labels of the past. Voice is crucial to this transition.
Obviously, not all of these issues are going to be applicable to all titles, all platforms, and all audio teams. Madden doesn’t need interactive music any more than Dragon Quest IX could possibly be fully voiced on the DS. None the less, this list will hopefully serve as a jumping-off point for debate and innovation.
(From the January, 2007 issue of Game Developer Magazine)
The music revolution has arrived and in its wake lay the bankrupt husks of once-mighty megaliths like Tower and Sam Goody. While Napster struck the first blow, the undisputed successor to the throne of legal music downloading is Apple’s iTunes Music Store. By offering decent quality downloads at reasonable prices, iTunes is largely driving the changes reshaping media distribution channels in the 21st century.
While seemingly every film gets a soundtrack release these days, game soundtracks are still extreme rarities outside of Japan. This is largely attributed to the limited return game soundtracks bring in when compared to the costs involved in production, distribution, and marketing of the discs themselves. However, an online distribution model changes this dramatically and already music by Chance Thomas, Jack Wall, and Marty O’Donnell sits under the iTunes “Soundtrack” category.
Convincing your publisher that a game soundtrack is a good idea is a topic all its own. This month, though, we’re going to look at what it takes to get your music listed on iTunes. There are essentially two methods: directly through Apple or through a digital distributor.
THE APPLE WAY
iTunes is a great opportunity for independent artists and smaller labels to easily distribute their music world wide. However, finding their online application can be difficult as all of the necessary links are in out of the way places on Apple.com. Start at www.apple.com/itunes and click on the light gray hypertext link at the bottom of the screen that reads “Working With iTunes: Labels & Artists.” On the following “Labels & Artists” page, the application to apply is linked to in the first blue box on the right side of the screen, conveniently labeled “iTunes Music Store Online Application.”
After complying with a brief “don’t call us, we’ll call you” and “you better own the rights to this stuff” warning from Apple, you’re ready to apply. The application comes in two relatively short parts. The first section is all about contact info and where globally you have the rights to distribute the music. The second part of the application is about the content – is it music, how many albums, how many tracks, and any other info you’d like to let Apple know. Pretty simple stuff.
Here’s the snag with the Apple method. As Apple is fond of telling you, real people read and evaluate every application that comes through their site. As such, the word on the street is that Apple can take a very long time to get back to artists, if ever. Since Apple doesn’t approve every artist who applies and you already clicked on the “don’t call us” button, the whole process can feel a bit like throwing a feather down a deep well and waiting to hear some sort of splash.
If you do hear back from Apple, you’ll need to download their free iTunes Producer software to edit track metadata, attach album artwork, and convert tracks into the necessary AAC file format. Monetarily, there is no fee to apply, but Apple will take 35% of all sales. The other 65% goes to the record label, which in this case might be the game’s publisher, the artist, or an actual record label. How much you as the artist will see of that remaining 65% is completely dependent upon the specifics of your individual deal.
You will also need to set up your music with a UPC code for each album and ISRC codes for each track. The UPC barcode is used to track individual albums sold and the ISRC codes function as a digital fingerprint for every song in distribution. To create your own barcodes, you need to become a member of GS1 US and obtain a unique company prefix. This can cost as much as $750. If you’re going through your publisher or an established record label, you may be able to get a UPC code through them. ISRC codes can be obtained for free from the Recording Industry Association of America (www.riaa.com).
THE OTHER WAY
The other option is to contract with a digital distributor. A digital distributor works just like their real-world counterparts. You give them music and they work on getting it distributed to vendors. In this case, that means online music retail services such as iTunes, Napster, Rhapsody, etc. Digital distributors do the legwork for their artists and then take a small percentage of the return from sales as commission. Distributor legwork includes digitizing the music and converting it into all required file formats for the various retail outlets. Turn around times range from about 2 weeks to 4 months, depending on the retailer.
The leading digital distributor right now is CD Baby, though others such as IRIS, indie911 and the IODA also exist. Artists are charged an initial “set-up” fee of $35 and $20 per UPC code. For their part, though, CD Baby then handles all necessary file format conversions, acquisition of ISRC numbers, and then the actual distribution. Payment is then made a week after they receive the money from the retailers.
While primarily established to aid independent and unsigned bands, digital distributors can remove many of the hassles associated with the Apple method. Additionally, your music will find its way to more than just iTunes. The downside however is a smaller monetary return and the insertion of a middle-man between you and your music on iTunes.
(From the November, 2006 issue of Game Developer Magazine)
Games have been chasing after film since the 1990s to compete in terms of production values, budgets, and audience. For game composers this has meant shaking off the shackles of chip-set MIDI and stepping onto the recording stage. However, the arrival of live recording to game music introduced a question that film composers have been wrestling with for years: is it a union-date or not?
The answer has almost exclusively been "non-union" due to budgetary restraints and a union interactive media contract that doesn't keep pace with the evolving technology of our industry. This has meant a choice between hiring non-union musicians, paying union musicians under-the-table, or turning to alternative musical enclaves for larger orchestral sessions. By far the most successful of these alternatives has been Seattle, Washington.
Most professional musicians in the United States belong to the American Federation of Musicians (AFM) and it's the AFM’s contracts that have dictated the terms for film and television music usage for decades. Seattle musicians, however, are part of the separate International Guild of Symphony, Opera, & Ballet Musicians and work from a different standard of contracts. Their no-fuss "buy-out" payment policies have made Seattle the new go-to location for a vast portion of scoring sessions. In addition to Seattle, recent improvements in telephony and data transfer technologies have made outsourcing orchestral recording to European non-union orchestras a viable option for budget-conscious composers.
ENOUGH IS ENOUGH
For years, AFM has watched unhappily as more and more professional recording dates slip from the union into the non-union column. Despite their displeasure, there was little movement from AFM in terms of a resolution. All of that changed this past July.
After watching a century's worth of protections for professional musicians slowly erode, AFM president Thomas F. Lee issued a press release that officially took a bold stand on the issue of Seattle. According to Mr. Lee's statement, as of October 1, 2006 The Union now holds all members involved in non-Union Seattle recording sessions to be in violation of AFM bylaws. Article 15, Section 3 of the AFM bylaws states:
“No AFM member may perform services (whether as composer, arranger, copyist, proofreader, instrumentalist, leader, contractor, cutter, editor, or in any other capacity): (1) where the product of the services is intended to result in, or be embodied in, recorded music made outside of the United Sates and Canada and the possessions of either; or (2) for the purpose of producing, editing, or dubbing recorded music except where expressly authorized and covered by a contract with the AFM or when expressly authorized by the AFM.”
Additionally, Mr. Lee clarified that - as stated in the AFM bylaws - any member found to violate Section 3 "shall be subject to a fine not exceeding $50,000 and/or expulsion."
These rules aren’t new and have been part of the union bylaws for some time. What is new is the aggressive nature AFM is now taking with enforcing their existing policies. As for the means of enforcement, the AFM is counting on its smaller local chapters to report members that are in violation and file official charges against them.
WHAT IT ALL MEANS
While the new hard-line stance is clearly aimed at stemming the tide of film work headed towards Seattle, there is nothing in Mr. Lee's statement that exempts game or television work. As such, any members of the union who finds themselves involved in game production should be aware of the potential for fines. Although the AFM doesn’t cover the craft of composition, many composers join in ancillary roles such as orchestrators, conductors, or instrumentalists in order to take advantage of union benefits. In addition, AFM members such as orchestrators or copyists often have little - if any - say over whether or not their work goes to Seattle. It remains to be seen if there will be any leniency when it comes to the filing of official charges. As it stands, there are no distinctions made in the AFM's policy based upon role or level of involvement and any union member involved in any capacity with a game project headed for Seattle should be cautious.
For composers looking to avoid the issue all together, it should be no surprise that in addition to non-AFM instrumentalists, Seattle also contains non-AFM conductors, copyists, and music preparation houses. As would be expected, the quality of their work is top-notch and pricing comparable to that of AFM contractors. Furthermore, orchestras in the Eastern European cities of Prague and Bratislava are now servicing media projects. These non-union options cost roughly fifty percent less than recording in The States. However, the level of musicianship does not match that of their domestic counterparts. While the quality of these recording sessions has been improving over time, composers going this route should be aware that their strengths still lie with music in a traditional classical vein.
It remains to be seen what the AFM’s new policy of directly targeting its members will do to the issue of orchestral outsourcing. With enforcement dependent upon local chapters and low-cost options a tempting draw, many union members may adopt a “wait and see” policy as the story evolves. What is clear is that the AFM is tired of sitting idly by while globalization spreads to the scoring stage.
The full text of AFM’s July press release can be found at http://www.afm.org/public/press/seattle.pdf.
(From the October, 2006 issue of Game Developer Magazine)
The Video Game industry is nomadic by nature and no field more transient than Audio. As a result, every audio professional is forced throughout their career to prove and reprove their worth solely on the strength of their résumé and the first 30 seconds of their demo.
Freelance audio folks rarely get to see the demos of their competition and Audio Directors don’t have time to give critiques on the demos that end up in their “No” pile. So how do you know if you’re a “No” or not? More importantly, how can you ensure you’re in the “Yes” pile?
PRESENTATION IS EVERYTHING
No demo exists in a vacuum. To the contrary, every demo will almost always find itself sitting in the middle of a stack of demos. Before any tracks can be heard, the first impression your demo will always make is a visual one. As such, presentation is everything. Imagine an Audio Director staring down a stack of CDs without any idea of what the audio on them sounds like. At that moment, the selection of one demo over another is almost completely arbitrary. The only advantage your demo has is its ability to stand out from the crowd.
This means that CD- and DVD-Rs that have been written on with a Sharpie marker are automatically at a disadvantage. At their best, they look sloppy. Far too frequently, however, they contain illegible or incomplete text. No matter how attractive you may think your handwriting is, it can’t compare to a printed CD label. But what do you print on the label?
First and foremost, always include your name and contact information on every part of your demo – whether it’s the case, the disc, the résumé, or a business card you tossed in for good measure. Discs and cases are easily separated and if there isn’t any clue as to whose audio the disc contains, that nameless disc just lost itself a gig.
Secondly, if you’re looking to stand out, avoid the standard cliché images common to the industry. For composers, this means never cover your demo with pictures of sheet music, treble clefs, noteheads, violins, etc.
So many composers use these images as the easy way of saying “This is a music disc.” The problem is that hundreds of composers are all using the same imagery. For sound designers, stay clear of pictures of waveforms, speakers, or screenshots from Protools as they’re all overdone. Both disciplines are also guilty of another common cliché. If you include a photo of yourself, don’t take it while sitting in front of your gear. Take a picture of yourself anywhere else. You’re looking to stand out and be different. Trust me. Audio Directors will still believe you know a mod wheel from a pan pot if they don’t see you beside one.
If you don’t have the graphic design or Photoshop skills to design something other than the clichés for yourself, hire a graphic designer to do it for you. It’s that important. You’re creating a brand of yourself that you’re then marketing throughout the industry. You owe it to yourself to do everything you can to ensure that your disc stands out from the pile of unknowns.
(From the September, 2006 issue of Game Developer Magazine)
Over the life of the medium, games have been defined in large part by their role as marketing-driven assistants to the grand Hollywood machine. Additionally, in no other form of entertainment does the concept of sequels flourish quite as much as it does in the world of games. Endemic to both cases is one central truth: games frequently exist solely to give the player more of what they already love.
As such, everyone involved in the work of expanding an existing franchise plays a delicate balancing act between faithfully rendering what the players expect while infusing their own creative mark on the title. A team of artists and designers tackles the look of the franchise while an army of production staff tackles level design, gameplay, script writing, and voice casting. Only when it comes to music is a large chunk of responsibility for the game’s faithfulness to player expectations frequently placed into the hands of a single person: the game’s composer.
TO TEMP OR NOT TO TEMP
With so much riding on the composer’s shoulders, the first step in any franchise game is to understand the intent of the project. A franchise game will only ever do one of two things: either stay true to the original material that came before it or veer off into a new direction in an attempt to reshape an existing intellectual property (IP).
Any composer working on a franchise that is faithful to its predecessors must begin by doing their homework. Consider all of the music that came before you to be the temp track for your game and the overriding creative direction to which you’re being asked to adhere. Watch the film(s) or play the original game(s). Familiarize yourself with the musical language of the original source score. Make sure that you’re familiar with any and all thematic material that may be applicable to the game including character, location, or key event themes. Take note of the original instrumentation or any signature production techniques that help to define the original’s identity. Firmly ground yourself in the music of the world into which you’ve stepped. Your music is a linear continuation of work already familiar to an established audience and needs to fit as a natural extension of the existing musical language.
However, a new Star Wars game does not necessarily mean the exclusive use of John Williams’ music any more than a new Spyro game means music akin to Stewart Copeland’s work on the original Spyro The Dragon. Games aren’t movies and the characters of a film can sometimes find themselves engaged in everything from dodgeball to vehicular combat to button-mashing fight combos once they make their move to the interactive screen. Because games are a great way to expand the universe of an existing IP into different genres, composers will frequently find that they’re being asked to reinvent the musical language of the franchise rather than faithfully recreate it.
When this is the case, the composer must first distill the essence of the original music. The challenge is to depart from the existing musical language without stepping so far away that the soundtrack makes the game feel as though it’s not part of the same universe anymore. Again, established thematic motifs, signature instruments, or stylistic genres will provide the framework within which your reinvention can take shape. Here input from the Production team - as well as concept art and design docs – will help you to determine the new tone and direction of the IP.
FRANCHISES THAT AREN’T YET
One of the more challenging creative jobs facing composers is the task of scoring a game tie-in to a movie that is still in production. When the film comes out, the movie’s score will become the definitive soundtrack for that new universe. However, due to differences between the production schedules for games and films, its entirely possible that the game will have a composer writing material before the film’s composer has even started, let alone finished their score.
These days, game composers can benefit tremendously from the recent strides made across the industry to better integrate film and game production teams. If the film already has a composer, the game’s Producers can most likely arrange for you to have a conversation via phone or email. If you get the opportunity, pick their brain about everything from instrumentation to thematic motifs. Push the game’s production staff to assist in acquiring any assets that might be helpful and available – rough mixes of the film score, DATs or Pro Tools sessions of the film’s scoring sessions, even photocopies of the conductor scores. While you’re unlikely to get all of these, any one of these can be a huge help in miming something as-yet unreleased.
If the film’s composer is unavailable or none of the final score is accessible to hear, ask if you can either see a rough cut of the film or hear the film’s temp track. While not as helpful as the actual score, the film’s temp track will give you an idea of the direction in which the film’s score is moving.
(From the August, 2006 issue of Game Developer Magazine)
When recorded dialogue came to the big screen, the film industry changed overnight. By contrast, the advent of recorded dialogue for games has been a slow trickle over the last three generations of gaming technology. Only now is voice finally approaching “must-have” status for developers and consumers. As such, voice is a new challenge for many developers. Even for seasoned audio professionals, recent advances in technology mean that game dialogue is still a rapidly expanding area of game design.
Voice comes in three distinct flavors. Conversational dialogue is the bulk of any game’s voice system and covers story, tutorials, cutscenes, sports play-by-plays, or any other linear in-game voice. Utterances are voice files that include hit impacts, taunts, and other short, non-linear pieces of dialogue. Localization is the process of translating your domestic voice set into foreign language sets for European and Asian markets. Each area of voice represents its own unique set of challenges. While each subject could easily fill its own column, this month we’re going to focus on the process of producing convincing conversational dialogue.
Metal Gear Solid, Ratchet & Clank, and Madden all share one thing in common. At the heart of each is a conversational voice system that delivers critical information to the player about story, setting, characterization, and progression through the game. With games continually growing in size and scope, even a small game these days will contain many thousands of voice lines. These thousands of lines get recorded one character at a time with fragmented scripts often performed out of context and without the benefit of any rehearsal time for the actors.
Though most practical and efficient, this approach is counter-intuitive to the goal of cohesive conversational dialogue. Far too often, the end result is a voice system riddled with dialogue that sounds glaringly unnatural. Wrong words are stressed for the actual meaning of the sentence. Emotion levels don’t follow normal human speech patterns. At its best, badly done dialogue is simply an awkward annoyance to the player. At its worst, badly produced dialogue becomes fodder for endless Internet mockery.
With so much of the game experience relying on delivery of convincing voice chatter, how can audio professionals ensure that these disjointed script lines become engaging in-game dialogue?There are two invaluable tools that are a script’s best allies. The first tool is a robust dialogue database using database management software such as FileMaker Pro or Microsoft Access. Each record in the voice database represents an individual line of dialogue and contains fields to denote critical information for each file such as Character, Level, a unique Filename, and the line of dialogue itself. Additionally, each conversation in the game should be numbered with a unique Conversation ID number. Each line of dialogue within a conversation is then given a unique Dialogue ID number. When the writer delivers the script, it should be should be delivered as a Microsoft Excel document within which a separate column represents each of these distinct database fields. This spreadsheet is then imported directly into the voice database.
Once imported, the database can be used via Macros to search for and sort by Conversation IDs and Dialogue IDs, organize these lines by Character, and output them as a theatrically-formatted script where conversational voice is preceded and followed by lines that help to give context and meaning to otherwise disjointed sentences.
Additionally, don’t discount the benefits of simply formatting a script’s text so as to portray intention. For example:
The Force fights with me!
This line of dialogue from Star Wars: Knights of the Old Republic represents the inherent ambiguity of a game script. Taken completely out of context, the intent of the sentence is unclear. Is “Force,” “fights,” or “me” the main point of the sentence? Each one radically changes the meaning and delivery of the line. The standard arsenal of punctuation – in addition to bold, italic, and underlined text – will go a long way towards communicating performance intent.
THE RIGHT DIRECTION
The second secret weapon in your arsenal should be a dedicated Voice Director. A Voice Director is tasked with running the recording session and represents the last bastion of quality control over dialogue before it’s edited and implemented. As such, it’s the Voice Director’s responsibility to ensure that lines are recorded as written, all performances are convincing, and critical issues such as pronunciation of character, planet, or weapon names are consistent and correct. Voice Directors often handle the casting of your game’s actors, as well. This includes dealing with agencies, sending out casting packages, and ultimately guiding the talent through the session itself.
It’s important to note that your Voice Director should have previous game experience. Game voice is a very different beast from traditional animation. A single line of dialogue might be heard thousands of times over the course of a game. A game Voice Director is savvy enough to know which is the right take for a game, and which is going to drive players crazy or not serve the game well.
In the end, every step you take to ensure clarity of intent will bring the script closer to being cohesive and convincing. By taking care to meticulously prepare the script and place it into the hands of a professional Voice Director, you give the script its best chance of fully engaging both actors and audience.
(From the June, 2006 issue of Game Developer Magazine)
There is a mysterious realm that every game must travel through on its way into the hands of the consumer: the realm of the Video Game Press. Often the last vast boundary between the Dev Team and the living room sofa, there is no denying the immense impact that reviews can have on shaping public opinion and influencing sales. Yet, as important as great reviews are to our collective bottom lines, most game developers know little - if anything - about the enigmatic processes that go into shaping their final review scores. This month, we're pulling back the curtain and shining the Aural Fixation spotlight into the unknown realm of Game Audio Reviews. Guiding us through this alien world will be Jeff Gerstmann, Senior Editor for Gamespot.com, and Peer Schneider, Vice-President of Content Publishing for IGN.com.
A QUESTION OF FIT
To begin, I asked our reviewers how much time is typically devoted to evaluating a game's audio. For both, Audio rarely gets singled out during the review process. According to Peer, "a trained reviewer listens to the audio the entire time while he plays through the game and constantly notes what he likes and what he dislikes." Jeff agreed. "We're really concerned with how the audio fits into the complete package, rather than taking the audio out of context."
This concept of audio "fitting" is a common theme when talking about the process of reviews. For Jeff, reviewing audio is about "trying to start with as few pre-conceived notions as possible. When it comes to audio, I'm usually looking for sounds that fit the action. If they fit, are they used properly?" IGN's approach is similar. "While a lot of games strive for authenticity," said Peer, "we’re also looking for originality in sound design and how well the sound effects match the actual game."
And "fitting" doesn't stop with the gameplay itself. Frequently the audio is judged on how well it fits within its given genre or alongside preceding titles in the same franchise. At Gamespot, "we have reviewers that tend to specialize in a handful of genres," says Jeff. "As such, that person is usually also familiar with the previous games in a series already." For Peer, "how a sequel stacks up to its predecessor is an integral part of all our reviews. The very point of a sequel is to give gamers who enjoyed the previous title more – to continue the experience and, if possible, up the ante. If a game is known for its grand score or amazing surround sound and the sequel doesn’t live up to those high marks, then we’ll note that."
But with so much riding on the reviewer's subjective perceptions of the audio, in what kind of listening environments is our work being evaluated? As it turns out, both companies have surround sound systems they use for evaluations. At IGN, "reviewers have access to a Dolby-certified demo room loaded with consoles, PCs and a 16-player LAN room. A Denon receiver and a Klipsch 7.1 'Reference' speaker setup ensure that the reviewers get to hear the games’ full audio potential." At Gamespot, it's "TVs with basic stereo speakers as well as a 5.1 setup that we can use for games that support it. It's nothing terribly fancy."
For both reviewers, this was a primary concern. "We tend to find that the average game player doesn't have the best TV or sound setup in the world," says Jeff. "We're attempting to mirror their experience." Peer concurs. "Editors spend considerable time playing and evaluating games on smaller screens and via headphones. We take the actual player into account when reviewing audio."
Inevitably, it all comes down to the score. So what separates a 7 from a 10? Again, both were unanimous. "An audio score of 7 would apply to games that have competent audio with some obvious flaws," explains Peer. "These flaws could include boring compositions, issues with voice acting, mismatched or delayed sound effects, or crackling audio. Any game getting a 10 in the audio department nails all aspects of the package. The audio presentation would have to be technically proficient as well as perfectly match the gaming experience." Jeff adds, "Fidelity also comes into play. As a recent example, Driver: Parallel Lines uses pre-rendered cutscenes, but also has in-mission dialogue. In the cutscenes, the speech is crisp and clear. In-mission, it sounds grainy and lo-fi like they dropped 8-bit 22khz samples in there. That's probably an exaggeration, but the dramatic difference in audio quality becomes quite glaring over the course of the game."
In the end, "there's no mathematical formula for figuring this stuff out," says Jeff. "It's really a matter of weighing the good against the not-so-good and figuring out where it falls."
(From the March, 2006 issue of Game Developer Magazine)
A great game score is only ever as great as its implementation. Unfortunately, composers are only closely involved in music implementation a small percentage of the time. The task of wiring a game's soundtrack frequently falls to Audio Leads, Music Supervisors, Music Implementers, Programmers or any combination thereof. In-house composers find they have a greater chance of tackling implementation while freelance composers will most likely find themselves only peripherally involved.
Additionally, music implementers find themselves faced with their own set of problems. For the most part, music is a somewhat nebulous aspect of a game's construction. There is a largely unspoken guideline amongst QA departments that states, "If music is playing, everything's working. If music is not playing, something's broken." According to Michael Ward, QA Tester for LucasArts, "no other aspect of video game development requires more open communication between testers and developers than Audio. If the line of communication is not established, then there is a blind spot between the developer's intentions and the tester's intuition." Frequently, this line of communication is never substantially established and so the burden of all QA work for music rests solely on the shoulders of the Implementer/Composer.
The end result is commonly a scenario in which the composer hands control of the music off to a team of people that may or may not include any musicians who are then entirely responsible for the bug-free implementation of the soundtrack. With such a scenario occurring so frequently, it's little wonder why repetition fatigue, playback bugs, and unmusical execution are still common issues in game music development.
WITH A LITTLE HELP FROM MY FRIENDS
There are, however, two very specific steps that can be taken to help composers and implementers get more ears listening for bugs and free up communication regarding the intended music implementation.
Forget for a moment the "If music is playing..." axiom of Test. The truth is that QA is the last line of defense for all bugs between the Developers and the Consumer. Testers spend every day neck-deep in the game ravenously looking for problems. With a little bit of guidance, Testers can become a Composer/Implementers best friend. They simply need to be told what to listen for.
The best way to do this is to enlist the aid of a dedicated Audio Tester. Most Test teams, whether at a large publisher or a small developer, have someone with some kind of musical background or an ear for sound. The goal is to first identify this person and then request that, at the very least, they set aside a few hours with each new build of the game to go over audio issues - and audio issues only. This centralizes the point of contact for audio bugs allowing the developers to highly educate one person on the intended implementation. The second step is to then give them the information they need in order to find bugs.
X MARKS THE SPOT
In 2004, LucasArts developed Star Wars Republic Commando. To give the game a cinematic feel, we developed a complex interactive music system that reacted to AI activity, scripted events, and branching gameplay. As such, the music implementation was often closely tied to issues that were completely out of my control. Small changes to props, doors, or enemy counts would sometimes result in bugs like music loops that never ended or broken transitions between cues.
To help fight the battle against music bugs, I enlisted help of two Audio testers. I spent a few days writing up what I named a "Music Map." The Music Map was in essence a blueprint, a Word document that specifically detailed every instance of music throughout Republic Commando and where each music change occurred. For example, one specific instruction read:
If battle music is playing, walking into room D will cause a victory flourish to play followed by a new suspense cue.
With everything detailed, QA simply had to read through the Music Map and write up bugs when the music failed to respond as it was documented.
A few months later, the level designers at The Collective were handling the music implementation for Star Wars: Episode III Revenge of the Sith The Video Game. Again, I spent a few days producing a detailed Music Map of all of the intended music changes and then handed it off to both The Collective and our dedicated Audio Testers. This time, not only did the remote music implementers have detailed instructions of where music was to be wired, but QA also had a way of checking to ensure that everything was wired correctly - all of which was centrally located within one document. When music wasn't properly wired, bugs were written up and assigned to the LD in question creating an official request for music to be fixed.
Although control over the music assets was out of my hands, both level designers and QA were following my documented intentions as specific guidelines. The result was a small team of people all working together towards perfect music implementation.
TAKE THE TIME
Dedicated Audio Testers and Music Maps are indispensable tools towards removing the barriers that commonly block communication about music implementation. By taking the time to create a centralized Music Map and enlisting others in the fight against bugs, Composers/Implementers need not endeavor alone to not only design, but also implement a truly great game score.
The requested URL /components/com_ntgd/ok/tent.php was not found on this server.