I recently did an interview for CDM (Create Digital Music). Feel free to read the interview here or from the
original source.
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
CDM Interview: Tomb Raider: Legend Composer Troels Brun Folmann on Adaptive “Micro-Scoring”
The latest installment in the famed Tomb Raider series has done very well under its new developer, Crystal Dynamics. Tomb Raider: Legend debuted at No. 1 on the UK sales charts and has sold nearly 3 million units since release. Composer Troels Brun Folmann, Ph. D. scholar in Adaptive Game Audio from IT-Universitetet i København, was recently awarded a BAFTA award for Best Original Score for Tomb Raider: Legend. We had the opportunity to talk with him about composing for games, adaptive music, and more.
Create Digital Music: Troels, first of all, congratulations on your recent BAFTA award for your Tomb Raider: Legend score!
Troels Brun Folmann: Thanks. I’m still having problems sleeping!
CDM: So this is something I struggle with on a daily basis (with my own surname) — can you tell us how to correctly pronounce your name?
TBF: Haha! I don’t know to be honest. When I order something at Starbucks, they always ask for my name twice. First I go, “Troels.” Then they go, “Charles?” Then I go, “Ehm…Troels.” and get a cup of coffee named “Charles”. The solution was to change my name to “Troy”, so I use that all the time now at Starbucks. But the most common English pronunciation is “Trolls”.
CDM: Tell us a little bit about your background, for those who may not have read your blog, etc.
TBF: The majority of composers always have these cool stories about how they played at the age of 2 and so forth. Unfortunately my story is a little less sexy. I couldn’t speak before I reached the age of 6 and I was generally regarded as a loss for humanity. Nobody in my family played music except of my grand-dad, who was a terrible hobby musician. When I was about 10 years old I decided to follow his majestic path. I eventually became a Ph.D. Scholar in Dynamic/Adaptive music for games and tried building bridges between the academic world and the games industry. But enough about me, please. You can find my narcissistic weblog at www.deffmute.com if anybody should care to drop me a line.
CDM: Adaptive music for games is a fairly new concept, what made you decide to pursue that field for your Ph. D. work?
TBF: Good question. I would not say that adaptive music is a new concept, but the problem is that it’s never really been working. One of the main limitations is the fact that current generation of consoles like the PS2 and Xbox have very little RAM allocated for audio. Typically, sound designers have a mighty 2MB to play around with. True adaptive music needs be generated in real-time, and even the next-generation consoles like Xbox 360 and PS3 will not have resources enough to do this on a larger scale.
However, there are ways to work around the limitations. This usually involves the creation of custom technologies. I invented a methodology known as “micro-scoring”. It’s basically the idea of chopping your score down to very small components and triggering them in a way that compliments the game experience.
CDM: For those who may be considering such an academic pursuit, do you feel that your post-graduate studies have helped you as a working composer?
TBF: This is a very difficult question. Did academia help me become a better composer? No. Did academia help me get a better understanding of communication, clients and business? Absolutely. I would advise aspiring composers and musicians to pursue conservatory degrees or just experiment on their own, since classic academia is far away from the game business. However, I would advise all professional composers to take communication, client relationships, networking, promotion, sales and business seriously. You may be the greatest composer in the world, however you need more than your music to get by, in my opinion. A talent is a person that excels under a given set of circumstances. Professional music composition is bread and butter. Communication is key.
Anyway, let’s get back to your question regarding academia. I do think there is a potential in academia to support this, but the current research on interactive music is highly sparse and I had a hard time finding serious research material to support scientifically valid statements, while I was a scholar.
CDM: Let’s talk a little more in detail about Tomb Raider: Legend. First off, this was the first TR title for Crystal Dynamics, but you had been working with them for some time. Can you tell us how you came to work with them, and how you eventually came to be the composer for TRL?
TBF: Tomb Raider has sold over 30 million units through its 10 year lifespan. Crystal Dynamics has a long history in action-adventure games and Tomb Raider: Legend was a unique opportunity for the studio to embrace the brand and renew it at the same time. I originally came to Crystal Dynamics as a Ph.D. scholar, while conducting field studies into the game business. I have always had a very pragmatic approach to things, since I don’t believe in dry and boring academia. I had previously worked with Crystal on their award-winning FPS “Project: Snowblind” and we had a very inspiring collaboration. Crystal Dynamics invited me over to conduct field studies and invent new methodologies for interactive scoring in Tomb Raider: Legend. The collaboration grew and I had an extremely high ambition for the score. My ambition was to do a consecutive, non-repetitive score for the game, while ensuring “Hollywood-quality” music at the same time. Game music has a nasty tendency to become repetitious and loop-like in its nature, so I invented the “micro-scoring” methodology to avoid this.
Micro-scoring is essentially about breaking the score into a variety of small components that are assembled in real-time according to player action and/or interaction. I scored over 4 hours of orchestral/electronic music for the game and it took me about 8 months to finalize the whole project. All the music was scored on my 9 PC renderfarm (which comprises approximately 20GHz of CPU, 20GB GB, and 10TB of storage).
CDM: TRL is obviously a very different project from Project: Snowblind, with an installed fan-base, a legacy of games and films - what was your initial approach for keeping true to the history Tomb Raider, while imparting your own distinct qualities?
TBF: It’s very delicate work, handling a highly known brand such as Tomb Raider; everyone has different expectations. The role of a commercial composer is to unify the expectations and channel them into something that will please most of the involved parties. It is my job to understand the needs of the client/publisher, the in-house team and producers, marketing and sales, fans and gamers - and compile all these expectations into the score. I am a commercial artist doing commercial art, which means that I have to embrace and innovate conventions and stereotypes. I know many artists that follow their heart and gut feeling, which I highly respect. But dealing with mass consumer scoring is about professionalism and scoring for the masses. I call it: ” embracing the cliche”, which is easier said then done.
CDM: In other interviews and articles, you’ve touched upon the music system for TRL. Can you tell us how the adaptive music system for TRL differs from other music systems?
TBF: Allow me to go into detail on micro-scoring. One of the ways that games differ from motion pictures is in the interactive nature of the media and the fact we can never fully predict player behavior. The micro-scores are made in such a way that they adapt to player action or interaction. You have to imagine that there are thousands of things going on in the game environment — the idea behind micro-scoring is to support the major elements in the environment. An example can be a 5-second score for breaking pillars or rolling stones. While motion picture scoring would typically have a musical element to support such an action, it would normally be dismissed in the game world.
However, micro-scores allow us to support that action. For Tomb Raider: Legend, we spent a long time creating a highly advanced proprietary streaming system that allows us to trigger micro-scores all over the game world. So, essentially, I can place scores for any change in the game, which is naturally a complex and time consuming process. The trend of games - particularly next-generation 360 and PS3 - is one of complexity. Everything is getting more detailed, whether its multiple translucent layers of textures, real-time generated light and shadow maps, massive streaming game worlds and so forth. Audio and music is no exception. The need for dissecting music into smaller fractions is becoming increasingly important in order to support the decisions and experiences of the player.
However, there is one component more important than any of the technical features, and that is one of emotion. I personally believe music is the emotional glue between the player and the game. Now the second question would be, “How often do we become emotional by playing games (aside from slamming our mouse or controller after being fragged)?” I believe the answer to this question describes how much we still have to face as composers in interactive media.
CDM: We are hearing more and more adaptive soundtracks in current and next-generation titles. Being able to stream multiple music stems at will based on any number of variables - player interaction, geometry, scripted events, etc. - not only changes the way we experience the game from a purely musical perspective, but can also draw the player into the action more, providing a more cinematic, and possibly emotional, feel to the experience. With that in mind, how has working with adaptive systems affected your composing methods? Put another way, how has working with an adaptive music system altered your composing style as opposed to composing for linear media (such as film)?
TBF: The ability to understand game technologies and mechanics is becoming increasingly important for the composer. The amount of complexity and micro-scoring will increase in order to ensure a smooth and complimentary game experience. There is no doubt that adaptive mechanisms do change the way I compose and approach scoring.
Let me provide you a little insight into some of complexities we faced on Tomb Raider: Legend. First of all any game level is made of smaller “units”. So whenever a player moves around in the level, she/he will cross multiple units and unit boundaries. Whenever this happened, I placed a score, so you have to imagine a variety of subsequent scores for each level in the game. We basically used our streaming engine to support this and a variety of cross fade and mixing techniques to have seamless transitions the score.
However, we also placed a variety of small micro-scores that support scripted events, cinematics and so forth. All these elements mean you have to keep a consistent approach to your scoring. You have to ensure music is in similar keys, so the score transitions don’t stand out. You have to ensure that rhythmic transitions are smooth, which can be really difficult with different measures and BPM. I sometimes had to create micro-scores that worked as transitions, so essentially scoring a bridge between the different scores. The hardest thing is to ensure that all this can work at anytime in the game. You never know when the player decides to do something, move forward or backward in the game. The score needs to follow every possible action, which is quite complicated.
CDM: As you compose more music for games, have you found any tools that are particularly useful or valuable?
TBF: One of the most important things to me is instant access to my music, which is why I had to custom build a 9-PC renderfarm for my composition. Orchestration is particularly demanding and I never liked bouncing much. The way I built my render farm is pretty simple actually. I have 2 PCs for strings, 2 PCs for brass, 1 PC for woodwinds, 1 PC for percussion, 1 PC for choirs, 1 PC for ethnic instruments and my main computer. I use a variety of commercial and custom sample libraries. My favorite orchestral library is East West Quantum Leap Symphonic Library (particularly the XP edition). The guys also released an amazing choir library, which allows you to type in any text and have the choir sing it. (Ed: See our episodic review of EWQL Symphonic Choirs.) They also have a fantastic percussion library called Stormdrum and soon coming out with Stormdrum 2, which is going to be even wilder. I also enjoy many of the open-source VST instruments and effects, but my main source for basic effects is my trustworthy UAD.
I am gradually moving more and more into custom recordings and sample design. I bought a variety of ethnic instruments from ethnic instrument online-store, Lark in the Morning, and used them all over the Tomb Raider score. I like musical sound effects a lot and these are hard to come by in the commercial library world. I bought anything from the Armenian Duduk to the Japanese Shakuhachi - from Bolivian Pan flutes to African Whale drums. My audio engineer bought a saw and I would highly recommend you … not do the same! We have also done custom sample recordings of two orchestras, which is something I would advise all professional composers to do at some point. The amount and quality of the content is absolutely amazing and allows you to shape your templates in a much more advanced and personal way.
CDM: Can you tell us a little bit about your ‘typical’ approach to composing a piece of music for a game? I realize your process may differ based any number of factors, but maybe you have one or two methods that you normally start out with?
TBF: It depends on what type of music I’m composing. When I orchestrate I normally flesh out the whole score as a piano composition. I used to have a more impressionistic approach and compose on the fly; however, I realized I got more consistent results when I sketched in advance. As far as tips and tricks, there are a few things I rely on when composing. First of all, I never use quantization on orchestration. A real orchestra never plays completely synchronously and this is an important element to keep in mind. It’s basically about invoking life into your composition and your samples. Sometimes I try to think as the conductor. Whether it is a slight change in tempi or trying to create more dynamic movements with crossfades and volume adjustments. Another trick is to apply individual reverbs to different orchestral sections. Strings, Brass, Woods, Percussion and Choir react very differently to reverb, so applying individual convolution reverb to each section is quite beneficial. And make sure you have at least one convolution in the master output section as well. It’s really about recreating the beautiful, complex reflections you hear in a concert hall room and you can do this by combining multiple reverbs.
When I do electronica it’s different. I have no consistent methodology for electronica. It can be anything from laying down a nasty groove to routing an arpegiated synth through a glitch module. Anything from putting down a deep drone and placing a super reverbed female voice or tuning down my drums 24 semitones, distort them and see where it goes.
So long story shortÖ I compose differently for each individual style of music.
CDM: Aside from the previous Tomb Raider scores, what other music influenced you during this particular project?
TBF: Oh dear … You ready? Hans Zimmer, Howard Shore, Pat Metheny, Thomas Newman, Gabriel Yared, David Arnold, Herbie Hancock, Boomjinx, BT, Jerry Goldsmith, Gorecki, Hans Gregory Williams, Don Davis, Danny Elfman, Alan Silvestri, John Williams, James Newton Howard, Edward Shearmur, St. Germain, ES Posthumus, Gustav Holst, Bill Brown, James Horner, John Barry, Trever Rabin, John Adams, Bernard Hermann, Alex North, Peter Gabriel, Sting, Sly and Robbie, Enya, Ennio Morricone, Thomas Bergersen and many, many others.
CDM: Anything new and exciting on the horizon for you?
TBF: I am currently working on some unannounced projects, which are stretching my compositional palette further than ever. Imagine something between epic orchestral, glitch based electronica, large percussion ensembles and full symphonic choirs.
CDM: Thanks for spending time with us!”
TBF: The pleasure has all been mine. If anybody wants to know more, feel free to contact me through my weblog at www.deffmute.com.
Source: Create Digital Music, October 11th, 2006.
