[In this special Gamasutra interview roundtable, the musicians behind Flower, Trino, Sushi Bar Samurai and more exchange views on the strong suits of the independent game scene and its music.]
At the Game Developers Conference in March, four composers met for an indie roundtable
to exchange views on the strong suits of the independent game scene and its music. The participants were Chris Schlarb (NightSky
), Baiyon (PixelJunk Eden
), Shaw-Han Liem (I Am Robot and Proud
, Jon 'Everyday Shooter
' Mak's new game) and Vincent Diamante (Flower
Now, in this recent follow-up conversation, musician, photographer and Gamasutra writer Vincent Diamante
returns to share thoughts on his experiences working on audio at ThatGameCompany. He is joined by Matthew Wasteland
, columnist for Gamasutra and Game Developer Magazine, who is currently working on a music title at Shadegrown Games
Also participating is Casey Muratori
, creator of Sushi Bar Samurai
, showcased as part of the PAX 10 selection of independent games last year. Including an interactive music score, the gameplay envisions sushi preparation occupying a central role in restoring balance to the universe. Representing this year's PAX 10, Soo Jeong Bae
produced and wrote the music for Trino
. An audience-pleaser at the Seattle expo, the game was a product of Carnegie Mellon's Entertainment Technology Center.
The group chat takes a look at how the Penny Arcade Expo has gone about creating a convergence of community-minded and game-related activities. The discussion also poses some reasonable requests for the Xbox Indie Games portal and concerns with the development processes behind many triple-A game scores.
Soo Jeong Bae
In the first years of the PAX 10, Carnegie Mellon has gone two for two in developing games featured at the Penny Arcade Expo. Soo, how in your observation has the design of the college program managed to contribute so successfully in fostering independent game design?
Soo Jeong Bae, Trino composer:
The Entertainment Technology Center
started from a class called ďBuilding Virtual WorldsĒ taught by a professor named Randy Pausch
. Itís meant to combine right-brain people and left-brain people: drama, arts, music people and the CS guys. It asks what happens when theyíre put in a project room and they have to make something within 14 weeks in a semester.
is the main faculty adviser for game design courses. He was pitched a project by a group of students for the IGF. (That game was Polarity
.) [Ed's note: The students pitched Polarity
to Schell and the rest of the faculty.] We, on the other hand, just wanted to make something for the Xbox Live Arcade. Somehow it ended up here after two semesters of work.
Which brain are you?
Iím a right-brain, left-handed person. I come from an art background and did music composition when I was younger. At one point I felt I wanted to be more specialized, because when you apply for a job the first thing they ask is: ďWhat do you do best?Ē This is part of the reason why Iím joining a start-up: When youíre in a start-up, itís not a bad thing to be doing multiple tasks.
How were design decisions made on your game Trino?
Whenever decisions had to be made, we would have meetings. People would share their ideas, and we tried to make most of the faction happy. Sometimes people disagreed, but we didnít have too much time to fight. We had to keep moving.
There were also two faculty advisers. Itís good to have advisers, because often youíre looking at the same stuff over and over for weeks. People outside the loop can often tell better what's working and what's not. We learned that playtesting early on, and it often helps design decisions to be made quicker. Like I said, we had to move fast. Trying things out was the easiest way to tell whether those decisions were good or bad.
This might be one of the advantages of having a semester schedule as a frame for a development cycle. In the making of Flower, did you feel your collaborators had a strong idea in mind of where to take the game, and were they open to your providing input?
Vincent Diamante, Flower composer:
During this last GDC, the Flower
team was showing off some of the various prototypes that existed over the course of development. Elements of the camera and of flying were very well defined at the beginning. Previously with Cloud
had a certain amount of trust in me. I also had a strong idea of how I would use the music to influence him to my particular way of thinking.
Matthew Wasteland, Shadegrown Games:
It's interesting you mention influencing Jenova with your music. That's really rare in the triple-A videogame industry. Usually it goes out to the composer with simple requests - ďWe need this kind of music and that kind of musicĒ - then it comes back, and they put it in the game, and that's it. Changing something in the game because of what the composer is doing is pretty unheard of. With a game like Flower
, though, I think it shows in the final product.
Vincent Diamante and Casey Muratori
What has been your experience being part of the PAX 10 this year?
We knew we needed more promotion, and being at PAX definitely helps. We submitted the game to IndieCade, Dream-Build-Play, and the IGF. Because we donít have money to market ourselves as an indie developer team, the best we can do is submit our game to these contests.
Being on the show floor, have you come across any feedback thatís been useful to you?
A lot of people ask if the game is available anywhere.
For the record, Trino is available now on the Xbox Indie Games portal, and it's been getting some great feedback. We've been hearing from a lot of people that Microsoft's portal should be doing more to get games like Trino into the hands of their intended audience. Can we take a moment to come up with a brief list of suggestions on how to improve the service?
More advertising. Also, it takes one too many steps to get to the games.
Casey Muratori, Sushi Bar Samurai composer:
I think curation would be big. In general, if you think about how many community people there are out there who dedicate a lot of time to writing indie game blogs, why canít I get on Xbox Live and see games selected by them? I canít go see what the top 25 most played are, or what the top 25 highest rated are... Worst 25? Iíd probably go look at that, too. I want to see the crappiest game on Xbox Indies!
Despite the Indie Game section being there, itís as if itís not actually integrated into the Xbox experience. Itís very hard to get to it in the interface. The first time it really struck home for me was when I saw that Xbox Indie games are not present in the Quick Launch menu. You can access the game thatís in your disc tray, or any Xbox live arcade games, but you canít access any indie games.
Did you both receive feedback on your music at PAX?
Not really, because you canít hear the music on the floor of PAX.
I wanted to get some feedback on the music, but itís too hard to hear on the floor.
In addition to bugging the people at Xbox Indie Games, we might need to agitate for there being headphones accessible at PAX 10 kiosks.
Or sound cones. There are these cones that basically focus the sound down on you. You stand in this circle and you hear it, but if you step outside the circle you donít hear it anymore. The first time I ever saw one of these was in the lobby of EA Redwood Shores.
Casey, in your Sushi Bar Samurai interview last time you controversially stated your opinion that the professional audio software on the market today is useless to game designers.
Itís probably more up to people who do music performance to complain about music software. However, in general if you want to compose game music, the idea is that you want to put as much interactivity into it as possible. This means in general, about the worst thing you can do is give someone an interface thatís just going to record what they do and then play it back as sequences.
You canít really do even straightforward stuff in most music packages. You might want to render out every measure separately and leave the overhangs so that when the programmer wants to put them together they just work. Trivial stuff like that is not in there. Thatís just a very basic thing that you might want, and no one has even remotely the tools for it.
Iím far from an expert in Ableton Live, but it apparently has more things where you can build blocks, and so do mod authoring tools, but still theyíre really primitive. The easy proof is that music tools are basically the same as they were ten years ago. Game music is way bigger and more complicated than it used to be and the tools are still exactly the same.
Modern music software just isn't designed with the idea that the music might be controlled algorithmically from a game. It's mostly geared towards recording, with some other programs emphasizing live performance. Game composers by and large have to use the same tools that film and TV composers use, and somehow make it work with, or around, an interactive medium.
For a big budget game, the reputation of the music score will often be assessed by making direct comparisons to non-interactive media, the features of film and television scores.
If youíre going to take interactive music seriously, you need to take it at the stage where youíre still working on the interactive music engine. You have to start thinking about it way earlier than just putting a score in after the fact.
In terms of interactivity, my experience has been ďthe more the better.Ē Thereís never the case where Iíve had interactivity in there, where something is composed on the fly, and Iíve not found it to be an improvement. It complements the experience better, because the more parameters that youíre feeding into the music system, the more itís going to sound like itís meant to sound at that particular time.
The other part is that it makes the music more interesting, because in general people are playing games for a lot longer than they are listening to a particular piece of music. People will play a game for 20 hours these days. Very few people will loop a CD for twenty hours. Having the interactive music changes things up a lot, and makes it so that I think you get less bored of what youíre hearing.
A lot of thinking about music in games is pretty lazy. It's a case of developers or publishers liking the emotional impact a good film score can have, and thinking that hiring the same guy, or getting the same sound, will necessarily lead to the same impact in an interactive medium. So you get a lot of hiring of famous film composers on triple-A titles.
For some reason, even though these people implicitly understand that hiring a famous Hollywood screenwriter won't necessarily make for a better game, they don't feel the same way about hiring a famous Hollywood composer. It's too bad, really, because I would argue that Hollywood has much less to teach us about scoring than most of the industry thinks it does.
The Secret Of Monkey Island
Vincent, at GDC you came up with an early example of interactive game music with Ballblazer. Any other pioneering game scores come to mind?
and TIE Fighter
. When the first games came out, they sounded fantastic. They were MIDI-based. Sometimes you would hear an audio cue from off-screen, and you would be like, ďOh, man, an imperial capital ship just warped in.Ē
It was fantastic. Then they put out X-Wing vs. TIE Fighter
on CD-Rom. There, they put in the John Williams score and they lost the interactivity. Sometimes the interactive music was more communicative than the status bar.
I could not agree more. The same things kind of happened with the adventure games. The score for Monkey Island 2
was insanely good. It had themes for every region, and it blended smoothly between them as you went. When new characters would come in, occasionally they would have a theme. Largo LaGrande would make his entrance, and it would switch to his theme.
They went to 3, and they still supposedly were doing interactive music, but it just didnít have that feel where you would walk from place to place and it would smoothly transition to the next track. Sometimes I would walk back and forth between scenes just for that.
Wasnít that cool? Starting with Redbook Audio, everyone began focusing on music that sounded good and wasnít actually really functional in the way that MIDI was. The previous music could respond to you in a way approaching what the visual feedback was. That just wasnít possible with CDs.
We have no excuse now. People donít want to take any steps backwards in terms of audio fidelity. We donít have to, but we need tools that can do interactive music while still sounding good. Someoneís got to provide those tools. It would be great to license a soft synth library that could do an orchestra. Whereís that product?
Iím using a tool right now called Synful
. Itís an orchestra thatís all CPU. Itís not sample-based at all. Itís got a pretty small memory footprint: itís a VST plugin. I asked them, "Whatís the possibility of making a runtime version of this thing for use in applications?" It turns out they werenít really thinking of that at all.
A lot more is possible in real-time now, but somehow we just aren't exploring that as much as we should. For example, Halo 3
actually uses a Waves plugin at the very end of the audio chain - the L360 Surround Limiter
. Waves licensed their algorithm, and Microsoft did most of the work to get it to run on the 360. But even though it makes the game sound really good, and it was clearly fast enough to run on top of Halo 3
in the Xbox 360, not a lot of other developers have expressed interest in pursuing the idea of getting real pro audio tools onto the console. Waves themselves don't see it as a big area of growth for them.
What interest do you have in CD soundtracks for music games, where interactive music is given a set structure?
My answer to that question is, Why be interested in taking something that someone composed to be interactive and putting it in a non-interactive form? Take your iPhone: itís got a bunch of stuff. Itís got a microphone (so it knows the ambient noise level) and an accelerometer (so it knows if Iím walking how fast Iím walking). Why not start shipping interactive soundtracks as interactive
, so that they still recompose on the fly? Thatís what Iíd rather do. The user could say, ďIím a little bit sad right now,Ē and it would do that. Why ship it in an old, crusty format that sucks? On the iPhone you could literally ship an app thatís the interactive soundtrack. You push Ďplayí and off you go.
I really like that idea. I was actually thinking back to the mods and demoscene of the early Ď90s. They didnít just release mods, they also released programs that would run through them in interesting ways. Now that we have iPhones and DVD audio discs, those have interactivity possibilities. It would be nice to take advantage of them.
However, about the stuff you could put in to make a score more interactive, themes are also important. Having those things that are static that people can latch onto are important, just as the design for a character doesnít really change in a game. Mario is Mario.
You can always take a theme and make fairly drastic musical modifications to it that a human ear still identifies as that theme. You can change a chord from major to minor, and people will hear it as ďthe sad version.Ē You can end it on a discordant note, and theyíll be like, ďOh, something bad happened at the end of that theme.Ē They donít interpret it as a whole new theme.
Interactivity, for me, has never meant that you have to give up your strength. Music is bendable, just like Mario rendered from the side and Mario rendered from the front are both identifiable as Mario.
I suppose for certain types of music it does work that way. As a music composer, sometimes I get really attached to this very specific way in which this theme is presented. This happened a couple times on Flower
, where I would rather create an entirely new theme than modify what I had.
Really? Iíve never had that thought. Usually Iím happy when the Sushi Bar Samurai
music system plays something back that I had never considered. Itís like having another sound designer on the project that does interesting stuff on occasion. Also, I can always go in and tell him to stop. ďThat was bad. Donít ever do that again.Ē
Thatís true. Sometimes Iíve been surprised by the things Iíve heard in Flower
, and then saw if I could tweak the parameters of the soundtrack to see if I could get more melodies like that. But, you know, sometimes I get really attached.
How did you overcome the obstacles you've observed with using audio software in creating the interactive music system for Sushi Bar Samurai?
Casey: Sushi Bar Samurai
does not use any audio software. Itís one hundred percent in the game. When youíre in the game, you can hit a debug key, edit your patterns in there, and itís all stored in an interactive format. All things are always decomposed at all times. There is never a time when it has to go through some kind of a sequencing tool or anything like that.
What do you think? Do you use music creation tools and have you noticed a lack of capabilities?
I agree. I started out with Digital Performer, and then I went to Cubase, because those were the two at the time that were big on what you could do with MIDI. Then I moved eventually to Cakewalk. Iíve decided to stick with it for now because I like it enough, but still there are those little things, like not being able to render out separate stems instead of everything. The interactivity in Flower
is actually fairly simple, but even rendering out the music for that became a chore, because the software is not made to do stuff like that.
It's a time-consuming process everywhere, and it shouldn't be. When Marty [O'Donnell, of Bungie]
is implementing a track, he has to spend a tremendous amount of time cutting up the track in Pro Tools, putting it back together in the game tool, and listening to it over and over again, making sure there aren't pops or clicks at loop points. And he'd have all these little audio files to keep track of then, ďLoop A,Ē ďLoop BĒ and ďLoop A to Loop B.Ē There are no tools that I know of that can be smart about how that kind of music works. Itís not super-complicated, right? Itís loops that transfer to other loops, plus ins and outs. Adding that kind of feature set to an audio program doesn't seem difficult, but so far it hasn't happened.
Vincent, youíre teaching music in a university setting. Do you agree with Caseyís opinion that the technology is behind?
Itís not so much that in the university setting people care about the products that are used being consistent, but thatís the result of a process passed down from interns to students that needs to remain stable. As a result, if equipment breaks down, what do you do? You go to Guitar Center and get something that plugs right back into that whole process chain that youíve developed over the years.
Some of these guys I know who have been recording engineers for a long time will make a new studio in a shiny new space in LA, and here they have an expensive new outboard reverb line. How does it compare with what we can do these days with convolution reverbs? It doesnít. But it fits right into that slot thatís really ingrained in their particular studio culture.
It even has a vacuum tube in it. Donít forget that if you put a vacuum tube in anything, that becomes instantly valuable to musicians. If you sold vacuum tubes on the end of a USB, every single musician would buy it.
Weíre kind of going on this side track, but it does kind of say something about how it feels like a lot of the game audio guys that I met in the LA area and other places are really separated from game development. After five or six years of being in this world of game audio, itís scary to realize that this is the norm.
A lot of game composers, especially the ones with a film background, are pretty divorced from the actual equipment. They put notes down and itís an arranger or synthesizer programmer who chooses the samples and records the audio.
When I was a student I was a fan of in-house musicians, like Michael Land
at LucasArts or Sakimoto
at Square. You donít see that so much here. You have a guy that comes in and fulfills a contract.
Bungie is known for having an in-house composer.
That's one of the big reasons they excel in audio. Marty is actually there, harassing people if the game doesn't sound right to him.
Sushi Bar Samurai
This reminds me of Caseyís "About Us" page on Molly Rocket, where he contends that due to the domination of impersonal development processes, the game industry is rotting from the inside out.
In the game industry youíre at a point now where itís big business. Itís good in that any time you want to play a first-person shooter, there are ten. Thatís not a bad thing. Itís just that if you play a lot of games, after awhile thatís going to get old. You might want to play a game that has some unique things that were maybe a little more connected to the people who made them.
To some degree I think that requires that you have a smaller group of people to make them and that those people really be invested in it. They care a lot about making this specific thing, and not just about making ďa game,Ē or making money. Itís tough, because youíre going to have less of that.
Weíve certainly seen examples of this. There are no shortage of inspired game designs coming from teams that defect from large game companies in the hopes of being able to make their personal artistic choices.
The definition of a personal artistic choice is that it was not made to appeal to the largest number of people. Theyíre juxtaposing forces. If youíre going to do something because it meant something specifically to you, then that is by definition not a decision that you made by playtesting and considering what would appeal to the widest audience.
There are some things you can do with software in general that are different from other sorts of things that are artisan-like. There isnít a cost for reproduction, and thatís kind of cool. Any artisan game can be played by as many people who want to play it, with no additional cost. Thatís pretty awesome.
The same is now true of music. In the age of digital distribution, if someone makes a really great piece of music, everyone who might want to be able to hear that, can. Games are in good shape in that you can seek out and find the kinds of artisan games that you want if they exist, and you donít have to pay an exorbitant price to get them.
To conclude our talk, what does everyone think of the Penny Arcade Expo overall?
This is my first PAX. It seems like a really well developed convention. Thereís a lot of energy. I love seeing fans having fun within the culture of gaming, and knowing the stuff that we make pervades more than just the television screen. Having a three day weekend where people can do that is just cool.
It has the feel of a community.
I did almost a complete rewrite for Sushi Bar Samurai
following PAX. Prior to going in, I had a pessimistic view of what kind of people play games, even though I myself play games a lot. Just from having worked in the industry too long, I think you kind of get this really soured opinion of everything. So many awesome people came up and played the game and had so many interesting things to say.
Overall, I was really surprised with how engaged people get with a particular game and how much theyíre willing to bring to it. It gave me a lot more confidence. You donít have to just make something thatís "easy to play, simple, addictive," and whatever. That goal of how you sell a million copies of something, you can forget about it. If youíre a small indie, you donít need to sell that many copies. Thereís a fairly large audience of people who want something interesting and different.
I think itís easy to lose sight of that if youíve worked in games too long. For me it was a formative experience, to be honest, because I was not expecting it. From now on Iíll be thinking of those 200 people who came up to play the game, and that replaces this image of a faceless dude at EB Games who picks up the controller and plays some stupid hockey game for five minutes. Thatís who you start thinking about after awhile. I canít thank PAX enough for changing my mind about this.
Itís fun to be around gamers. They get these twinkling eyes and you receive the kind of appreciation you want to have. Sometimes you develop a game and you have no idea whoís going to play it or how they feel about it. Here, theyíre playing it right in front of you and they give you all this feedback. I love those moments.
The show has a very different vibe than E3 or GDC. It's great to be reminded concretely just how many people love games and want to share their enjoyment with others. There's no pressure to do business or prove you're some important person, you can just relax and be a fan like everyone else. PAX reminds me why I wanted to make games in the first place.
[Images courtesy of ThatGameCompany, TrinoTeam, LucasArts and Molly Rocket. Photos by Jeriaska.