Categories
Audio&Games

Game audio stack vs design

I watched the new Unity5 audio system stuff, it’s pretty cool. They don’t dig so much in how performance is when you use a granular synth and sub mixes on top of an actual game with AI and input though. They are going full native avoiding the managed code part so it’s promising but even “offline” audio processing using native plugins is heavy on recent CPUs, I don’t think we’ll be able to use those effectively in-game soon at 60fps.

They are a bit if not a lot overlapping with well known and used third party solutions like FMOD and Wwise.

Miyamoto once said:

A good idea is something that does not solve just one single problem, but rather solve multiple problems at once.

I like this quote because advanced audio features tend to solve one single particular problem while adding tons of complexity while simpler solutions can not only solve a sound problem but also a game design, UI feedback problem at the same time.

My focus as a designer is not to use fancy tools to the max. More features doesn’t mean it’s going to make things so much better. Two things make me think that:

– Most game audio that people remember is not related to anything technical, it’s usually from games where audio is played or stopped. No convolution reverb or complex variables to make something duck –4.5 dB in the background.

– As Iwata, a fantastic programmer said after working with Miyamoto: “it’s about content”. Even though technical mastery is useful ultimately it’s about what people will hear, the content.

And that works for me: I’m enjoying Else Heart.Break() made in Unity which has basic audio integration but the content is great: the music is awesome, the foley and little sounds are cool (I really like the fake TV show sound). Content first.

GunPoint has some clever FMOD layering going on but if the music was bad, it wouldn’t have mattered. And the layering (happening when you switch to hacking mode) can actually make it tiresome from time to time.

Kentucky Route Zero is the master of the past few years in terms of ambiance and mood, achieved by fantastic audio design with about zero fanciness or parameter-enabled function. It’s just a great sounding game.

With time I feel like people enjoy and remember two things from game audio: music they can hum, and sound/voice effects they can remember because they heard them a billion time (“KAY-O” and other “the bomb has been planted”). It seems like the loop/repeat paradigm is part of game audio’s DNA and we can’t escape it.

The only games that require massive audio systems and tech are the CoDs and Halos of this world: big battles, hundreds of sounds being played, the need for a robust system to control what’s going on is inevitable.

But for most games? Nah. Audio designers, focus on making great sounds. That’s what will make the game better.

Leave a Reply

Your email address will not be published.