From DAW to Game Engine. There must be a better way!

Question

Hey game audio folks. I'm wondering what you do.

Here's what I'm looking at: alt text http://www.wraughk.com/DropboxPermanent/SSD1.png

Here's what's happening. I have a sound effect that I've mixed a bunch of different layers of sound into (the first. I know I want some of these randomized, or separated into looping/ramp-up-to-loop assets. The first cluster of sounds is the whole effect, the next few clusters are the same thing with elements muted so I can render out regions that will be re-mixed together in the UDK.

But this isn't ideal, as I can't get instant response on the effect of what I'm doing in game? Once we implement WWise, I'll be able to play the sound effect (in WWise) and render over the assets (from Reaper) to hear new results, right?

What do you do?

Answer 1

I think you're asking a workflow question. I think I know what you're talking about and the frustration of working in a multitrack DAW only to deliver to another multi track audio engine where you have to rebuild it all anyway.

The DAW is only 1 part of a larger scheme of tools you need for "modular iteration and contextual previewing". Same thing with a wave editor that's only 1 part. I would also add a sampler / sequencer and the actual middleware to the mix to thats at least 4 independent tools. I run these on 3 separate machines networked by digital audio and MIDI over IP.

I make sure I spend the time to precisely balance the levels and match output configurations with all these tools and the game engine / audio middleware. So as I'm creating its within the context of the middleware and other reference playback sources all balanced and placed as they would be in game. Basically it takes time but you have to create the environment where the tools are flexibly routed allowing you to preview source material and your creative ideas as though they are "in game". I also have had a test kit and a retail PS3 balanced out in the the mix. I work like this in surround at medium volume. I also have a downmix to a TV set to low volume. When I need more precise detail I put on headphones that are set fairly loud. I NEVER change the volume my 5.1, TV or headphones .. EVER.

Once you're working more with the middleware you can decide if breaking the sounds into smaller parts works for the game and the content. If you decide to use the middleware to design the sound then your use of the DAW and other tools might be much more simple. The downside is that you'll be limited to the middleware technology. With wise that ain't so bad because its so good. It's not as great if your working in FMOD.

Answer 2

The downside to wav editors is they are clunkier than DAWs when it comes to layering (especially sound forge). Based on your session, you do a lot of layering.

My answer is not a very pleasing one, but trial and error leads to experience which leads to quicker development of assets.

wWise is an awesome tool. Judging from your session, I am guessing that you do a lot of your 'randomizing' in your DAW (it looks like you have 5 versions of the above effect). wWise gives you the opportunity to do some of that work in the wWise tool itself.

Also, randomizing the relative start times of a layered sound is just as effective at creating large numbers of unique FX as randomizing pitches and volumes of an unlayered (single wav file) effect. This can be done in wWise and can actually save RAM.

Either way, I find DAWs the best solution for now.