Mixing In-Game Cinematics





Under normal gameplay conditions, the listener or “virtual microphone” is typically positioned at or near the location of the camera, and the sound sources are modeled where they really are in the environment. Distance-based attenuation, direct and indirect sound path determination, voice limiting—all are determined using these realistic positions. However, during an in-game cinematic—a portion of the game in which player control is suspended so that a story moment can take place—the camera often pans out away from the player’s head. This kind of thing tends to wreak havoc with our 3D audio system. We could just keep the listener/mic locked to the location of the camera; but this is not always appropriate. For 824 13. Audio example, if there’s a long shot of two characters speaking, we probably still want to mix so that the characters’ voices can be heard, even though physically speaking they are too far away to be heard. In this case, we might want to detach the listener from the camera, and artificially position it nearer to the characters. Mixing in-game cinematics is a lot closer to mixing a film. As such, a sound engine needs to be capable of “breaking the rules” and doing things that aren’t necessarily physically realistic. 13.5.9 Audio Engine Survey It should be evident by now that creating a 3D audio engine is a massive undertaking. Luckily for us, lots of people have already put a great deal of effort into this task, and the result is a wide range of audio software that we can use pretty much out of the box. This ranges from low-level sound libraries all the way to fully featured 3D audio rendering engines. In the following sections, we’ll survey a few of the most common audio libraries and engines. Some of these are specific to a particular target platform, while others are cross-platform. 13.5.9.1 Windows: The Universal Audio Architecture In the early days of PC gaming, the feature set and architecture of PC sound cards varied a great deal from platform to platform and vendor to vendor. Microsoft attempted to encapsulate all of this diversity within its DirectSound API, supported by the Windows Driver Model (WDM) and the Kernel Audio Mixer (KMixer) driver. However, because vendors could not agree on a common feature set or set of standard interfaces, the same functionality would often be realized in very different ways on different sound cards. This required the operating system to manage a very large number of incompatible driver interfaces. For Windows Vista and beyond, Microsoft introduced a new standard called the Universal Audio Architecture (UAA). Only a limited set of hardware features are supported by the standard UAA driver API—all remaining features are implemented in software (although hardware manufacturers are still free to provide additional “hardware acceleration” features, as long as they provide custom drivers to expose them). Although the introduction of UAA limited the competitive advantage of prominent sound card vendors like Creative Labs, it did have the desired effect of creating a solid, feature-rich standard, which could be used by games and PC applications in a convenient way.

Post a Comment

0 Comments