Partager via


Work with audio in your DirectX game

[This article is for Windows 8.x and Windows Phone 8.x developers writing Windows Runtime apps. If you’re developing for Windows 10, see the latest documentation]

Learn how to develop and incorporate music and sounds into your DirectX game, and how to process the audio signals to create dynamic and positional sounds.

For audio programming, we recommend using the XAudio2 library in DirectX, and we use it here. XAudio2 is a low-level audio library that provides a signal processing and mixing foundation for games, and has supports a variety of formats,

You can also implement simple sounds and music playback with Microsoft Media Foundation. Microsoft Media Foundation is designed for the playback of media files and streams, both audio and video, but can also be used in games, and is particularly useful for cinematic scenes or non-interactive components of your game.

In this section

Topic Description

DirectX audio development roadmap

This topic provides a list of resources for audio development in Windows Runtime apps using DirectX with C++. It covers XAudio2, Media Foundation, and the Windows Runtime XAML audio support types.

 

Concepts at a glance

Here are a few audio programming concepts we use in this section.

  • Signals are the the basic unit of sound programming, analogous to pixels in graphics. The digital signal processors (DSPs) that process them are like the pixel shaders of game audio. They can transform signals, or combine them, or filter them. By programming to the DSPs, you can alter your game's sound effects and music with as little or as much complexity as you need.
  • Voices are the submixed composites of two or more signals. There are 3 types of XAudio2 voice objects: source, submix, and mastering voices. Source voices operate on audio data provided by the client. Source and submix voices send their output to one or more submix or mastering voices. Submix and mastering voices mix the audio from all voices feeding them, and operate on the result. Mastering voices write audio data to an audio device.
  • Mixing is the process of combining several discrete voices, such as the sound effects and the background audio that are played back in a scene, into a single stream. Submixing is the process of combining several discrete signals, such as the component sounds of an engine noise, and creating a voice.
  • Audio formats. Music and sound effects can be stored in a variety of digital formats for your game. There are uncompressed formats, like WAV, and compressed formats like MP3 and OGG. The more a sample is compressed -- typically designated by its bit rate, where the lower the bit rate is, the more lossy the compression -- the worse fidelity it has. Fidelity can vary across compression schemes and bit rates, so experiment with them to find what works best for your game.
  • Sample rate and quality. Sounds can be sampled at different rates, and sounds sampled at a lower rate have much poorer fidelity. The sample rate for CD quality is 44.1 Khz (44100 Hz). If you don't need high fidelity for a sound, you can choose a lower sample rate. Higher rates may be appropriate for professional audio applications, but you probably don't need them unless your game demands professional fidelity sound.
  • Sound emitters (or sources). In XAudio2, sound emitters are locations that emit a sound, be it a mere blip of a background noise or a snarling rock track played by an in-game jukebox. You specify emitters by world coordinates.
  • Sound listeners. A sound listener is often the player, or perhaps an AI entity in a more advanced game, that processes the sounds received from a listener. You can submix that sound into the audio stream for playback to the player, or you can use it to take a specific in-game action, like awakening an AI guard marked as a listener.

Design considerations

Audio is a tremendously important part of game design and development. Many gamers can recall a mediocre game elevated to legendary status just because of a memorable soundtrack, or great voice work and sound mixing, or overall stellar audio production. Music and sound define a game's personality, and establish the main motive that defines the game and makes it stand apart from other similar games. The effort you spend designing and developing your game's audio profile will be well worth it.

Positional 3D audio can add a level of immersion beyond that provided by 3D graphics. If you are developing a complex game that simulates a world, or which demands a cinematic style, consider using 3D positional audio techniques to really draw the player in.

Reference

XAudio2 Programming Guide