Share via

First Steps in DirectSound Programming

  • Article

[The feature associated with this page, DirectSound, is a legacy feature. It has been superseded by WASAPI and Audio Graphs. Media Casting have been optimized for Windows 10 and Windows 11. Microsoft strongly recommends that new code use Media Casting instead of DirectSound, when possible. Microsoft suggests that existing code that uses the legacy APIs be rewritten to use the new APIs if possible.]

This topic provides a brief overview of the steps you must take to play a sound by using the DirectSound API. For details, see Using DirectSound.

Playing a short sound requires the following steps:

1. Create an object that supports the IDirectSound8 interface

Call DirectSoundCreate8 to create an object that supports the IDirectSound8 interface. This object usually represents the default playback device. Optionally, you can first enumerate available devices to obtain the GUID of the device you want to use, and pass this to DirectSoundCreate8.

DirectSound is based on the Component Object Model (COM). However, you do not have to initialize COM explicitly unless you are using effect DMOs.

2. Create a secondary buffer

Use IDirectSound8::CreateSoundBuffer to create a buffer object that will contain sound data. Such buffers are called secondary buffers, to distinguish them from the primary buffer, which is the object that mixes all sounds.

3. Obtain PCM data

Read data from a WAV file or resource into a private buffer.

4. Put data in the buffer

Prepare the secondary buffer for a write operation by calling IDirectSoundBuffer8::Lock. In most cases, this method returns a single memory address. Copy data from your private buffer to that address, and then call IDirectSoundBuffer8::Unlock.

5. Play the buffer

Play the sound by calling IDirectSoundBuffer8::Play. You can instruct the buffer either to stop when it reaches the end, or to continue playing over and over again until you call IDirectSoundBuffer8::Stop. You can start and stop the buffer repeatedly; however, only one instance of the sound can play at one time, unless you create other buffers containing the same data. Any number of buffers can be played at a given time, and mixing is done automatically.

The preceding steps apply to the simplest possible scenario, that of a short sound that fits into a buffer of a reasonable size. Typically, sounds that last more than a few seconds are streamed: that is, data that has already been played is periodically replaced by new data.