389a2f7b4b
The fontsound still maintains its own start, end, and loop offsets, so that the same buffer may be shared between multiple/all fontsounds. Ideally a single buffer should be used for all fontsounds to avoid memory fragmentation and help CPU caching, although higher quality soundfonts may need more memory than a single buffer can hold.
Source Install ============== To install OpenAL Soft, use your favorite shell to go into the build/ directory, and run: cmake .. Assuming configuration went well, you can then build it, typically using GNU Make (KDevelop, MSVC, and others are possible depending on your system setup and CMake configuration). Please Note: Double check that the appropriate backends were detected. Often, complaints of no sound, crashing, and missing devices can be solved by making sure the correct backends are being used. CMake's output will identify which backends were enabled. For most systems, you will likely want to make sure ALSA, OSS, and PulseAudio were detected (if your target system uses them). For Windows, make sure DirectSound was detected. Utilities ========= The source package comes with an informational utility, openal-info, and is built by default. It prints out information provided by the ALC and AL sub- systems, including discovered devices, version information, and extensions. Configuration ============= OpenAL Soft can be configured on a per-user and per-system basis. This allows users and sysadmins to control information provided to applications, as well as application-agnostic behavior of the library. See alsoftrc.sample for available settings. Acknowledgements ================ Special thanks go to: Creative Labs for the original source code this is based off of. Christopher Fitzgerald for the current reverb effect implementation, and helping with the low-pass filter. Christian Borss for the 3D panning code the current implementation is heavilly based on. Ben Davis for the idea behind the current click-removal code.