Latest 'Steam Audio' beta taps AMD GPUs for spatial audio in games and VR

AMD has teamed up with Valve to bring its TrueAudio Next technology to Steam games and VR applications. As such, Valve announced that its Steam Audio 2.0 beta release now supports TrueAudio Next, which allows developers to use a reserved portion of a support GPU to generate spatial audio.

"Combined with Steam Audio's ability to model a wide range of acoustic phenomena, this enables increased acoustic complexity and an increased sense of presence in games and VR applications," Valve said in a blog post.

It was around this time last year when Valve launched its Steam Audio SDK as a free beta for developers. Steam Audio uses physics-based simulation to model two ways in which sound travels from a source to the listener—a direct path, sometimes through solid objects, and indirect sound that bounces off of in-game objects.

"Simulating and rendering indirect sound is the most computationally intensive part of what Steam Audio does, and TrueAudio Next helps manage this workload," Valve explains.

TrueAudio Next is a software library that implements a high-performance filtering technique called convolution for audio data on supported GPUs. It also provides a feature called Resource Reservation that allows a developer to use a portion of the GPU's computational resources—20 to 25 percent—on time-sensitive tasks, in this case audio processing.

According to Anandtech, TrueAudio Next can be enabled and disabled on a scene-by-scene basis. This would allow a developer to disable TrueAudio Next in graphically intensive scenes that need every bit of computational power available.

What all this ultimately means for gamers is more immersive audio experiences, at least eventually. Offloading audio chores to a portion of the GPU also means the CPU will have more resources to throw at other tasks, such as AI and physics.

Developers who want to play around with this can grab the latest Steam Audio beta from GitHub.