AMD Reveals AudioKinetic Optimisations Which Reduce CPU Strain From HQ Audio
Ryan Martin / 3 years ago
As part of AMD’s new TrueAudio technology (which was revealed earlier – check details on that here) AMD have revealed AudioKinetic optimisations that will help to improve the way audio is delivered. The optimisations build on what we’ve already seen from AstoundSurround (check that here) and AudioKinetic offers ways to reduce the CPU load of the audio experience and increase the overall quality of the audio experience further.
AMD’s TrueAudio technology really steps up the hardware demand required to operate this new level of HQ audio. Therefore it’s important that AMD finds ways to offload the CPU demand of the audio experience in innovative ways and that’s what the new AudioKinetic technologies apparently do.
One of the main new developments for improving audio quality is Convolution Reverb which allows game audio to use impulse responses. These are based on actual recordings of real sound or on “audio photos”.
However, this ultra realistic high quality audio comes at a price and the longer the audio clip is, the higher the CPU usage becomes.
The AudioKinetic optimisations have ways of averting these bottlenecks by using two runtime FFTs and introducing more memory and shared data caches to speed up the transfer processes.
In AMD’s opinion the AudioKinetic optimisations combined with the AMD TrueAudio package will allow developers to create better in-game audio which should create happier gamers!
Images courtesy of eTeknix via AMD LiveStream