Lately I have been presenting Spatial Audio panning/decoding shoot outs to the engineering team at Abbey Road. I used a variety of material, from mono microphones, to Ambisonic source recordings and even 5.1 surround mixes from both music and film. One thing that quickly became apparent to me when I started exploring Spatial Audio is how different all the various software options sound. It was really reassuring that during these tests all the engineers and myself felt the same way about the various options this sounds a little phasy, this sounds most like the original source, this has the best sense of spatialisation.

At the end of the day, the current crop of software seems to be a balancing act of timbre vs. spatial resolution. The area in which there were conflicting views regarded around the sense of centre image and top to bottom panning. This makes sense, as we are all technically listening to something slightly different due the size and shape of our heads and ears. This is something that will eventually have to be overcome if we are going to inspire a sense of trust in Spatial Audio. So, personal HRTFs (Head Related Transfer Functions) will need to become the norm, hopefully some sort of Machine Learning will sort this one out for us. Hooray for tech!