ORTF-3D Ambisonics vs ORTF-3D SPS: Comparison with Sam Hocking's Test
This is my response to Sam Hocking's "ORTF-3D SPS 8 vs ORTF-3D Ambisonics" listening test, published here: • ORTF-3D SPS 8 vs ORTF-3D Ambisonics
In our labs, we used the same signal routing in my ambisonic pipeline, using our production-ready ambisonic pipeline from SoundFellas studios.
The switcher tracks are:
1. My render.
2. Sam Hocking's render.
3. The original ORTF-3D Binaural Prerender that was provided by Boom Library on their demo recordings and used here as the experiment's control condition.
Switching between 3-1 (Ambisonics pipeline) sounds better than switching between 2-1 (SPS pipeline).
On my extended test, not shown in this video, the SoundFellas ambisonics pipeline offered a natural wide image without loss, unlike the IEM cube decoder in Sam's video. I believe that we need to test the IEM cube decoder using different energy models and perhaps add some imaginary speakers to guide the energy distribution better. Perhaps some people from the IEM team can help us set the decoder better for a more accurate cube configuration.
The sound files used to conduct this listening test are found in this blog post from Boom Library https://www.boomlibrary.com/blog/why-..., which provide some excellent ORTF-3D recordings on their sound libraries.
I use them here only for the sake of educating the public. If this use is not considered fair, I will be happy to remove this video.