In this manner, all sound editing and mixing could be done without any elaborate object-based encoding and rendering. You would have a format that would “immerse” the audience to a degree as to be indistinguishable from Dolby Atmos or DTS MDA for 98 percent of the running time of most films.
<!--end paragraph--><!--begin paragraph-->
I think that the missing two percent are not just an acceptable compromise, they’re a desired one. What object-based mixing does uniquely—move sounds either down the length of the theater or to specific locations away from the screen—is precisely what I don’t like. In the mid-1990s, some mixers took a few movies to get out of their system, putting silverware Foley or snare drums in the surrounds, and the thought of point-source sounds barking at me from the ceiling or walls is almost too much to bear. In fact, even the 7.1 format, dividing the surround tracks in two, bypasses my ken.
<!--end paragraph--><!--begin paragraph-->
Now that I’ve said this, I admit that the toothpaste is out of the tube, and with hundreds of theaters worldwide putting in immersive sound systems, the question is not if immersive sound should be implemented, but how. First, let’s look at the two approaches: Auro-3D, with its emphasis on height layers, vs. object-based Atmos and MDA.
<!--end paragraph--><!--begin paragraph-->
In the multiple Auro-3D demonstrations that I have attended, I have not heard anything that I consider to be a radical improvement over 5.1. The most impressive parts of their test material were in sections recorded with their custom mic arrays, with various height levels. As good as these sounded, I think that custom mic arrays highlighting the strengths of Atmos and MDA would be much more impressive. Besides, as film sound history has shown us—such as with CinemasScope’s three-track production recordings—literal reality is often not desired or practical.
