Yes, there are certainly details that can’t be emulated outside of binaural recording, but the argument that the plane cannot extend beyond the screen without it feels overly assertive to me. I also feel the issues with HFR 3D has little to do with the sound presentation. It’s a fundamental shift in composition of the visuals, and the process of cinematography needs time to adapt and learn to use it to best effect.

Hi Shaun thanks for the response,

I’ve done extensive work with binaural and I disagree that true sonic depth can be created with conventional and existing cinema sound methods. You certainly can create a facsimile of depth but there’s a simple physics problem that you can’t overcome: a sound that is further away from the screen plane interacts with everything else that’s further away from the screen plane to tell your brain where it sits in space. This is extremely easy to demonstrate (I’m surprised that you say you have yet to hear an example of it, in fact): set up a camera on a quiet path among some buildings and get someone wearing hard shoes to walk away from you. With a binaural recording you can plainly hear the person walking off into the distance. They _sound_ distant. Now attempt to emulate that effect with a mono recording, volume control and any verbs and delays you choose (since this is what you’d do in a film mix). You’ll never match it. You might get something that sounds acceptable – in fact, in this day & age I should hope you could* – but the binaural recording preserves complex elements of the real environment that your ears and brain can very accurately place – and not just laterally, but vertically as well. This situation becomes acute when you’re dealing with multiple sounds. Right now, as I sit writing, I can hear a very complex layering of sound outside – birds in a tree close, a more distant bird calling, a dog barking, traffic and the local highschool PA playing music. All these sounds are clearly layered in their appropriate distance. If I close my eyes I can tell you exactly where they are in respect to one another in terms of distance. No-one, no matter how clever, could emulate this complexity to simulate the reality sufficiently. But a single, simple binaural recording would capture it all.

You quite correctly point out the wealth of problems that binaural presents, and I’m not suggesting that it would be a desirable or even plausible solution for a cinema environment, only that it points out the failing of conventional sound models to provide accurate representation for sound beyond the screen plane (and yes, I am aware that the speakers sit behind the screen – a full 3 feet of awesome extra depth if you’re lucky :p)

If you get a chance, try and catch a screening of The Hobbit in HFR and 3D (I guess this might be tricky now, until the next one comes out). It vividly demonstrates the issue. The mix on this film is excellent – there’s nothing to fault as a standard 2D cinema mix. I’ve watched it on Blu-ray in my studio and it’s a great job from all concerned. In HFR 3D, however, the sound does a weird thing – it becomes flat and featureless. It’s like the detail is sucked out. Complex scenes feel like washes of sound with effects poking out in a very ugly way. It took me three viewings to figure out why – most of the visual depth is beyond the screen plane, and the sound doesn’t stick with it. It’s acceptable in 2D because we’re used to the convention, but as 3D image approaches very realistic levels, your ears and brain are trying to deal with a cognitive dissonance.

(*I’ll make a slight caveat here: convolution verbs will be hugely helpful, I accept. The thing is, a stereo convolution verb is actually providing aspects of binaurality, due to the nature of the convolution recording process).

You raise some interesting points, Peter, but I’ll play devil’s advocate here. I’ll grant you that this is somewhat dependent on listening position, but I have yet to hear a binaural example of distance in the frontal plane that cannot be recreated with a more “conventional” playback systems. If you really want to get semantic in the discussion, all theatrical sound is emitted from behind the screen to begin with (…though, obviously, not exactly your point). There’s also the issue that effective binaural playback is highly dependent on how well the HRTFs calculations used align with the individual listener. I myself have listened to binaural presentations that work perfectly for others, while presenting issues with consistency of image for me. From my perspective, creating depth in non-binaural systems is not terribly difficult.

I wrote a little bit about this on my sound blog Hummadruz:

https://hummadruz.wordpress.com/2013/01/14/the-sound-of-distant-drumming/