Peter’s excellent blog is something I check regularly. His journey on the road of ambisonic discovery and a PhD shares many correlations to mine, both in the fascinations of the subject and the manner in which it (ambisonics) is and can be used.
Initial experience with 3rd Order ambisonics allows for enhanced spatial mixes allowed Peter to think on what to use ambisonics for, and how best to present his own audio recordings with the physical placement opportunities available to those using spatial audio.
To answer, the first point to note is that we’re not talking about the production of electro-acoustic music. It’s not art based noise pieces or synthesis for limited audiences or unique performances. The discussion is about commercially made music for the masses available in a viable surround format. Something that any record engineer understands the process of, and any consumer understands the business model. Pop(ular) music. It’s the sort of thing that Tchad tried to use for Pearl Jam’s Binaural but didn’t quite make it.
Delving into the past details the traditional music production paradigms stemming form from archaic, unscientific, product led, and are almost meme theory-based in their propagation throughout the last 70 years. Monophonic recordings were heralded as unsurpassable, stereo was deemed ‘too real’, surround was an aural indulgence and anything beyond is unfathomable.
So – why are we adding more channels? And to that; what’s the point of using every new technology that comes along?
Well, to answer that – my argument has always been that the recorded medium of sound has been part of the documentation of time. With respect to music this translates directly to the idea that live music acts could not perform everywhere in the 1950s but their popularity grew exponentially – and therefore a recording could allow people to buy into the experience. Although the dates and some facts smear the hard line opinion; live music came first. A recording is then simply a translation of a group on a stage with the placements a relative levels set to how their work should be heard.
Popular records that transcend the listener often make it into the psyche of a generation and therefore the methods used to achieve that sound become the new emperors clothes. Trigging samples, auto-tune, layering guitars, lead vocal harmonies, compression and even panning have all came from one record or another. The people who understand how techniques flow through the field of production best are often DJs. Every single one DJ where the amen break came from.
So, knowing the conventions of music production has been fairly stable over the last 70 years. Techniques and available tools have changed – but they’re all built on years of repetitions by professionals. Everyone knows how to mix onto two speakers. It’s easy, as your favourite record was mixed for two speakers. This theory of meme-based knowledge and phases of production techniques becomes problematic when we start to add more than 2 channels into the mix. Not many of us have been to a surround gig, or experienced a piece of popular music like Björk’s 2006 Surrounded where an artist dares to break convention.
Placement of instruments within the ever expanding virtual space, however, as Peter points out is an interesting point of discussion. Additional channels and the recordings produced with those channels has not results in divergence from where instruments were placed in the 1960s. The stage of sound has changed little. Art-noise has embraced the technical possibilities, as any student of Cage and Stockhausen should. Classical recordings have moved the reverberation of the room to the rear. Pop music hasn’t accepted surround. The last one is something that motivated me to investigate further into the mirky world of traditionalist views on surround.
Peter points out that 5.1 mixes commonly use the rear speakers as classical music would, with ambience. Perhaps audience noise if the recording is live. Spatial audio formats, however, offer more than a de-correlated positioning.
Many spatial audio formats hinge on the use of regular, or equidistant, loudspeakers arranged around the listener(s). This is a stark contrast to the ‘irregular’ ITU-R BS.775 recommendation where the speakers are at various angles and distances from the listener(s).
Having all directions being equal and the speaker placement allowing for sound reproduction equality – the use of the equidistance within ambisonics is key to true stereophonic sound for the listener, and Peter’s original question is well put; just because we can, does it mean we should?
Knowing popular music and the never-changing sound stage presentation helps to answer the question slightly. Blurring the lines between what we expect in traditionalist presentations and what we can do in noise pieces would give the answer that yes, break the rules. Give the audience a hook, say a vocal, and allow them to focus on that whilst submerging them with sound.
Defining the lines between noise music and pop would suggest that no, the traditionalist approach would not give in to the temptations of allowing the listener to be a part of the band – but they are watching the band. It’s an experience of music in their own home but through a controlled window. Peter’s agreeable idea that the sound stage theory should be applied stylistically, with adherence to the context. An example of this would be a live recording could be remixed to include the listener as part of the band onstage. Live concerts using quadraphonic and ambisonics have been known in the past and to great acclaim. Much of the great work has been done by Funktion One at festivals using dance music.
Personally I think that all music should be widened a little bit. The idea of left and right is only applicable to technologies using headphones (or ambiophonics) where acoustical merging of signals would only smear the image. Broadening the boundaries of music allows for greater immersion, understanding and enjoyment of the piece. The qualities that many would argue are so crucial to those in electro-acoustic and noise-genres are in fact ones that should be applied to more modernistic methods of production.
Controlling the scalability and translation between listening systems is perhaps one of the last fundamental bastions of the mastering engineers. The mixing engineer, traditionally, should pay attention but not limit themselves to Peebles’ tube radio. Aim it for the highest quality and it will filter down. Crucially though, with surround it falls to the mixing engineer to ensure that translation between a spatial audio and stereophonic system is at its highest.
With that in mind; I pay homage to my production forebears and say ‘go for the highest quality’.
Reducing channels is acceptable to many causal listeners and the quality of spatial impression is still increasingly high when in direct comparison with ‘traditional stereo’. A new generation of MP3 addicts are looming – so perhaps all of this technology will not even be noticed unless it’s on a small 4″ screen with an RMS of -0.3dBFS.
My personal approach to mixing in surround is placement of audio objects is always relative to the hook. Translation of the message is key. Placing a vocal front and centre is where it should go if the message is carried there. The listener shouldn’t have to delve into the mix for the purpose of the music. As for the instrumental, any place is good. Making things interesting keeps listeners returning to music and providing some substance in a world full of auto-tune, quantisation and digital limiting. Every project I’ve worked on within the last 3 years has been placed into an ambisonic sound field. The tools are there, and I use them. I know that the band will never ask for the mixes. I know that if they did, the listener base would be small. And even then, I’d have to render for an ITU-R layout rather than the regular one it was mixed on.
So, I arrive at the end of my blog post on positioning of sound in space with a thought:
If art is a reflection of life, shouldn’t we just make it happen and see what happens? The evolution of sound, perception and meaning should be explored as much as possible.