Well Isn't That Spacial

TL;dr. My take on this whole spacial thing after doing it for for a while. A long while; Not about specific brands; About specific methods or general technologies; Is it “spatial or spacial?”; I hate the term “immersive”; Example of a specific use case you could deploy at most shows; Death to dual channel mono; Thoughts on hardware lock in and proprietary all in one solutions; Random shot of VRAS client for no apparent reason; Mild snark toward demos; Who to see at Infocomm to hear some of these boxes.

This isn’t so much a comparison of gear as it is an overview of techniques with some musing in general about spacial audio. For the grammar police either spelling, spacial or spatial is correct according to some old Brits in Oxford. I used spacial because it fits Enid’s tagline. You can get the info needed to configure a specific piece of hardware or get a more in-depth overview of a particular system from the people selling the boxes. There’s an assortment of quality, capable devices available. It’s my take on where we are right now from the perspective of mixing these shows.

“Immersive” is the latest buzzword in pro audio and isn’t exactly an apt description of some of these multi-channel or dimensional designs. I try not to use the term immersive when talking about these types of designs. I’m staunch and ardent about not using the term. For many of these designs using the dictionary definition of the word immersive is a misnomer at best misleading at worst. Who came up with that term? Could it be SATAN?

I prefer to call them spacial designs as the audio is spread throughout the performance space in one way or another. Immersive suggests that one is immersed or enveloped in a sound field for an indeterminate amount of time. That works well in a gaming aspect or a movie theater but may not work for the butt rock kings of 1987. There are some enveloping designs but more often than not it’s a combination of soundscapes. When I first landed on the block some used the term soundscape an many others referred to that part of the design by the name of the tool we were using. We’d say “it’s a Spacemap” or “we Spacemapped it”. It’s giving the art a depth and dimension that’s not possible in a traditional left/right concert system design. Or as I often call it dual channel mono.

Regardless of what some marketing types say there aren’t a lot of people that have worked on larger designs or deployments. Most are from themed entertainment, theme parks, theater or a combination of the above. In Vegas the largest concentration of these types are now at Cirque du Solotech having been paroled from circus of the sun. There aren’t a lot that have mixed in this way live outside of those folks mentioned above for a number of reasons. Cost is certainly one as is the learning curve. The most significant barrier is if one has an appropriate show or content. Until right around the pandemic and post pandemic you didn’t find these designs often or at all for pop concerts. That makes the pool of talent in tourville pretty shallow. Can’t learn what you aren’t doing. The times they are a changin’.

Traditionally concerts haven’t had defacto sound designers like a production show or theatrical production. The initial designs came from sound designers. Concerts have relied on working with the act based on how the song was recorded or how now the artist would like that song presented. There are more aspects of what a sound designer might do being incorporated into modern mixes and concerts. I see this particularly in large scale residencies where the consistency lends itself to the application. Think Adele or Katy Perry. It’s a lot easier to do once a rig and show is installed and programmed at the Gigantor Theater for a few weeks a few times a year than it is traveling from enormo-dome to enormo-dome overnight.

One method of providing a spacial mix that may be appropriate for most acts is a flat panorama format. This is a method currently in use for concerts and festivals. By flat I mean it happens across a single vertical plane across the front of the stage. Instead of traditional left and right clumps of PA on each side the PA is separated into smaller arrays that span the width of the stage or performance area. Using a renderer the system is combined into a larger area that acts as a single radiating zone across the speakers. The result is a wider area for panning separation and vertical placement which results in more space and clarity in the mix. Assuming of course the mix is up to the task. That’s the art thing. It’s also possible to introduce a field of depth using varying degrees of ambient processing derived from inside the processor.

This mixing technique can also employed using wave field synthesis. WFS is a technique where a sound field is created using many smaller drivers to synthesize a wave front. It’s a sophisticated form of beam steering. In this implementation the audio itself isn’t rendered rather the sound field or beams are generated by the array. You then add your audio to each beam(s). It’s a different approach to what is more or less a horizontal and vertical soundscape. There are only a few concert systems deployed at this point with this tech.

In the evolution of spacial audio I see the technology being in the third or even fourth generation of tools. Not new as we’ve been using it for 20 plus years but not quite fully mature. It’s only been within the last 5 years or so there have been more advances. The advancements in toolsets have been pretty impressive over the last few years. The biggest advancements have been tools that allow for an aural representation of the design by way of a binaural bounce to headphones. It’s not foolproof as variables like room config and acoustics or real world speaker interaction and coverage still impact the design. It’s a metric assload better than showing up in a big room with a big rig and and having the ideas not fully fleshed out particularly in front of the artist or client. Big party foul there. The tools range from plugins similar to how one would render a surround mix to comprehensive tools included in some of the other proprietary solutions.

The largest manufacturers are pushing solutions that are baked into their respective product lines. That leads to product lock in. While it’s conceivable one could design a large system using enclosures from one brand and the proprietary processor of another I’m not sure how effective support would be or if one would have full range of features using another drive system and enclosure. Or even if some of those manufactuers would be amenable to that option. There is a big install completing right now that is using the spacial design hardware of one manufacturer with the enclosures of another. It’s a unique case at least right now as the particular enclosure manufacturer relies on beam steering not audio processing to create the sound field. In this particular case the manufacturer doesn’t offer a spacial audio processor.

It takes some computing horsepower to render spacial solutions. Except of a couple of tools it’s all on proprietary hardware platforms. That’s a big cost driver. The plus side is the manufacturer has a base config they know the solution will run on so they aren’t troubleshooting every computer and their dog. That got me thinking not too long ago about how modern computers particularly those with robust graphics processing unit cores may be a better platform than a proprietary hardware solution.

The issue with processing audio with a GPU is that video is processed in parallel and audio is processed sequentially. A GPU is designed to process parallel threads. Processing audio sequentially with a GPU may slow the process compared to using a CPU. There is a startup working with NVIDIA that has a process scheduler that ques data to be processed in parallel then assembled in the correct order on the other side. It would be great for those of us that aren’t using all or that use hardly any of the GPU to help process audio. Think about it. A 60 core GPU that gets used for audio as well and is closer to $5k price point than $10k price point.

There are hurdles to using GPU for audio processing but I remain optimistic. As a manufacturer you could still only support one computer or type of computer and spec a base config. That would get them out of the boutique hardware business (at least as far as spacial audio processors are concerned) and allow them to focus on software and usability. It will also make the price point more accessible to those that aren’t like us in the 8 to 9 figure range of shows. Most of the market is priced out of the tool. Some box floggers proclaim everyone will want it but not when the boxes are tied to a specific system and are well into the 5 figure range for the processor alone. We’ve got a ways to go before widespread adaptation. If it ever happens at all but I remain positive. I’m positive the industry will find a way to screw this up.

Using commodity hardware like a Mac you’re trading cost and availability for fault tolerance. No dual PSUs, no internal auto failover. You could run two in parallel with a sensing switcher but then the cost advantage to dedicated hardware starts to starts to disappear. If you can tolerate a less streamlined failover commodity hardware could be the answer. Likely though the answer is something that’s not out yet or even a gleam in someone’s eye.

I’m largely avoiding the comparison of two of the methods of rendering these designs, channel based and object based. There are advantages to both, disadvantages to both. Right now the overwhelming bulk of spacial designs for live entertainment are presented on older channel based systems. Channel based designs made 20 years ago still hold their artistic value today. The tech is changing and likely some form of object based or hybrid designs will become the norm sooner rather than later. Of the designs out in the last year or so for live audio object based designs are taking hold. More likely it will be something that’s not identical to the tools used today. The more it’s used and the more we learn the better it will get. We can only hope the software interfaces will have more in common than not though I’m not holding my breath.

With objects come an ease of use and portability to like environments. An example would be a tour that plays a different hoop barn every night. Most modern NBA palaces are similar enough you could use the same general config. Perhaps a front with 5 zones, some sides and maybe some overheads and/or rears. With an object based system you have an easier flexibility to change the design to better suit each particular room. You can do it with a channel based system as well though depending on the design it may be a bit more laborious. Don’t get bogged down in the tech. Use those synapses for the art. The trick is to learn to use whatever you render on/in to the fullest or have a good tech to help. You can do that by leveraging any hands on manufacturer training you can get. Be true to your art. The solutions will follow.

If you’re going to Infocomm this year you’ll be able to kick the tires or at least see a dog and pony show of the available technologies. L-Acoustics is there with L-ISA and plenty of buzzwords in tow. d&b is there with Soundscape. Meyer is there with D-Mitri/Spacemap/Constellation. Holoplot is there and even if you aren’t into the spacial thing check it out. It’s a lot different than anything on the market today. Fulcrum just announced a dive into spacial audio. I’ve seen/heard all of those but the Fulcrum device and knowing a couple of those blokes from the old country I’m curious to hear what they have and see the interface. Look in the show directory for the demo rooms. Pro Tip: In previous years Meyer had comfy chairs and snacks.

If you’re getting into this sort of sound design or are curious stop by and check it out. I’ve seen demos where someone blows enough smoke up your ass it can chap your cheeks. Don’t let someone bullshit you into thinking your art can’t be done without a particular platform. I’d reckon the people presenting these demos while presenting the product in the best possible light (because that’s what they’re paid to do) will show more steak than sizzle and give you enough info to make an informed decision. Listen and judge for yourself. Just don’t tell them I sent you.