In the late 60’s the struggling Capitol Records took advantage of the burgeoning home audio market by releasing its back catalog in stereo for the first time. Represses of The Beatles and The Beach Boys LPs flooded record shops, juicing dormant sales- and boosting adoption of home-use stereophonic units. What a revelation it must have been to hear Meet The Beatles, in full stereo glory, vocals on the left and drums on the right, as God intended.
But those old Capitol records: they weren’t true stereo. They were Duophonic. This pseudo-scientific phrase was spun up somewhere in the bowels of the Capitol building. Unlike real stereo, which involves mixing multiple recordings across a two dimensional “stage”, Duophonic mixes use all manner of electronic trickery to fake stereo. Mastering engineers took the original mono recording and split it in two, panning all the high frequency sounds to the right, and the low sounds to the left. To mask these edits they then drenched the whole recording in reverb and delay. Listening to a Duophonic album is like looking at a painting smeared with vaseline.
Capitol’s competitors followed up with their own branding: RCA featured “electronically reprocessed stereo”, and Columbia’s releases were “re-channelled” audio. (So pervasive were these fake stereo mixes that record companies have been known to accidentally re-release new CD and Vinyl remasters sourced from Duophonic editions.)
By the late 70’s, studios had largely adopted stereo recording and mixing technology, and releases were justified in emblazoning “STEREO” across their headers. Duophonic records fell into the past, an oddity only remembered by music engineering fanatics.
But records didn’t stick around; stereo did. Decades passed. CDs shunted vinyl into a fetish market. Streaming annihilated its progenitor, digital sales. Ownership went the way of the dodo. Consumers are satisfied- nay, thrilled- to rent their music libraries. My dad still has binders of CDs, but they’re in our attic, shiny little plastic disks destined to last a thousand years. (My dad streams Meet The Beatles now.)
Spotify was first on the scene, but once they posted a profitable quarter their business model was copied six ways to San Francisco. Bay Area giants and startups alike built competing platforms, but by the mid 2010’s they were nearly identical in catalog and pricing. Cough up ten bucks a month to any one of a myriad of services, and you have access to nearly everything ever recorded. This amateur economist would term this market saturation.
What’s to be done? The Apple Musics and the Spotifys and the Tidals and the Deezers and the Amazon Musics and the Rhapsodies and the Pandoras? For some, the answer was bankruptcy or acquisition. But the big boys went to work, ginning up celebrity endorsements and super bowl ads left and right, but to no avail. But the music market of today has a lot more in common with 1961 than might be immediately obvious. Enter Spatial Audio.
You know when you see a movie in a theater and the gunshots whip past you and helicopters roar overhead? That’s thanks to surround sound, which traditionally works by placing 5-9 sound sources (which may be just one speaker or a cluster) around the viewer. With it filmmakers can place you inside a scene and dynamically adjust the location of sounds. Spatial Audio is surround sound for music. Listening to a spatial album, the marketing copy claims, is like being in the studio with the band while they recorded it, or standing at the barricades at an incredible concert. Nowadays, the best theater surround format is Dolby Atmos, which is capable of positioning a sound in any of a thousand places with only a handful of speakers. In fact, it’s this same Atmos technology shrunk down to palm-size that powers Spatial Audio.
Now, there’s a lot of competitors in this space. Amazon, Sony, etc. They originally all had different names for their unique, proprietary version of spatial (Amazon’s was something silly like 3D HD Audio Ultra Plus Max Pro), but they’ve all adopted the Atmos format, either under that title or their own brand names.
The clear forerunner in this space is Apple- they’ve made the biggest fuss, but they also have built Atmos into their massively popular Airpods and made it the default audio engine in their operating systems. For the purposes of this article, we’ll examine their platform, Apple Music.
To fully take advantage of the Atmos format, the original stems (individual recordings of instruments and sounds) for a record have to be manually reorganized by mixing engineers. This ranges from simply opening up the project file to manipulating fifty year old magnetic tape recordings, depending on how old the record in question is. The tracks have to be imported into a digital audio workstation (Apple would prefer you use theirs, Logic Pro) and then moved around a 3D space and rendered. (To put it simply). This is a full remix/remaster of the original material.
To make a Duophonic record, engineers actually had to retrieve the original mono master tapes from the label archives and go to work. But the magic of AI allows for original stereo releases to be “converted” to spatial. As opposed to the professionally remastered releases from above, any track- any audio recording, actually- can go through this filter and come out the other side in three dimensions. In practice, this blatantly anti-artist-as-arbiter-of-their-work function (something like f(x) = x * tons of digital delay multiplied by sneaky EQ’ing, where x is whatever song, and sorry to be cute about it like this), comes out just messy. For the gear-heads out there, it’s like if you ran a track through whatever reverb pedal and out your amp. It sounds artificial, off-kilter. Sometimes it smothers the vocal or pans the drums way out or separates different halves of the same instrumental. This is not a remaster- it’s more like a filter.
A lot of the spatial audio fervor is just marketing hubbub. These companies, struggling to distinguish themselves in a crowded market, are again resorting to the same tired tricks and gimmicks that Capitol & Co. used in the 60’s- renting the same album back to you with a fancy new marketing-approved pseudo-technologic improvement, except now it’s the distributors doing most of the work, not the labels- a fascinating shift in the music economy.
But I won’t let my corporate cynicism get the best of me- what if this really is a new way to experience music? I A/B’d more than two hundred spatial tracks and their stereo counterparts. These pieces ranged from the 60’s to the present, from dozens of genres- they only to understand what Spatial Audio really accomplishes. The results varied wildly. There’s really only one generalization I feel comfortable making, which is that the newer the release, the better it usually sounds in space.
(A quick note: the average loudness level of spatial mixes is quieter than their stereo equivalent, which I’m guessing comes from a need for greater dynamic range. Because volume can radically influence perception- people usually rate the same song 5db louder as better- I tried my best to ride the volume rocker between mixes.)
Most of Apple Music’s spatial catalog is basically ultra-disposable top 40 (no hate if that’s your thing). It’s not to my taste, and it’s not exactly where our best music ends up (with some obvious exceptions). Because of that, my review palette was slightly constrained, but workable, I think.
With that in mind, let’s examine the discography of all time pop great Taylor Swift. Her releases from 2018’s Lover on have spatial mixes. There’s no beating around the bush here: her electropop albums are totally marred by the remixing. Even in Midnights, which theoretically should have been mixed in stereo and spatial at the same time, there’s a huge quality discrepancy between versions. Take “Bejeweled”. The instrumental is constructed from three simple elements- a basic kick-snare drum beat, a default-preset sounding synth bass, and those wonderful arpeggiated synth sparklies in the chorus. In stereo, this combination works. The sparsity of the arrangement lends itself to a carefully balanced mix; perfectly EQ’d, shiny; you can hear every element clearly. It’s economical and efficient.
The spatial mix is a disaster. The only way I can describe it is that the bass sounds like fog. It’s huge, hollow, seeps into every corner of the soundstage. This may sound neat, but what it means is every other instrument is muffled by those low frequencies. Those wonderful high end sparkles in the chorus- which gives the song its sonic character and hook- are almost entirely absent in the spatial version. It’s not unlistenable by any means, but what worked well in the original is gone. “The Man” and “Karma” suffer the same fate.
But credit where credit’s due: folklore in spatial is much improved. The original folklore instrumentals were a highly delicate balance between clean electric guitars and basses, pianos, electronic and acoustic drums, organs, and synthesizer pads. But because it’s a T. Swift album, her vocals were front and center, and overly loud compared to the instrumentation. In the spatial mix, every tender detail of the record is open to the listener. This is not achieved just by turning Swift’s voice down, but by careful positioning of instruments across the soundstage. The shimmering guitar loop that opens “mirrorball”, in stereo nearly inaudible, now consumes you, swirls around you, before Swift devours you. Some tracks take this a step further: “august” actually fades Swift’s voice out in the second chorus (anathema to pop music!) and allows the song to breathe. Spatial Audio allowed me to reevaluate folklore; from an okay album to maybe a great one.
Other recent spatial releases rival or match their stereo versions: “Spitting Off the Edge of the World” from the new Yeah Yeah Yeahs album envelops you in its lush keyboard layers; the drums in The Weeknd’s “Blinding Lights” punch right through your stomach. Take any track from Nas’ new King’s Disease record and its spatial mix is tight, coiled, potent.
In fact, every hip hop track I sampled sounded as good as, or even better, than the stereo version. Most of these songs avoid any cheesy spatial tricks; their instruments and samples reside more or less in the same section of the soundstage. But the clarity is dialed up to ten. Listening to both Outkast’s “Ms. Jackson” and Biggie’s “Hypnotize”, I felt like I had just upgraded my headphone setup to the tune of several thousand dollars.
When it comes to remastering older material (besides the above), however, spatial fails nearly a hundred percent of the time.
Weezer classic “Buddy Holly”’s drums and guitars are hard-panned left and right- in, I think, conscious homage to those early stereo Beatles records- but it robs the guitars of their weight and heft. They do recenter for the chorus, but it’s all the more jarring when they splinter back out.
(Speaking of the Beatles, their ‘canonical’ albums have all received spatial make-overs, that while largely unoffensive, are also completely unnecessary.)
But, by far the most egregious: Spatial Audio completely destroys Talking Heads’ Remain in Light, one of the finest albums of the eighties. Opener “Born Under Punches” is competent: the kick drum comes from way down deep in your chest; the skittering polyrhythmic guitar licks whirl around you. But it eviscerates “The Great Curve”, a song so exquisite Karl Ove Knausgaard wrote about it in the fourth volume of My Struggle- “And, toward the end, on top of all this, like a bloody fighter plane above a tiny dancing village, comes Adrian Belew’s overriding guitar, and oh, oh God, I am dancing and happiness fills me to my fingertips and I only wish it could last, that the solo would go on and on, the plane would never land, the sun would never go down, life would never end.” It’s that type of song.
In it, Robert Palmer absolutely slams at the bongos for six minutes straight, and it’s from here the track derives its bounce. In Spatial Audio, the percussion is rendered thin, impotent, and lifeless.The groove is pulverized. Because Remain in Light was recorded on tape, I’m guessing there was some ‘creative’ eq necessary to separate out the instrumentation. I don’t envy the engineers; I can only imagine the amount of knob-twiddling necessary in remixing the album. But if the result is so utterly bloodless, why do it at all?
This is my main concern: the tension that underlies all this technological fetishism and needless fiddling with cultural documents of the past. The reason for reworking these albums isn’t because there were great flaws with their stereo (or mono) mixes. It’s money. Were a great band to intentionally record an album making innovative use of spatial audio- here I recall The Flaming Lip’s Zaireeka, an album requiring you to play four separate CDs on four separate systems- I would applaud it. But no such work has arrived, and I don’t think it will. It’s not that the technology isn’t ready- surround sound has existed for decades- but it’s the dynamics and economics of the industry that prevent this. It’s not labels pushing spatial, and it’s certainly not artists- it’s tech giants in CA trying to differentiate their services and make money off of you. It’s that simple, and yes, that cynical.
The labels of yesteryear cultivated close working relationships with their talents; the adoption of true stereo was the rare union between the commercial and artistic, and it was only possible in that symbiotic economy. Now, the music money rests with companies firmly outside the entertainment sphere, and spatial audio is something they foist onto recording performers at the expense of their work.
But I’m no better. So many paragraphs ago, I gushed about tiny reworks to folklore, an album that needed no tinkering (it won Album of the Year at the Grammys and enjoys a reputation even among the anti-pop crowd). I too can be suckered by slick marketing and minor tricks; Apple still gets ten bucks from me every month. And I listened to folklore by streaming it- renting it- instead of buying a record or a CD. Apple can make money from me in perpetuity.
And money in perpetuity is the ultimate white whale.


Leave a Reply