David Bowie’s classic Ziggy Stardust album has been reinvented in three dimensions.

On July 6th, 1972, an unearthly figure in a blue and gold quilted onesie made music history. David Bowie’s Top Of The Pops performance of ‘Starman’ influenced countless future hitmakers, taking glam rock in a strange and thrilling new direction. The song and its parent album The Rise And Fall Of Ziggy Stardust And The Spiders From Mars established Bowie as a superstar, and would form the cornerstones of his incredible musical legacy.

Fifty‑two years on, that legacy is being celebrated with a box set of outtakes and rarities entitled Rock & Roll Star, and an entirely new Atmos mix of the Ziggy album. Both were overseen by original co‑producer Ken Scott, who worked with Emre Ramazanoglu to carry out the immersive remix from the original multitracks.

Star Quality

Back in 1972, Ken Scott was an ambitious house engineer at London’s Trident Studios who was looking to spread his wings. “I had done two albums with David as an engineer, with Tony Visconti producing, and David and Tony split ways. And I was working with David, who was taking some time off after having very little success with the two TV albums. He came in to produce a friend of his, a gentleman by the name of Freddie Beretti, who finished up being a clothes designer for David.

“Freddie wasn’t a great singer to say the least, but it put David and I back together again and it was at a time when, as an engineer, I was getting very bored with what I was doing. Everything seemed the same, and I wanted more artistic say. So, during one of the breaks in the sessions, I happened to mention this to David, and he said ‘I’ve signed a new management deal. They want to put me into the studio to record an album whilst they shop a record deal. I was gonna produce it myself, but don’t know if I’m capable of doing it. Will you co‑produce with me?’ An instant ‘Yes!’ and that was the start of a four‑album relationship.”

Ken Scott: “David was brilliant at putting together great teams. For the Ziggy period, that was Trevor Boulder, Woody Woodmansey, Mick Ronson, myself and then a pianist."

Co‑producing, as far as Bowie was concerned, meant joining a core group of creative people who worked together. “David was brilliant at putting together great teams. For the Ziggy period, that was Trevor Boulder, Woody Woodmansey, Mick Ronson, myself and then a pianist. That changed on three of the albums, but we all had our input, and then the final thing for me was David’s trust, because I think of the four albums that I did with him, he came along to only two mixes. The rest was just left up to me. Whatever I chose to do in the mixes was fine, I guess, because once the album was completed, we never spoke about it again!

“Some of the songs had been routined, but not all of them. Quite often he would just teach the band the song in the studio, and then they got to work out what they were going to play themselves. I’ve heard that from various people, like Carlos Alomar and Mike Garson. They all say the same. When he put a team together, he picked them because he knew what they would bring to it. And he wanted them to do that. He didn’t want anyone to do it the way he wanted it, if you get it — it was the way they would normally work. He didn’t want to direct them that much.”

Tracking

Ziggy Stardust was recorded at Trident, with Ken Scott engineering. A single 16‑track machine proved ample for capturing everything. “We made decisions, then, as to what things were going to sound like!” Scott laughs.

Tracking sessions usually involved the rhythm section being laid down live, perhaps with one or two additional parts and a guide vocal. “Bass and drums would always be live, with possible punch-ins for the bass. There might be an electric guitar, acoustic guitar or acoustic piano along with them, then overdubs from there.”

Woody Woodmansey’s drums were mixed live either to two or three tape tracks. “I think some of it was stereo and some of it was three tracks; the kit would be stereo and the kick would be on its own track. I think more often than not the bass was DI, but at times I’m sure I used the amp as well, mixed in with it.”

For Mick Ronson’s guitar, meanwhile, Scott says he used “One Neumann U67, about that far away [he indicates a distance of a foot or so] from the cabinet, and we got his sound. With Mick it was always the same. He played his Les Paul through a Cry Baby wah‑wah pedal into his 100 Watt Marshall, and the way we’d get the sound is that he’d start the wah‑wah pedal at the bottom and slowly bring it up until we said ‘Stop!’ And that was the sound we’d use for that particular section, or the whole song maybe.”

Bowie himself, meanwhile, was an engineer’s dream. “With regard to his vocals, on the four albums I did with him, I think probably 95 percent of the vocals were one take, first take, beginning to end. I would get the level, the sound, hit record on the tape machine, and what he sang that one time through is what you still hear today. He was amazing. I think I probably used a little compression; I can’t remember what we had back then, I’d need to look at pictures to remind me what we had, but probably an LA‑2 or something like that, and the mic would have been either a Neumann U67 or an AKG C12A. He was a one off!”

Ken Scott: “With regard to [David's] vocals, on the four albums I did with him, I think probably 95 percent of the vocals were one take, first take, beginning to end. I would get the level, the sound, hit record on the tape machine, and what he sang that one time through is what you still hear today... the mic would have been either a Neumann U67 or an AKG C12A. He was a one off!”

There were also orchestral overdubs, which at that time had to be recorded in a less than ideal way because of the preferences of the musicians. “The orchestras refused to wear headphones back then, so we had to feed the track through large speakers. If it was loud enough for them to hear it, it was picked up by the mics. I remember when I was doing the 5.1 remix, I hadn’t listened to the multitracks or anything for 30‑odd years, and then the first time I brought up the orchestra faders I suddenly heard this track coming out. I thought ‘What the hell is going on? What’s this?’ And suddenly I realised ‘That’s coming through the speakers!’ I had completely forgotten about that. I got so used to them wearing headphones. But yeah, that’s something that we used to learn, the whole thing of blending things like that so that it doesn’t appear as if there’s too much spill coming in when you start to cut live. You just learned to deal with it back then. Now, you get everyone having to be in their own little booths, you can’t have anything crossing over because who knows what we might want to do with it later!”

From Surround To Immersive

Ken Scott carried out the 5.1 remix of Ziggy Stardust in 2003 and, in retrospect, sees it as something of a missed opportunity. “I said to the label, ‘OK, there are two ways I can do this. I can modernise it. Or I can just do that as it was originally, just putting things in different places. Which would you prefer?’ And they said ‘We’d love both versions, but we can only afford the one version — which has to be the same as the original.’”

Two decades later, the call came to revisit the album for the next generation of surround formats and, specifically, Dolby Atmos. “I insisted going into this that it had to be a whole different thing. It was going to be a new album, basically. Same music, different album. I believe that if the listener is in a different environment, you should try and make it that they’re in a different environment.

Ken Scott: “I insisted going into this that it had to be a whole different thing. It was going to be a new album, basically. Same music, different album."

“There’s also the whole thing where Apple have to have the stereo matching up exactly with the immersive audio. Well, that’s OK as long as you’re working totally digitally, but with regard to the old stuff that was done on tape, no tape machine ever runs at the same speed twice. So there is no way of getting the stereo to match up to the Atmos mix unless you go in and you do things like Elastic Audio to match them up, and then you’ll start to lose quality. So what I insisted was ‘OK, we’re going to do the immersive audio mix — but there’s also going to be a brand new stereo mix as well, which will be based around the immersive audio.’ Anything we changed on the immersive audio will be reflected in the stereo mix, so that they will match up perfectly.”

The New Team

There was never any doubt that Ken Scott had to be the person to oversee the Ziggy Atmos remix. “David insisted that if anyone was going to do any future work on his product, it had to be the people that did it originally. It had to be the people that knew what he was going for, that was part of that original team.” But there was a problem...

“When it was first brought up about two years ago that we might be doing this, I thought ‘Oh, great. I’ll start learning about it.’ But I learned very quickly there is no way that a 76‑year‑old would learn Atmos well enough to make a classic record into a classic Atmos record.

“Luckily I was introduced to Emre at an event, and a bit later I said to a friend of mine: ‘Can you put me in touch with that guy I spoke to at the awards ceremony?’ He said ‘You mean Emre?’ I said ‘Yeah. Yeah. OK.’ And we met up, we were speaking and it was amazing. It was as if Emre was part of me. We work together so well, it was so easy. It was such an enjoyable 10 days and it’s turned out to be something I am exceedingly proud of. And once again, I keep on going back to the team thing. There was the team of Emre and myself, and it worked perfectly.”

Praise indeed — especially when you consider that it’s the first time this has happened in Ken Scott’s 60‑year career. “I started off as an engineer, and then decided in 1971‑ish that I wanted to move into production. I have tried since 1971 on several occasions to work with another engineer, and it never seemed to work for me, because it would take my mind away from the music. I could always reach over and do something quicker than explaining it to the other engineer. So I have always engineered my own projects — until now!”

Precision Matching

Emre Ramazanoglu is a London‑based musician who has enjoyed varied success as a producer, songwriter, drummer, programmer and engineer. Lately, he’s become something of an Atmos specialist, with credits including the immersive remixes of Brian Eno’s classic 1970s albums, and he brings an impressive degree of rigour and technical understanding to the often confusing business of immersive mixing.

Emre Ramazanoglu: “A lot of the work I do is exactly matching the stereo remix, making a fully immersive version where the binaural fold‑down has exactly the same balance as the stereo mix."

“A lot of the work I do is exactly matching the stereo remix, making a fully immersive version where the binaural fold‑down has exactly the same balance as the stereo mix,” he explains. “Often artists aren’t involved, and I’m not gonna change the artist’s mix when they’re not in the room and the producer’s not involved and the mixer’s not involved. It’s certainly not my job to go and reimagine their mix, when they’ve spent a year getting that one approved. That’s my gig on those jobs, and it’s very clearly stated especially by UMG and Sony that that’s the job, because it’s very very difficult to do correctly. Matching the mastering from pre‑master stems is not straightforward in immersive audio.”

Immersive audio formats can be channel‑based or object‑based. [For more info, see An Introduction To Immersive Audio - SOS January 2022.] In the former case, a master recording contains a fixed number of mono channels, each of which is mapped directly onto a specific loudspeaker. In the latter, the channels within the master typically represent individual sources or groups of sources, and are packaged with metadata that allows their spatial position to be assigned on the fly during playback. Dolby Atmos is a hybrid format, containing both surround ‘bed’ tracks and mono or stereo 'objects'. Engineers who mix music in Atmos have very different opinions about the relative usefulness of these elements, and Emre Ramazanoglu insists that if you want the binaural fold‑down to sound right, you need to avoid using the beds altogether.

Dolby Atmos is a hybrid format, containing both surround ‘bed’ tracks and mono or stereo 'objects'. Engineers who mix music in Atmos have very different opinions about the relative usefulness of these elements, and Emre Ramazanoglu insists that if you want the binaural fold‑down to sound right, you need to avoid using the beds altogether.

“The bed doesn’t translate to headphones very well, and wasn’t really intended to. That was not the purpose. It’s not a really useful tool if you want any headphone translation. You need specific control of the binaural metadata per element, and precise control of objects, in order to get a deeply translating stereo companion in binaural, and also being aware of what that does with Apple Spatial. So now the only thing I use on the bed is a touch of LFE. And very little, often, because you can’t rely on LFE for an Atmos mix because several playback systems just don’t have it, or reproduce it in unexpected ways.”

Instead of using the conventional Atmos beds, Ramazanoglu creates what he calls ‘object beds’: groups of mono or stereo objects routed to specific positions within the 3D space. Individual sources or stems that don’t need to be treated as objects in their own right can then be fed to these object beds as required. “I’ll have multiple object beds depending on what binaural positions work for the intention of what I’m doing. So you end up with quite large object‑based mixes. But again, I don’t even use the object beds that much for anything positional, because then you’re lying inbetween two often different binaural positions, and if you have to hear it in headphones, it won’t be a useful translation. So, I’m much more definite and will be very specific about where things go.

“Another issue in Atmos that’s really constant is that you’ll get a heavily front‑loaded system and then small satellites or surrounds, and all the bass management’s at the front. You have to bear that in mind a little bit when you’re moving elements around. You technically will get a non‑discernible location from your low end, but it depends how high your bass management is going up, and what speakers there are. You have to think a bit carefully about where the weighting is going and how it’s going to feel, because you can go absolutely mental and it’ll maybe sound quite nice in your room, but in headphones or other rooms it could be just a real mess.

“And this is the problem you get with Atmos a lot. People say it sounds ‘watery’, and it doesn’t. It doesn’t sound any different from stereo — unless you make it sound different by not making the mix feel coherent. What’s often happening is people are mixing from pre‑master stems, and these are from a track that has been mixed with a limiter or more on throughout the mix process. This mix bus is then removed and not replaced, so you’re not actually hearing a mix that has ever been approved or is even finished!”

Future Proofing

By the time Ken Scott and Emre Ramazanoglu started work on the Atmos mix at RAK Studio 4, with its state‑of‑the‑art Genelec monitoring rig, the producer had already done a good deal of detailed preparation work. “The original Ziggy sound was very much — quote unquote — the ‘glam rock’ sound. And that was fine for the ’70s but it needs to be beefed up a bit. It needs to be modernised. So one of the things that I did was I went in and I added samples to the drums. Initially just a snare and kick, not replacing anything, just adding to the original sound, to beef it up a bit and make it more modern. As it went on, for one thing, back in the day, I hated cymbals. I now quite enjoy cymbals. So we found in doing this: ‘Hang on, there’s no cymbals. What’s going on?’ So I then had to go in and program in hi‑hats and cymbals and all of that kind of thing.”

“I was slightly concerned when Ken mentioned samples and adding stuff,” admits Ramazanoglu. “I’m very very fussy about that kind of thing. Especially with the hi‑hats, which is really not often done sample by sample — but I don’t think you’d tell they were there. It was just part of the drums. If you’d put them up, I wouldn’t have questioned it. I might not even have known.”

“It’s ridiculous!” laughs Scott. “It literally is hit by hit, making sure it’s on there. I use the Massey DRT to get it into MIDI and then trigger from there, and there are always some things that are going to be off, and you have to move it around. Obviously when you’re coming to hi‑hat, especially when it’s very quiet with regard to everything else, it’s going to be off more. So it really was going in and moving around almost every beat to make sure it’s exactly the same as the original. Otherwise, then it starts to sound fake. And also, doing it via MIDI, you do get the builds and the quiet passages, all of that kind of thing, a bit better. So that makes it more human.”

Scott’s painstaking efforts also addressed another respect in which the original sound has dated somewhat. The album’s relatively tight low end made it perfect for cutting to vinyl, but arguably unnecessarily bass‑light on modern playback systems. “The kick drum sample I used the most throughout had a lot more low end than the original, and there were a couple of places I used a third kick drum, which was really sort of low end, just to emphasise a couple of spots.”

Getting Wet

Once the sample augmentation process was complete, Ken Scott bounced stems of important elements and took them into Abbey Road Studio 2. “I can’t remember what it is they use at Abbey Road these days, but we had a pair of their regular monitors down in the studio, and I had them set up Neumann U67s and AKG C12s and U48s all the way along the studio, and played back the stems.

“As luck would have it, there was an engineer came along and it looked as if he was going to be working on this project with me. He came along and he brought this mic called a Core Audio OctoMic, and he said ‘Would you mind if I set this up in the middle?’ I said ‘Fine.’ He set it up, we recorded it, we never listened to it when we were doing all the stems. We just listened to the Neumanns and all of those. And it wasn’t until we came to start mixing that we finally managed to pull up the OctoMic — and that was the one. It was so much better than the rows of mics that I had along Studio 2!”

“This mix was really interesting,” continues Ramazanoglu, “because when we first talked about it, my very first thought was ‘God, that’s a dry album!’ So when Ken said ‘I printed rooms’, I was like punching the air going ‘Oh my God, yes!’ Because you won’t listen to this and think it’s wet. Not even the stereo mixes. You don’t think ‘It’s a different record. It’s all huge reverbs!’ It’s nothing like that at all.

Emre Ramazanoglu: "We used Altiverb’s plate more than I thought we would. They did an unbelievable job that I think is slightly unsung, taking Altiverb 7 and making it immersive.

What reverb there was in the original mixes came from an EMT stereo plate, and although Scott and Ramazanoglu weren’t trying to recreate those mixes exactly, they did end up leaning on modern equivalents. “We did quite a bit of looking for reverbs,” says Ramazanoglu. “We were making sure it sounded natural in the track. It wasn’t just whip up that plug‑in and it’s done. We had to search a bit for a couple. Actually we used Altiverb’s plate more than I thought we would. They did an unbelievable job that I think is slightly unsung, taking Altiverb 7 and making it immersive. And so they’ve got sort of combinations of all the plates to make a 9.1.6 plate, which is really good. They’ve done a lot of really clever stuff in there that isn’t being shouted about really much, but it’s absolutely brilliant. And otherwise, Liquidsonics’ Lustrous Plates were used quite a bit.

“We didn’t really reference the original mixes that much until quite near the end, where we were having a listen going, ‘I’m not sure these are present enough,’ and then we had a quick listen to the original mixes, and I would sit with headphones and make a small tweak here and there just to get a bit more edge, because I think we were feeling out the room as well. It was some of the first mixing we’d done in RAK 4, which is fabulous. But we were working out, ‘OK, what sort of reference level are we working at? And could it just have a little hair more presence?’ and that’s the first time we really listened to the original mixes. When Ken was saying it was nice working together, it really was, because we’d both go ‘1dB on the vocal!’ at the same time. ‘This word, take the ess down!’”

Third Dimension

Freed from the requirement to match an existing mix, Emre Ramazanoglu found he needed to use much less processing than usual on the Ziggy Stardust remix. “Normally I will be recreating the mix bus or at least the mastering, so I’ll have linked sidechain processing going across. If I have 100 mono objects, it’ll be 100 mono channels of processing all sidechained and all working together and creating 'glue'. Here, we used one compressor on one channel on one track on the entire record, which is testament to the recording I have to say. There were a few EQs — a few — and it was mainly subtractive.

“There was so little done across the whole record, but this is why I was so impressed by this, because you could move things. I did very little of what I would often do to create space. There’s all kinds of ways of making space immersively that feel satisfying and coherent that mean I can spread effectively a mono or stereo source around a bit more, but we didn’t really do that, because of the way it’s put together. It’s slotted into a mix that was in 3D rather than in stereo, and the whole basis was thinking in 3D and thinking about... We found some interesting things with masking. Some things that we hadn’t heard before. Even some melodic stuff I’d never heard before, because it was interacting before with other elements. So there was recreating that feeling of interaction.

Emre Ramazanoglu: "One of the things I hadn’t done before is have the vocal ‘floating’. It’s actually slightly forward and up. It made it all quite spacious-feeling immediately, not having it all just pinned to the back wall."

“One of the things I hadn’t done before which is quite interesting is have the vocal ‘floating’. It’s actually slightly forward and up. It made it all quite spacious-feeling immediately, not having it all just pinned to the back wall. I really don’t use the centre speaker much in Atmos, because of headphone translation issues. But this worked! There’s a sense of belonging, every piece belongs in its new position for a reason, and it’s not just sort of plonked there. It’s like a 3D supported mix.”

Ken Scott: "I wanted it to feel like you were in the club or the theatre with the band, and it’s the way they’re positioned on the stage. The vocalist will always be out front, and the band just around that, not all in one line, because when they’re live they’re not all in one line."

Ken Scott sums up the philosophy behind the immersive mix thus: “I wanted it to feel like you were in the club or the theatre or whatever with the band, and it’s the way they’re positioned on the stage. The vocalist will always be out front, and the band just around that, not all in one line, because when they’re live they’re not all in one line. So it was very much that kind of thing that I was after, more of a live experience.”

“It’s quite immersive in that there’s a lot of sides and rear,” concludes Emre Ramazanoglu, “but unless you turned your head, I don’t know if you’d notice massively — except if you turned it to stereo, you’d go ‘Oh no, put that back.’ If you didn’t even notice it was Atmos, I wouldn’t be sad. As long as you felt enveloped by the music, then I’ll be delighted, personally. In the nicest possible way, it’s not about making a Dolby advert!”