You are here

Al Schmitt: Recording Bob Dylan's Shadows In The Night

Inside Track: Secrets Of A Mix Engineer By Paul Tingen
Published May 2015

Bob Dylan’s album of Sinatra covers is an unlikely triumph. So good, in fact, that it didn’t need mixing!

The news that Bob Dylan was set to release an album of Frank Sinatra covers left many people scratching their heads. Dylan’s gritty, nasal voice has always been the antithesis of Sinatra’s mellifluous crooning, and over the last couple of decades, has deteriorated to the point where it’s at times little more than a tuneless rasp. To tackle 10 classics of the Great American Songbook armed only with this weapon seemed a bridge too far, and many feared the worst.

However, as so often happened during his long and chequered career, Shadows In The Night confounded expectations. Perhaps because he’s singing more softly and not straining his voice, the 73–year old Dylan manages to hold the often elaborate tunes quite beautifully, and the fragile and wavering quality of his singing only adds to the poignancy of the songs. “It shouldn’t work,” wrote one critic, “but Shadows In The Night is quite gorgeous.”

Uncover Versions

The surprising depth and quality of Dylan’s vocals is only one of several factors that make Shadows In The Night such a distinctive artistic success. Another is the minimalist and sensitive arrangements, played by Dylan’s live band, with the occasional addition of some understated horns — there are none of the orchestral or big–band arrangements that were the essence of Sinatra’s approach. Many reviewers also noted the compelling general ambience and “amazing” sound of the album. This stemmed from Dylan and his band being recorded in a straightforward, old–fashioned manner, resulting in a sonic warmth and intimacy that perfectly complements the late–night melancholy of the songs and performances.

“I’ve wanted to do something like this for a long time,” wrote the singer, “but was never brave enough to approach 30–piece complicated arrangements and refine them down for a five–piece band. That’s the key to all these performances. We knew these songs extremely well. It was all done live. Maybe one or two takes. No overdubbing. No vocal booths. No headphones. No separate tracking, and, for the most part, mixed as it was recorded. I don’t see myself as covering these songs in any way. They’ve been covered enough. Buried, as a matter a fact. What me and my band are basically doing is uncovering them. Lifting them out of the grave and bringing them into the light of day.”

Al Schmitt, teaching a  masterclass at Mix With The Masters.Al Schmitt, teaching a masterclass at Mix With The Masters.Inside TrackIn other words, while Dylan’s musical approach was very different to Sinatra’s, the recording process for Shadows In The Night was very much of a piece with the way records were done in the ’50s. And if you intend to record in that way, who best to call than someone who was actually there at the time, and who even worked with Sinatra? Enter Al Schmitt, most likely the recording engineer with the longest career of any still active.

Harry’s Game

Schmitt first set foot in a recording studio during World War Two when, as a young boy, he assisted his uncle Harry Smith at Harry Smith Recording in Manhattan. He has since gone on to become the most venerated engineer and mixer in the history of music, with a record–breaking 23 Grammy Awards to his name, and a set of credits that could easily fill a book, including legendary names like Henry Mancini, Sam Cooke, Jefferson Airplane, Barbra Streisand, Steely Dan, George Benson, Toto, Miles Davis, Natalie Cole, Diana Ross, Elvis Costello, Luther Vandross, Neil Young, Brian Wilson, Paul McCartney and many, many others.

Surprisingly, however, Schmitt had never worked with Dylan before the call came from the singer’s manager, Jeff Rosen. “Unfortunately, I was busy at the time they had planned,” remembers Schmitt, on the phone from his Los Angeles home. “I was really disappointed, because Dylan was still on the bucket list of artists I’d love to work with, but never had. The very next day I got a call back, saying that they were prepared to move their schedule to a time when I was available, because Bob really wanted to work with me. Evidently he’d heard my work, including the Duets [1993] album I had recorded with Sinatra, and he must have figured that I was the right guy to record it.”

Capitol Cities

The company’s next step was to decide where to record the album. “They had gone around looking at studios,” explains Schmitt, “and when Jeff got to Capitol Studio B, he just loved the way the live room sounds. We all went there later on, and Bob also commented on how nice the room sounded, and he asked, ‘Where would be the best place for me to sing?’ My assistant, Chandler Harrod, said, ‘Well, right where you’re standing would be a good place.’ And so he did, and it was just great.” Inside Track

In addition to its sound, Capitol Studio B also had symbolic and practical advantages, respectively because Sinatra often recorded there and because Schmitt regularly works at the studio. According to the engineer, the sessions were conducted in extremely good spirits, and he repeatedly mentions how enjoyable they were, “It was so comfortable for everybody, we all had such a good time doing it, it was great. I had a great time!” This relaxed spirit inarguably shines through on the final result. But what allows this spirit, and the music, to shine, was the recording approach taken by Dylan and Schmitt. It was, says the engineer, “just the way records were made in the old days”.

In practical terms this meant a recording approach that was, by modern standards, almost comically simple: seven microphones going through Capitol’s 56–input Neve 8068 console,Only eight of the 56 inputs available on the Neve 8068 in Capitol Studio B were used.Only eight of the 56 inputs available on the Neve 8068 in Capitol Studio B were used. a bit of reverb on the electric and pedal steel guitars, reverb and compression on Dylan’s voice, and everything going down live to 24–track and two–track tape. There were no edits, and for three songs, the two–track laid down during recording became the final master. In the other seven cases Schmitt made a few level adjustments to the 24–track balance, but no effects were added. These few modest alterations don’t even warrant the term mixing.

Uniquely, then, this is the first album we’ve featured in a series subtitled ‘Secrets Of The Mix Engineers’ that did not involve any mixing!

No Fixes

According to Schmitt, the procedure for the recordings of Shadows In The Night was the same for each song. “We did two three–hour sessions every day, for three weeks, five days a week,” remembers Schmitt. “In between each two sessions we had dinner. Dylan had this small music player, and at the start of each session we’d all be there in the live room to listen to the old Frank Sinatra recording of the song we were going to record, not to approach it in the same way, but to get an idea of the interpretation. We then would talk for maybe a couple of hours about how we were going to do the song.

“The bass player, who is the musical director, had the chord charts for everybody, and would sketch things out for each player. A lot of the time was spent on making sure that each musician was playing the right parts, with the right performances. We also wanted to make sure that everyone was comfortable and could hear each other. Most of all, Dylan had his input in how he wanted things to be and how he wanted the guys to play. He was the producer, so he had total control of what went on. He would comment on tempo, on how the rhythm guitar played, what he wanted from the pedal steel, and so on.

“Then they’d do one or two takes, and would come into the control room for playback. People had ideas for how to improve their parts, and Dylan might think he could do a better vocal, and we’d discuss the balance, and they would go out again for one or two more takes. Sometimes the very first take would be the take, so there was nothing to adjust, but most of the time after listening to it, they had their ideas, and I would say that I would need a little bit more volume here, or little bit less there, and I asked them to adjust that in the room. When there was a guitar solo, he just played a little louder. I did not want to be riding faders, I wanted it to be natural. I rode faders on the vocals, but for the rest, once I set it up they balanced themselves in the room. After this there was very little for me to do. That was it. There was no editing, no fixing, no tuning. Everything was just the way it was.”

This involved making sure that the musicians and the microphones were set up in the right place and the right way, something Schmitt for the most part took care of during the first session. However, his job was complicated by an unusual request from Dylan, which was communicated by his manager.

“Jeff told me that Bob did not want to see a lot of microphones around,” recalls Schmitt. “So I had to use as few as possible, and to try and disguise the ones that were there as much as I could. I don’t know why this was. Perhaps because he wanted to have more of a very relaxed, living–room atmosphere, with him and the musicians not so acutely aware that they were being recorded. Dylan also did not want to go through all the other things that are involved in making records today, like earphones and every musician making their own little mix, and so on. So there were no headphones. At one point Dylan couldn’t hear the acoustic rhythm guitar enough, so we just moved the guitarist closer to him. We balanced everyone in the room around Dylan simply by placing the musicians in the right positions.”

Super Warm

On Shadows In The Night, Dylan’s band played the following instruments: de facto musical director Tony Garnier was on double bass, Donny Herron on pedal steel guitar, Charlie Sexton on electric guitar, Stu Kimball on acoustic guitar, and George Receli on drums and percussion. The musicians were set up in a semicircle in front of Dylan, who was facing them.Capitol B’s live room had been used for many of Frank Sinatra’s own recordings, and its fabulous acoustic made it an obvious choice for the project.Capitol B’s live room had been used for many of Frank Sinatra’s own recordings, and its fabulous acoustic made it an obvious choice for the project.

“The entire setup was really simple,” explains Schmitt. “I loved it all these years ago when I watched my uncle record in New York, how they moved the musicians to get the right balance in the room, because there were no earphones and overdubs and mixing were not possible. Especially because of the request to make the mics as invisible as possible, I did have to draw on my years of experience of making records. But once I had set everything up and they heard the first playback, that was it. Bob just loved it. When he heard his voice, he said that it had not sounded this good in 40 years. He was knocked out by the sound we got on his voice and just totally thrilled.

“I used Capitol’s Neumann U47 on Bob’s vocals. It’s the very same microphone that was used on Frank Sinatra, Dean Martin and Nat King Cole. It’s a great–sounding microphone. Fortunately, because I work a lot at Capitol, I get to use it frequently, and I’ve had it on Paul McCartney and on Diana Krall and others. The mic was maybe nine or 10 inches away from Bob, with a windscreen. Pops and esses were no problem. I used the Neve 1073 mic pres in the old Neve desk at Capitol on all the mics. It has an amazing sound, with lots of punch and warmth. So Bob’s 47 went through that. The only compression I used on the entire album was on Bob’s voice, a tiny bit of an old mono Fairchild. I barely touched it, I used it mainly for the tube sound. It just added some warmth. On the desk I also mixed in some of Capitol’s live chamber number four on his voice.

“I tried to use ribbon microphones as much as I could, because I wanted to get it as warm–sounding as possible. On the band I had, from left to right, looking at the band from the control room and Bob’s perspective, an Audio Technica AT4080, which is a really nice–sounding ribbon mic, on the acoustic guitar, placed three feet away, aimed towards the sound hole. I had one Neumann M149 on the upright bass, at least four feet from Tony, placed kind of low. Normally I use two microphones on the upright bass, one on the F–hole and one higher up on the fingerboard, but because we were trying not to have so many microphones out there, I had just one microphone, aimed at the F–hole. The drummer mostly played brushes on a pad, and occasionally sticks, plus timpani on one song, and I used a stereo AKG C24 on him.

“I also had an AT4080 on the electric guitar cabinet, which I placed behind the guitarist, so the mic wouldn’t be visible to Dylan. Next to that, on Dylan’s right, was the pedal steel, with yet another AT4080 ribbon on the cabinet. The steel guitar amp had some echo, but I felt it was too much, so right at the beginning I asked Donny to switch that off, so I could add and control the reverb. I used some Bricasti M7 reverb on both the electric guitar and the pedal steel. That was it in terms of effects. I did not use any EQ either. It is very rare for me to use EQ while recording. I try to make adjustments to the sound like that using different microphones and microphone placements. In general, also during mixing, I try to use as little EQ and compression as possible.

“We had horns in for a few days [the album has two trombones and French horn on two songs, and two trumpets and one trombone on another song], for which DJ Harper had prepared the arrangements. Dylan and he had already discussed these beforehand, so there were very few changes. We had the horn players in a booth to the side of the main live room, and left the glass doors between the two rooms wide open, so the sound leaked into the main room, and Dylan could hear it. The horns were mostly picked up by the Neumann M49 ambient mic that I had placed in the middle of the band, close to Dylan’s vocal mic. It’s the one that Barbra Streisand uses, and I set it to omni. It was this mic that gave the album such an airy sound. I also had a Royer 122 on the trombones, a Neumann U67 on the trumpets, and a Neumann M149 on the French horn. But again, most of the horn sound comes from that M49 room mic. When there’s a trombone solo it sounds like a Tommy Dorsey solo!”

Two Tapes

As already mentioned above, all the sessions were recorded simultaneously to a 24–track Studer A827, running two–inch tape at 30ips, and a two–track Ampex ATR half–inch machine, also running at 30ips. No noise reduction was used. Everything was also going down to Pro Tools, running at 192kHz, but as Schmitt explains, this was purely for backup. “Digital sounds excellent at 192, but tape still adds something. It makes everything sound a little bit more pleasing to the ear. The problem is that not everyone can use tape. Today not everyone sings or plays as well as they did in the days of Sinatra, Nat Cole and so on. Those singers sang in tune and had microphone technique and if something was a little out, it was OK. But many people today need all the editing and tuning capacities that digital offers. Obviously, for Dylan and his band this was not an issue.”

At one point Schmitt did suggest some kind of mixing process, but Dylan had other ideas. “We wanted everything to sound like it was done at the same time in the same room,” the engineer recalls. “I rode the fader on his vocals, and I panned everything pretty much as it was in the room, apart from the electric guitar, which I panned to the left, opposite the pedal steel. I placed the bass where I felt it should be, which was not too loud. At end of the session we listened back to the final takes, and that was it. Dylan decided which take of each song he liked best, and that one would immediately be locked as the master. When I mentioned mixing Dylan said: ‘No, I love the way this sounds.’ We did spend the final day doing some touches on the other songs, like rebalancing a few things slightly, maybe bringing his vocal up in a spot or two, basic things like that.

“In some cases Dylan preferred the original rough to the version with the final touches. We thought we had improved it, but when we played it to him, he’d say, ‘Let me hear the original,’ and he’d like that better. Dylan did not want anything fancy done. He wanted it to sound as natural as possible. He completely fell in love with the roughs, and wanted me to stay very close to them. Again, it really was just the way records were made in the old days! In those days you could not edit or fix things, and so you had to do the take when things were emotionally right. And you chose the take that had the feel on it. This is why so many records from back then are so much more emotional and touch you so much more deeply. Today everything is perfect, and in many places we have taken the emotions out of records.”

Listen Before You Judge

All in all, 23 songs were recorded during three weeks which sound as though they were thrilling and magical, even for someone as dyed–in–the–wool as Al Schmitt. The engineer confessed, though, that he also wasn’t sure at the very start of the project how things would pan out. “Right at the beginning I was a little concerned too. I wasn’t sure whether Dylan would be able to sing these kinds of songs. But I had goosebumps the first moment I heard him and his band coming through the speakers. I still remember the first time a take went down. I was just totally amazed at how good it sounded, and how emotional it was.

“Bob loves these songs and put his heart and soul into singing them. If there was something slightly off–pitch it didn’t matter, because his soul was there and he laid the songs open and bare the way they are. He also wanted people to experience exactly what was recorded, hence no studio magic or fixing or tuning things or moving things around, and so on. For the same reason I did not add anything before sending the two–track tapes off to mastering, which was done by Greg Calbi at Sterling Sound in New York. [The album credits mention Doug Sax, but Dylan’s manager confirmed that this is a mistake.] I understand that very little was done during mastering. Bob loved the sound of the reference CD that we gave him, and that is what he wanted the final record to sound like. He did not want a lot of level put on, nor did he want any EQ added.

“Many people I know were sceptical when they found out that Dylan was going to release an album of Sinatra covers, and when they asked me while I was working on the project how things were going, my reply always was, ‘Don’t judge until you’ve heard it.’ After we finished, I got a call from Diana Krall, who had heard the material via Bob’s manager, and she said that it touched her so deeply that she had just started to cry. Then Elvis Costello called, and then T–Bone Burnett. People who had heard it were raving about it, and after the release several of the other artists and people I work with and who had been sceptical called me and said, ‘Al, you were so right.’ I knew from that first playback that we had something really special.”

Neil Young’s Storytone

Another high–profile project Al Schmitt conducted last year was the recording and mixing of Neil Young’s most recent album Storytone, which was released in November. It consists of 10 original songs, seven of which were recorded with an orchestra, and three with a big band. On the deluxe version of the album a CD is included with all 10 songs performed by Young alone.

“This was recorded in many ways similar to Bob Dylan’s album,” says Schmitt, “because also in this case everything was recorded live in the room, doing just a few takes per song, using no headphones, doing only minimal mixing and things going live to 24–track and two–track tape, with Pro Tools as backup. Again, in a few cases, the two–track rough mix that went down during the recordings was used. I recorded all sessions, the orchestral and big band sessions, and also the solo songs with just Neil on ukulele or piano. We had one day with the big band at EastWest Studios and two days at the Barbra Streisand Scoring Stage at Sony Pictures Studios, with a 65–piece orchestra and a 34–piece choir. Neil stood right next to the conductor, and sang into the same Neumann M49 that Barbra Streisand used. I figured that mic would work well on his voice, and it did. This setup sheet drawn by Al Schmitt’s assistant, Chandler Harrod, shows the arrangement of the musicians within the live room. The brass section, where used, were in the isolation booth but with the door open. Note that two mics are shown on the bass in this drawing, but only one was in fact used.This setup sheet drawn by Al Schmitt’s assistant, Chandler Harrod, shows the arrangement of the musicians within the live room. The brass section, where used, were in the isolation booth but with the door open. Note that two mics are shown on the bass in this drawing, but only one was in fact used.

“I brought a lot of my own gear to the Neil Young sessions. I have preamps by Upstate Audio, Neve, Studer, Mastering Labs, and more, some vintage and valve compressors and EQs, and more than 50 microphones, including a bunch of M149’s, a dozen Royer mics, several Audio–Technica mics, and so on. I just got the new AT5045, which is great for overheads on drums. I used my Neve 1081 mic pre on Neil’s voice, though I did not use compression in his case. The main mics on the orchestra were three M50s on the Decca Tree, and I had several spot mics. The choir was recorded with four M149s. The mics went either through some of my preamps, or the Neve 88R desk there. Because it all was recorded live, I really had my hands full! But it’s how I learned. Mixing was just a matter of putting up the faders and making some minor adjustments, with very few compressors or EQs used, and that was it. It’s the way we used to make records, and it’s a lot of fun!”

The Best Time For Sound

“In terms of sound, I think the era from the ’60s to the early ’80s was the best, when we were still recording on analogue tape,” says Al Schmitt. “The early digital stuff just sounded dreadful. But digital has since come of age, and I think 96kHz is acceptable, and 192kHz terrific. I love the sound of digital today. And it will only get better. Plus digital of course has the edge when you work with people who can’t sing or play as well as was more common in the analogue days, and you need all of digital’s editing and fixing tools.

“Having said all that, digital still has drawbacks. I totally don’t like working in the box, for example. I’ve mixed one album in the box, and people said it sounded great, but I far prefer using a console. What happens now is that people don’t have the budget any more to afford a project with a console, and as a result, sound quality suffers. Of course, there also are all the lossy consumer formats, which I don’t like, and the loudness wars have made many records unlistenable. Records come on the air with radio compression and the compression that was put on during mixing and mastering, and often just sound terrible, with distortion and so on. Also, sometimes they have so much treble that the sound goes through my ears like a razor blade. I don’t understand it. I belong to an organisation called Turn Me Up! and that’s what we believe in. There’s a volume knob, and if you want it louder, turn it up!”