Music Mixing
Unlike a conventional orchestral recording, Paul's score recordings exist primarily to serve the movie (though he also spent three days remixing the score in stereo for the soundtrack release), and a successful music mix is one that underpins and enhances the dialogue rather than fighting it. Getting this balance right is something that is uppermost in everyone's mind right from the initial composition stage. "The composer writes with the dialogue on the whole time, pretty much. Then the majority of the mixing is done with the dialogue and the atmospheres, the versions I have at the time, and working at the level that I think it's going to be mixed at by Nigel.
"One thing we discovered early on with the television series was there were lots of pizzicato parts in certain themes, and they used to write them very, very dynamically. I spent my whole time turning them up, because when I put the mix through the television against an atmosphere or a dialogue, you just couldn't hear them. So I kept telling the composer and the orchestrator, 'You need to stop being so dynamic in your scoring, because it just doesn't work.' It's got to be present. You want dynamics for emotion, but you've got to flatten them out.
"Obviously, I will flatten them out when I mix it, and much more extremely than anyone will ever realise! You tend to use more compression than you would imagine, to facilitate sitting behind atmospheres and to make it sit there rather than being too dynamic. Also, you tend to use more reverb [or, in this case, the ambient room mics] than you would if you were just mixing for music for pleasure, again, to take the music behind the atmospheres and the dialogue. I find that you can actually have the music louder then, in that case, and then it does its job better. If you mix something that's too dry, the dialogue will be dry, usually, and the atmospheres. So you'll be in the same space, which means that Nigel will have to turn it down. So it won't have the same effect."
"And on what Paul delivers to me," says Nigel, "I have to use very little compression, if any. I mean, I don't think I did any on the film; and on the television, just a bit, to give it a second bite of the cherry for the kind of small speaker, low-level, off-axis television listening."
In general, Nigel worked with the music mix that Paul provided, but there were moments when having access to stems or individual channels from the recording proved useful. "As part of the story, there's an orchestra that plays in the ball, and in three or four scenes, the orchestra is in the next room. Paul printed off the high ambient mics for me, and we were able to use a plug-in called Pentio to turn that stereo signal into a 5.0 signal, and we just used that for the sound of the orchestra in the next room. Again, we used no artificial reverb on that. It was just the sound of Abbey Road Studio One up near the rafters, and it sounded absolutely delicious. People have commented on that sound. You know, how did we do it? What did we do? What did we use? Well, we used the real thing, and it's absolutely beautiful.
"Then, when you cut to the in-vision orchestra there's a couple of hard cuts where the camera goes right in the middle of the orchestra, and we cut to Paul's full mix with the spot mics. You get this beautiful transition from the ceiling of Abbey Road One to being right in amongst your spot mics and the Decca tree, and it's just spectacular. Totally real."
Atmospheres
Nigel and Brad carried out their initial dubbing mix at Hackenbacker, working throughout at cinema levels, so they knew that what they were hearing was as close as possible to what the cinema audience experiences. "We had a few weeks of balancing dialogue, music and sound effects," says Nigel, "and then we exported the mix to the Mix Stage here at Abbey Road with the intent of having a second crack at it in this beautiful room, and also doing the Dolby Atmos process."
The Dolby Atmos-certified Mix Stage is Abbey Road's newest mix room, designed for film dubbing.
The Mix Stage is the newest mix room at Abbey Road. Designed for movie balancing and certified for Dolby Atmos, it's a hugely impressive space. "The joy of working here, for me, is that I was able to listen in a slightly different way," says Nigel. "We made very few changes, but the changes we made were important. I was able to totally relax. The room sounds beautiful."
We often think of Dolby Atmos as coming into its own in action movies, but it has more subtle applications too. "Obviously, there's nothing flying around in this movie," says Nigel. "There's no missiles above your head. But we used it to enhance the feeling of air and space. When you leave the kitchens and the servants' quarters downstairs and you go up into the gardens, we used the Dolby Atmos just to open up atmospheres, and it's so effective. Then, when you go back under stairs, you get this feeling of the kitchen as hot and claustrophobic, and it really, really reinforced that. So it was more when you didn't hear the air that the Dolby Atmos was effective. It's a good advert for Dolby Atmos, because to be sitting in a 1000-seater auditorium and to feel claustrophobic when you're in the servants' kitchen, but to feel that you've stepped outside into the fresh air when you go into the garden, that's pretty cool, I think."
All dialogue, Foley and effects were recorded dry, so the movie's contrasting internal spaces — claustrophobic kitchens, stately dining rooms and lavish ballrooms — were evoked using artificial reverb. Perhaps surprisingly, Nigel and Brad don't use impulse responses of real spaces; nor do they use software reverb plug-ins, relying instead on banks of TC Electronic and Eventide outboard processors. "If you've got a scene set in a prison cell, I think it should just sound like what we all think a prison cell should sound like," explains Nigel. "Many years ago, I did a film called Mike Bassett: England Manager, and we had convolution samples of Wembley Stadium. I did samples with the big PA rig and the mics and everything — and it sounded so unimpressive. I wanted it to have that kind of weird, Dopplery, multi-reflection thing, so we just dialled it up on an Eventide and banged it back through a Yamaha or something, and made this totally fake thing that just sounded like what you'd really want crowd announcements to sound like."
Using algorithmic rather than convolution reverb also gives Nigel more freedom to shape the sound. As he explains, sometimes quite radical processing is needed to create the effect that's appropriate to the context. "In a big environment, you pre-delay it enough so it's not stepping on itself or stepping on the thing that made it. I tend to roll off top end very early, so that teeth and tongues don't really touch the reverb, and that's a good way. I can use more reverb to create an artistic impression of a scene if I roll it off quite early, and you're just hearing the chest cavity stuff coming back off the walls, which is kind of how it works in nature. It just avoids clogging up the scene."
Deeper Downton
Feature film sound generally involves multiple specialised facilities: music studios, Foley houses, dubbing suites and so on. It's unusual for a single studio complex to be involved at almost every stage, and both Paul and Nigel found it a welcome change. "I think it's really cool that a movie like this was scored in Studio One, mixed upstairs, and then it comes out to my place," says Nigel. "I do a pass of it there, then bring it back home to Abbey Road in a room that I can relax in and listen through the room to what's actually on the tracks. To be able then to let Brad sit on the board and do all that clever stuff down there whilst I can take a back seat and enjoy it, and have fresh ears on it, that's really remarkable, and I'm really pleased we did it here."
Summing up their work on the movie, says Nigel. "It's been great to see the number of people who say it's like going to see old family friends. It's been truly wonderful to see that nobody's kind of going, 'They've changed it and I don't recognise this. What is this thing'? We wanted it to be a familiar experience for cinema-goers, but grand enough to make the ticket worth it!"
Judging by the number of tickets sold, it would seem they succeeded.
Spot The Studio
A shot from the Downton session at AIR Lyndhurst.Photo: Simon Whiteside
The entire music score for the Downton movie was recorded in four days. For one of those days, Abbey Road Studio One was unavailable, so Paul Golding used AIR Lyndhurst instead. You might think this would create obvious discontinuities, but this was not the case, even when the same cue was tracked at both studios. "We did compare the mix of 1M1 and the mix of 1M1 from the two different studios," says Paul. "They were rough mixes at the time, but you could hear a difference, and there's possibly slightly more low end at AIR Lyndhurst, but there's more clarity in Abbey Road One. The room was kind of divided about which one they preferred. As a stand-alone piece, it wouldn't sound completely different at all, because the tone of it was the same, the players were similar, the music was similar, the mics were similar, the mic amps were the same. Obviously, I'd mix it to actually sound the same, so the balance between the ambients and all the rest of it would be the same sort of thing.
"So, side by side, if you really wanted to listen, you could tell the difference, actually, but just hearing it on its own, you wouldn't know. If you had to edit between the two then you would possibly get away with it, but you'd possibly have to cheat that, to a degree, with maybe adding some fake reverb to a situation like that, rather than using so much of the ambient side. So that would be the trick you would use to make it all sound homogenous and coming from the same world, even though it wasn't, because the reverb sound would then traverse both recordings. In this scenario, we didn't do that. They were either whole cues or they weren't. I don't think we actually edited between them."
Cue Mixing
As is normal in Paul Golding's world, the Downton Abbey score was recorded to a click, with all the musicians on headphones. "I've done TV series before where we have used video streamers for the conductor to try and get a more musical thing, but that is laborious and actually it's very tricky to get that to work. If you're aiming for sync points but you're trying to do it just on the baton, it's hard. So you rely on the metronome! We're lucky that we've got musicians that can manage to play with this thing thumping in their ear and actually bring some musicality to it.
"Once the sound is set up for the recording, the majority of the work then for me will be to do with headphone balance. I'll be constantly riding the level of the metronome, depending on the level of the music, with a view to helping them play to that, but also with a view to when suddenly something stops and it's coming from a loud section. If the click is roaring away, I need to bring it down very quickly so we don't hear it [bleeding]. Occasionally you might be riding individual headphone sends.
"I mean, you could try and program all that and automate it, but the composer might well say, 'Well, I want that bit to be louder, I want that bit to be quieter,' so you could have programmed something that's not going to help you anyway. So you're better off just winging it and doing it from experience. You learn and you can read the score as well, and so you know what's going to happen next.
"I usually run five or six different headphone mixes, depending. So, for example, a string section will generate a certain volume, but for a French horn section, their headphones will be a lot louder, because they're making so much noise they actually can't hear anything. Drummers, again, another level up. You actually can't put them on, they're so loud! Certain people want to hear themselves in a room, certain people don't. For example, some of the harpists I work with want to hear what they're playing, some of them don't. I kind of know from experience that it's probably good to give a harpist their own feed. You know, that type of thing, you just learn, and you do whatever you need to do to get through the session."
The value of a large-format console such as the AMS Neve 88RS in Abbey Road Studio One is obvious when it comes to creating this sort of complex cue-mixing setup. "The beauty of a console like that is that you can very quickly set something up, because you're dealing with analogue and you can very quickly adjust it on the fly if you have to. Obviously there'll be a separate feed for the conductor, so I'll often be looking at the conductor and he'll be looking at me, and he might sort of gesticulate something that I'm going to interpret to us to turn his headphones up or down or something. Or he might do a sign for piano like, 'Too loud!' and I'll be doing that live, so you have sort of to be prepared to wing it and give yourself the flexibility to do that.
"I normally run everything pre-fade, apart from the click, because I'm running that live. There's eight on the console, so I normally use five or six. It's usually fine, unless you get a finickity, particular musician that you have to deal with. I tend to do it in mono, because half the time they only listen in one ear anyway, or one and a half, so you don't need to worry about stereo. It's a tool to get them to record. It's not about sonic pleasure!"