Issue link: https://digital.copcomm.com/i/1543098
30 PRODUCTION SOUND & VIDEO – Spring 2026 That mindset carries through all the way to Andy and John on the dub stage. Andy does not treat "dialog" and "vocal" as separate species. For him, as for me, it is all performance. Recording "For Good"—Leaves, Puddles, and a Six-Mic Duet If Wicked: For Good had a single scene that embodied the entire philosophy of the film, it was the balcony sequence for "For Good." On paper, it is a simple idea: two friends saying goodbye. In practice, it is a delicate blend of dialog, freeform singing, tightly structured musical material, and camera choreography, all set on a balcony that needed to feel like a real autumn evening, not a soundstage. Real Leaves, Real Puddles, Real Wind Before we even began to talk about how we would record the vocals, we had to understand what Nathan Crowley's production design and Jon's visual language required. Jon's vision of Wicked and For Good was built around vast real sets. Nathan's stunning designs were built by real craftspeople and artisans. They did not want a CGI balcony with digital leaves and simulated puddles. Those elements are beautiful for the eye. For sound, they are treacherous. Crunching leaves underfoot and water squelching in shoes are exactly the sort of mid-band transient noises that fight the same frequencies as the human voice. Add wind machines, and you have the classic recipe for a scene that "looks great but will all be ADR later." We were not prepared to accept that trade-off. Working with the Special Effects team, led by Paul Corbould, we deployed the Silent Wind system that we had refined across the show. Instead of parking huge fans on the stage, we kept all the noisy machinery outside the soundstage and brought the airflow in through stage walls with runs of flexible twelve-inch tubing, each handled by an SFX technician who could aim and feather the airflow with great precision. What we heard on the mics was not mechanical motor noise; it was a broad, smooth "shhh" of air movement and the sound of leaves shifting. We adopted a very simple rule: • Only put leaves where the camera can see them. On a wide shot, you might see a carpet of leaves stretching to the back of frame. On the reverse, that area may not be visible. We became ruthless about clearing any part of the floor that was not in shot. This is where Taz Fairbanks came into her own. Shot by shot, lens by lens, she would liaise with the camera team, look at the monitor, then quietly sweep leaves out of the zones that were off-camera, while SFX kept them alive in the areas that mattered visually. The same was true for the puddles. Visually, they sell the idea that it has been raining, but water and shoes are not friends of live vocals. Every time the lens went tighter, Taz and the art and props teams would go in, mop out the puddles that were now out of frame, and leave only what was absolutely necessary in the composition. You cannot say you care about original performance and then allow unnecessary noise sources to exist in parts of the set that the audience never even sees. Beginning as Dialog, Behaving as Song "For Good" does not start like a "musical number." It begins like a conversation between two people on a balcony. Visually, Jon and DP Alice Brooks approached the early part of the scene exactly as you would treat a dramatic dialog sequence. We were largely on Steadicam, floating with the actors as they walked and talked, with the camera weaving in to find their faces. Technically, I treated the opening like dialog too, but with one crucial difference: I knew that at any moment, Ari or Cynthia had the freedom to step into singing at any given moment, and I had to be ready. We were using Schoeps CMD 42 digital microphones with MK41 hypercardiod capsules on the booms, sending AES straight down cables to the cart. That gave us an incredibly clean, modern boom sound with all the transient detail we needed, and a consistent digital path right into the Scorpio. On the early beats of the scene, Arthur and Robin were routinely working as close as they could; twelve inches above the actors' hairlines on wider framings, down to three inches above the hairline on tighter shots, always dancing around the Steadicam. From a sound perspective, the most interesting thing about those opening passes is that the performances were not locked. Cynthia and Ari made a creative decision to allow certain words, certain lines, to just slide into singing, depending on how they felt in the moment. On one take, a line might be spoken, on the next, it might soften or swell into melody. To support that, both actors wore IEM's from the start. They knew that if they felt compelled to sing a word or a phrase, Ben Holder would be there, following them on the keyboard. Sometimes he would only give them a single note or a tiny fragment of harmony, just enough to catch the emotional weight of what they were doing. But the key point is that his job was not to lead; it was to follow. Technically, the workflow at this stage was: • CMD 42's providing our primary vocal capture. • DPA 4061 lavaliers on each actor as backup and as an option for some angles. • Ben's keyboard feeding Pro Tools which in turn fed the IEM's. • My dailies mix presenting the scene to Jon as an integrated whole. Creatively, the opening is already a piece of live sung cinema. We just have not formally "started the song" yet.