Progress on Gerry’s cabin

Gerry’s cabin or Act 2 is very close to it’s final iteration. The length was cut from 9:00 minutes to 8:15, 45 seconds. We lost some lag and tuned the performance by splitting several two shots.

My first foray into tuning was way back, during an early pass at Gerry’s shots during the ‘chinese animal signs’ riff. We see Gerry facing camera frame right and Christina over the shoulder frame left. In some spots she was moving too much, so I split the screen using a garbage mat and replaced her with a calmer version from a different take.

A bolder approach was in making imperceptible jump cuts on Gerry’s CU, “You move so quick…”, snipping pauses where his facial expression and position we’re very close. One jump was subliminally problematic and I returned to it during the latest pass. The in and out don’t quite match. Based on my knowledge of roto, I can isolate an edge frame, rotate it against a clean plate of the background and weld the jump together.

With this insight, I tackled the Phil Dick delivery. In this shot Christina faces camera and Gerry is delivering from profile to over the shoulder. I wanted tighter integration between their interactions – her speaking, him turning and giving his line, her turning and giving her line, then tight pacing for the delivery of his lines matched to her changing reactions. It’s a lock down with lots of room between him and her, so it’s an easy split. I could handle most of his tuning with subtle jumps, which could be tuned with roto welds.

I’m proposing a radical series of welds for her head turn. When she says, “It’s god”, we cut from a wide shot of them both to a mid of C full and CU of G in profile. There’s a lovely match of her in both shots, but then she does a weird hesitation when she turns her head that screws up the tuning. I could forgo the match and cut to her later in her head turn, but then I loose my lovely match and it’s not quite as convincing. The plan is to use the best frames from her head turn and throw out the rest, creating a smooth head turn out of non contiguous frames. This will require a trip into AE  for sure.

Once the roto team is cranking again and the roto weld technique is proven, I’ll shine my editorial light on act 1. It’s a little backwards, we’ve already rotoed several shots from act 1, but I know it could be tighter still and am itching to got there. I can refocus the team on material for act 2 in the interim. I’ll take a gander at 1 today and see if the remaining dead bird shots can be completed, be good to have one meta-task done.

What’s the point of all this tuning with imperceptible jump cuts and roto welding? First inventing new terminology is super fun. Second, I can bring out the latent power of the moment. Where direction or acting falls apart, editorial can be a method of repair. Cutting full frame is powerful and cutting split frames even more so. Actors can be more tightly integrated, brought into a new time space and the scene re-performed. This approach can make up for inexperience on set. If the roto weld can be perfected, I’ll have significantly advanced my craft in the process of birthing DOG.