Filming Chap 1&2: Lessons Learned: Cinema Production Conventions at Odds with Our Form.

Danny's POV rig for canon 7D

Just wrapped on the first fast-paced weekend on the set of Penumbra . Our first shoot was a product of many factors including time constraints and the effort to find what works using experimental methods. A constant question we struggled with was the polish of POV. It is hard to get right, to feel "authentic" as a frame of vision (especially when filmed in portrait). Often it reads as amateur; however, in viewing the footage on the device it was intended for (the ipad) it somehow feels more appropriate than simply reviewing it on the laptop screen. This shouldn't be surprising, but I am still working on accurately articulating why this is. The tangible quality of holding and manipulating the device seems to ground the footage in a way that viewing it in post on a laptop screen doesn't. All in all, this makes editing and choosing footage from a laptop screen feel a bit disconnected from the final form. Should one keep building the app with new footage, each time a clip needs to be scanned for it's usability? This method would be time consuming. 

Selecting "good" footage is not influenced by the structure of its playback, but also by the hurdles in its creation: what can we salvage from a shoot that didn't go as anticipated? Now is the time to regroup and reconsider our approach. Therefore, the final version of the first two chapters of Penumbra will end up looking completely different than their scripts, drafts and planning notes. On paper they worked but when edited in light of our form and project structure. . . the conventional structure of the script paired with POV made the material feel too heavy. It will be interesting to see what material will be kept from this shoot. I suspect my next posts about Chap. 1 & 2 will be about our strategy forward rather than an analysis of past hurdles. There were two hurdles that have shaped how we edit the first two chapters. These hurdles are large influences because they inevitably will force us to reshape and reconsider our interface to accentuate the strengths of the footage.

There are two reasons we must reconsider the footage and our interface

1. Writing Conventions of Standard Cinematic Dialog at Odds with our Form

2.  Challenges of Acting for POV and Role Fit.

1. Writing Conventions of Standard Cinematic Dialog at Odds with our Form

Originally, this post was meant to describe the progression of Chap. 1 and 2 from rough draft to final iteration. However, it is best to post the evolution later and instead focus on what lessons can be learned from what didn't work. Given our form, I immediately should have dropped the effort to write a conventional screenplay in favor of accentuating the needs/properties of the “eye open/closed” interface. We are not working within the traditional language of cinema that screenplays are meant to express and streamline. The written structure of a conventional screenplay + dialog did not benefit the qualities of the interface we are using. Perhaps the months of work learning character/story development, script format and shooting procedures will inform the final material, but in terms of these chapters...it's back to the drawing board to find something that works.

To understand why, the first thing that needs to be examined is the unique properties of the form we are working in. The basic mechanics of the interface originally imagined for the first, second and final chapter are as follows. 

The user first encounters a (seemingly) dark world of thoughts/sensations/perceptions. This internal world is comprised of free floating thoughts (conscious/subconscious) usually in text. If the reader pinches open the eyes, they see an external world of video that describes James' surroundings. To remain in the outside world, the user must continue to pry apart James's vision. They must force him to engage. However, James is progressively going blind. This is a literary trope and device we are adopting into the interface. I will speak more on this later. If the user attempts to "watch" the external world, then the vision whites out after a period of time (increases as the story progresses) and forces the reader back into the internal world of text thoughts. Alternatively, the user can tire of the effort of prying open the eyes (holding the gesture) and return to the default internal space. If James' vision whites out, he must rest his eyes before they can be opened again. The reader can open the eyes at anytime as long as this rest period has been eclipsed. Additionally, the reader my "crunch down/collapse" any of the internal text thoughts to reach a more subconscious space of fragmented content or sensations of light that respond to the internal space. As the story progresses, these two spaces, the internal and external, become increasingly complicated and conflated.

Our shooting process consisted of uninterrupted POV takes of the script with the Cannon 7D head-rig pictured above while re-shooting parts of the script with a canon 5D for close-up details. On paper, It took a few months of story development and a month of script writing to get the scene right. In the end, it seems that I wrote the script at least two levels removed from the reality of our project. A. Our project does not have the cuts and variance of shots that a script like this demands. For example, the script had many small subtitle moments that would have worked with cut-ins but just seemed too forced and melodramatic when the actor has to force the POV head-cam to notice them artificially. B. The interface did not accommodate a flow of dialog or conversation as well as we hoped. Though the script was very minimal to begin with, the constant switch between the internal and external world made a stream of dialog feel incompatible. I'm still discovering how to articulate this, but I think it has something to do with the constant flow of conversation going on in the external world vs. the jarring rupture of having to switch between modes of internal/external that the interface encourages. The hope was that we could play up this rupture to create a separate stream of internal text commentary that James would be censoring (only accessible to the reader) during his conversation with Luke. However when we converse, so many cues are not given through language and rely often on visualization or tone. This was the same with the subtleties of the script. If James is unable to see Luke to respond to his cues, then it makes the conversation feel more stilted and somehow overdramatic. Ultimately, the interface/form felt strongest in moments of downtime, menial tasks, tension and silence between characters.

2.  Challenges of Acting for POV and Role Fit.

Dialog, delivery, and timing also significantly influenced the artificial or forced quality of much of our results. The constraints of the set and filming in wide-angle POV dictated that Danny and I could not largely see the actors during takes or redirect results after. More time to review the footage would have been helpful. This is understandably the benefit of having a bigger crew. I rehearsed takes with actors, decided blocking and conducted readings for a few hours prior to the shoot.  These efforts didn't always take, especially because adjustments should have been made between each take. Ultimately, line delivery significantly impacted our ability to find one continuous POV shoot to represent the external world that felt "natural". I'm sure this is also a reflection on the heaviness of material and extreme subtly of dialog. Perhaps, the whole thing should have been played lighter, but carried an edge. This was the original intent. James' voice is so essential (it is the representation of the character, it is what you hear when all is dark on the internal world). Added inflections and dramatic stage emphasis served only to kill the quiet, contemplative intentions of James' character. There is a significant difference in the demands of acting for camera and the stage. Admittedly it was a difficult general fit for our hire. He was definitely accustomed to acting for distance theater and placed great emphasis on conveying that he was "acting". Our rig also impeded a bit of the actor's vision in a way that his motions did not feel like natural observations when the footage was reviewed. The gaze would have been better if it was subtler, rather than set on accentuating and communicating "how" we look. Just looking would have been sufficient. There is a definitive art to acting for a POV camera on your person. This is admittedly a challenging task that more training should be provided for in the future. Especially, since we  identify POV rig with amateur aesthetic. The combination of overacting for camera and forced delivery of lines created a disassociation from the character (read as amateur) that served to alienate the reader when the interface was applied. This was far from the visceral intimacy we were hoping for.

Two results were obtained from our first day. 1. The story, interface and mechanisms have to be rearranged from what is salvageable. The first two chapters will no longer be uninterrupted literal takes of the external world. Instead, they will be fragmented (more representative of memory) and reintroduced in a way that is not fluid/constant. They will no longer be in the outside world, but some of the footage is still being kept for the expressive/impressionistic subconscious internal spaces. Hopefully, the concept/content will not be worse, just different from the way it was originally intended. Such are the types of jumps that must be made when transitioning from page to the screen. The difference between the script and output will evidence that these details not only changed how we had to think about our attempt at combining writing and media, but also the story: thus, the structure of the project itself. 2. Unfortunately, but necessary, James needs to be recast or written out as a visible character. This will severely impact our future shots when James is seen acting in 3rd person during his reflective memories. This is disheartening. I do not want to give the news, but I need to think what is best for the project.

How do we move on?

• Edit current shoots and rewrite project structure so it is less fiction and more poetry dialog isn't mostly for plot points, just set emotional tone…Make it about tone. 
• Write out James from acting intensive scenes... ?*I think it's important for the later scenes that are a conflation of memory space and past to be in third person and have James visible. Also, selfishly  I’ve spent so long creating my characters that it’s going to hurt to short change them!  Answer: Try Recasting James.
• Find an different actor or a non-actor that has the natural mannerisms of the character?
• Reshoot? with firmer instructions and "say it exactly like this", "look exactly here"
• Overdub James' very coached dialog over 1st person POV scenes?