Skip to content

Scrappin’ and Testin’

January 26, 2017

Around four months have passed since the last post, where I wrote about a new movie I’m making featuring fighter jets and a monster. What’s happened since then? Some fairly early VFX tests. Mostly a ton of screenplay revisions and self-critique. I’m gonna write about the last few months of Firewing (current working title), some stuff I learned, and the state of things today.


Houston takes a beating!

When I read one of the first Firewing drafts Wyatt and I had, it was super exciting. Here was this nebulous idea that was translated into something that had characters and relationships, progression and stakes. I was stoked because the movie became tangible in a way simply having an idea in your head wasn’t. Sure, it was around 20 pages, which is generally way too long for a short film. But it told a story, it felt thrilling, and it fulfilled an important aspect that I really wanted the movie to have: to look like it was all shot in one very long and dynamic take.

With the early script looking satisfactory, Jason and I moved on to the digital previsualization process. The script came to life before us with rough staging, camera movement, and timing. You can catch a few segments of it below.


It was exciting seeing what the film would look like as we put it together in a 3D environment. We completed 5 minutes of the film’s running-time in previz when a problem started to arise. Wanting a more complete picture of the film before making any decisions, I pushed ahead with Jason until we arrived to around the 8 minute mark of the previz, when the film’s problem became undeniably apparent:

There were only around two minutes of engaging content in the 8 minute running-time we had previsualized so far. The rest was cripplingly uninteresting.

This movie I was thrilled to make, ready to spend a great deal of money and the sink next year of my life on, that also included fighter jets and monsters, was boring. I felt a bit defeated but eager to find solutions. We scrapped the previz and we opened up the script for revisions. We need to make it shorter, I thought, so a new draft came out that was a couple of pages shorter, then a new draft came out with dialogue revisions, then a new draft with a different ending, then a new draft with the beginning taking place somewhere else, then several more drafts with locations shifting around. Twelve drafts of Firewing went by, and I didn’t feel any closer to having a script that solved our initial problem. It actually felt like the script was getting worse.


A slide taken from the VFX presentation.

Questions ran through my mind. What was the cause? Were we trying to cram this huge story that simply couldn’t fit into the running time of a short? It didn’t seem like it. Our story was painfully simple, and all the plot points could probably fit in a paragraph. So what was the issue?

Despite letting go or changing various things in the revisions, the single shot concept remained in all them. This had big implications on how the story could unfold, implications that I was willfully ignoring for a long time. It meant that the whole script had to move in real-time, and we couldn’t really skip ahead. It meant that, barring any clever transitions, which we did have a few of, the camera is essentially locked to one character or location, and if we want to go somewhere else, we would have to do so in real-time. In the story, our main character flies a fighter jet from San Antonio to Houston. I had done months of research and I wanted the movie to be realistic and immersive, so the movie included all the real-world procedures necessary before take-off, things like strapping in and engine start-up, preflight checks, clearance to taxi, a slow taxi to the runway, clearance to take off, the process of taking off, the climb to altitude – all of those things happening, again, in real-time (though shortened as much as I could manage).


Another slide from the VFX presentation – rendered with Element 3D.

Though those things may be interesting to aviation nuts, they were not moving the story forward or showing anything meaningful, and they were really boring as a result. Omitting those things would kill the sought-after realism of the movie in my mind, so the only way around it was to begin the movie after all these things had taken place. But that didn’t work because we needed the first few minutes to establish a relationship that was central to the story and the emotional context that it needed to work, something that was sort of impossible to do after she boards the jet.

It took around fifteen drafts written over the course of three-ish months to let go of the idea that it all needed to be in one shot. It was a hard idea to let go. From the outset, it was a very appetizing challenge that the small team was thrilled to tackle – taking this big concept and telling it in this seamless way would be a huge achievement, and the movie would be more immersive and engrossing to watch as a result. It was an extra layer of ambition on top of an already ambitious project, the icing on the cake. It became so important to us, this idea, that it took all that time to let it go. It became important to people outside the project too. I asked a couple of friends about their thoughts on giving it up, and the response was generally something like “no way, it would be so awesome to pull it off, there’s gotta be a way to do it somehow.” And so there was additional fuel for more revisions trying to make the idea work. But it never did. And when I started wondering if I should scrap the whole project and work on something else entirely, I finally made the decision to abandon the one shot. The thing I had considered a tenet of this project ended up being unnecessary.

The draft that was written shortly after that decision was a big improvement. Gone were the minutes filled with unavoidable aviation procedure or idle travel time, replaced now by stuff that advances the story. The new draft includes a back-and-forth between two people’s situations that play off of each other. The tension builds and it doesn’t slow down. And the best part – the movie’s running time got cut in half. There’s a saying among writers that, at certain times, it becomes necessary to “kill your darlings,” the darlings being sentences, characters, scenes, chapters, elements that you’ve become attached to but aren’t working out in the grand scheme of things anymore. I’d say that’s the most succinct way of putting the development of the movie so far. The last few months were spent killing a rather large darlin’.


The aquatic monster in its underwater swimming pose.

So I’m back to being excited about the movie I’m working on again, after the growing pains of writing a decent script. In that time, some cool preproduction stuff happened as well – we commissioned an amazing creature designer who works in zbrush to build the movie’s monster, which you can see above and below. She’s called ATOYO.


Atoyo is only barely evolved to walk on land.

We’ve started preliminary testing with 3D integration of our to-scale digital fighter jet, the relatively new and soon-to-be ubiquitous F-35, with live-action footage. A very early test with shoddy motion tracking can be seen below.


And we’ve started to experiment with the scale of our (untextured and clay-like) monster, trying to find a size that feels right when tracked and mixed into live-action footage, which you can see below. Again, early stuff. We’re just getting a feel for things.


What’s next? Jason and I will be building the previsualization of the film based on the new and improved screenplay. As the movie comes together in this way, some additional changes and decisions will be made to reach the ultimate goal of being able to watch the previz beginning-to-end and feel satisfied with the choices we’ve made. After the previz, we move on to the pre-light, which is the process of taking the previz and lighting the scenes using physically-based rendering, which mimics the way light behaves in real life and produces photo-real results – crucial for developing the final look and feel of the CGI in the movie, which there is quite a bit of. The pre-light also serves to inform Jason on real-world choices and solutions he has to make for practically lighting our live-action composite shots so they end up looking right. After that, the more traditional preproduction aspects come into play, like casting, building the sets, rehearsal, and all that jazz.

The future with this thing is looking good today. Stay tuned.

One Comment
  1. permalink

    I love youuuuu


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: