AWS for M&E Blog

AWS Artist Series: Laying out the opening shots of Spanner

This is the 5th post in our AWS Artist series, find the other four below:

AWS Artist Series: Compositing Spanner­­
AWS Artist Series: Creating characters with dynamic wrinkles 
AWS Artist Series: Animating Noa 
AWS Artist Series: Rigging the bridge for “Spanner”

Project introduction

Spanner is a short film about two bridge makers: Ulysse is an overly confident worker whose insecurities become very evident when he encounters Noa, a bridge building expert. Spanner was created by FuzzyPixel, an internal AWS team that tests the tools and services that AWS is developing for animation and visual effects studios. Our team ensures that these tools and services stand up to the rigors of real-world production. To do so, we create animated content using industry standard tools, and we aim to meet the level of complexity of the productions that our customers work on.

The finished short film can be viewed here: Animated Short Film: Spanner.

In this post, I discuss the process for laying out the cameras for the opening shots of Spanner. I specifically focus on the experiments and creative decisions that happened along the way to create the final composition of the shots.

What is layout?

Layout, sometimes referred to as “previs”(short for pre-visualization), is a step in film making where shots are planned out before actual production begins. In the context of computer animation, layout involves translating the 2D storyboards into 3D for the first time, using simple versions of the characters and environment. During layout, initial camera and character placement, and rough animation is determined, to be used as a blueprint for planning the final animation of the film.

Layout/previs is all about experimenting and refining the scope and composition of the shots in the film. Because it is usually done with simplified assets and environments, the layout can be produced quickly and efficiently. This early experimentation and planning helps save time and money during production by providing a solid foundation to build on.

The task

From the beginning, Spanner was conceived with numerous goals in mind. It had to reflect the high bar for quality set by our customers, both with its visuals and its storytelling. It needed to be done under “real-world” conditions, which meant a production schedule, budget, and real deadlines, just as our customers would have. And it needed to introduce new characters and a new environment and leave the viewer wanting more…all within 1-2 minutes. While that’s all a tall ask for a 1–2 minute film, it is the same kind of task that our customers face every day.

For layout, the first step in introducing a new world to the audience usually begins with an establishing shot. As an artist who has worked in layout and previs for over 20 years, I think I can say with confidence that every layout/previs artist loves to work on an epic establishing shot. Sweeping vistas…check. Compound camera moves that evoke a sense of wonder…yes, please! Making sure that the shot you create meets the overall storytelling goals for a 1-2 minute film? Well, that’s not always as easy.

The challenge for the opening shots of Spanner was to establish the environment, while at the same time introduce our characters without taking up too much time. Getting to the final shots involved some creative problem solving, along with some compromise and lots of iterations.

Experiments and feedback

The storyboards for Spanner called for a wide establishing shot that showed the environment – the bridge with Noa hanging underneath – and the airship arriving with Ulysse in tow. Our director, Rex Grignon, expressed that he wanted to try something a little more complicated and see if we could get it to work. He described a shot that would start a bit closer on the clouds and the environment, move up to show Noa hanging under the bridge, and finally reveal Ulysse arriving on the airship.

Of course, there are many different ways to accomplish those goals. I started by thinking about what, if any, restrictions I had on the composition and blocking of the shot. Knowing any restrictions would help me narrow down ideas and focus my experiments.

First of all, I knew that I wanted each element (clouds/environment, Noa and the bridge, Ulysse and the airship) to have its own distinct moment, but all of those moments would be tied together with one seamless camera move. I also knew that Noa’s location on the bridge was predetermined: she was to be hanging at the lowest point of the bridge and would not be moving. Lastly, I wanted the end framing of the shot to match up to a similar/same as shot later in the storyboards where Ulysse jumps off the ladder of the airship. So one way or another, the airship needed to end up over the bridge with the camera on its starboard side.

However, that left the starting location of the camera and the airship undetermined, so that’s where I focused my experiments. See the following videos for a few of the different camera moves that I tried, each of which attempted in different ways to meet the goals that I set. (1).gif

Kelvin-Helmholtz instability creates rolling waves when velocities differ in fluids

shot 1 test, v2

shot 1 test, v3

shot 1 test, v4

shot 1 test, v5

Of these tests, Rex liked v5 the most since it had a good balance of strengths. It started with a nice view of the clouds and environment and then showed the precarious nature of Noa hanging on the bridge, but not from so far away that the audience wouldn’t be able to tell that there was a person hanging there. And finally, he found the choice to have the airship turn towards camera to be the most interesting way to introduce it along with Ulysse.

What didn’t quite work was the animation of the airship itself. The motion of the airship was not very realistic, especially at the end, where it seemed to slide sideways into place. So my next set of iterations focused on improving the airship animation and adjusting the camera move accordingly. In addition, the aspect ratio changed from 2.39 to 2.2.

shot 1 adjustments, v1

In the next set of changes, the aspect ratio of our short was changed from 2.2 to 2:1. I continued to refine the camera movement as well. We started out with a wide aspect ratio, with the idea of showcasing the environment. However, since our story is ultimately a conversation between two characters, Rex felt that a narrower ratio might work better. As a result, we chose our final aspect ratio of 2:1.

shot 1 adjustments, v2

At this point, we felt that the camera and airship animation was in a pretty good place in shot 1. However, looking at the opening shots as a whole, there was a feeling that it was taking a long time to get to Noa, our protagonist. Shot 1 did a fine job of introducing our environment and showing the arrival of Ulysse and the airship, but the next shot of Noa (shot 2) is still relatively wide. It’s not until shot 4 that we get to see Noa up close. With only 1-2 minutes to tell our story, it felt like we were waiting a long time before introducing our main character.

At this point, our creative director, Jason Schleifer, had an idea. He tried splitting shot 1 into 2 shots and inserted a close up of Noa in between. After cutting back to the very end of the original shot 1 to see the airship and Ulysse, he cut to a medium shot of Ulysee and back to another closeup of Noa. Just like that, Jason introduced Noa much sooner, while still showing the key beats of Ulysse and the airship arriving. And he was able to do it all in 10 seconds less than the original sequence of shots.

opening shots, edited

The final result

The final version of the opening shots is very similar to Jason’s edit. The medium shot of Ulysse was ultimately cut, but with the time gained from this and Jason’s trims, we had room to extend the beginning of shot 1 to add in the title of our short and to allow for a bit more time to enjoy the scenery before Noa enters frame. I also made a few final adjustments to the camera in shot 1 as well. The director was thrilled with our creative problem-solving to keep the pace flowing and support the story. Win-win!

opening shots, final

The takeaway

While the cinematographer part of me was sad to see the long, sweeping camera move that I worked so hard on get split into two separate shots, the filmmaker part of me knew it was the right call. Ultimately, it’s all about the story, and how to tell the story in the best possible way given the constraints of your project. In the case of Spanner, our main constraint was time. Given a longer story and a lengthier runtime, the original opening shot could have made sense, but with only 1-2 minutes to work with, it was vital to introduce the audience to our main character sooner, rather than later.

Just because you can do something doesn’t always mean that you should. In the end, even the coolest, most complex camera move isn’t doing its job if it is detracting from the telling of your story in the time that you have to tell it. When you keep your focus on the characters and how to tell their story first, often the other pieces of the film making puzzle will naturally fall into place.

Have a question for one of our artists, authors, or compositors on this series? Let us know in the comments or on social media. Once the series is complete we will be posting a roundup of Q&A.

Dave Murphy

Dave Murphy

Dave Murphy has worked in the animation industry for over 20 years as a layout artist, set dressing artist, and supervisor, spending most of that time at DreamWorks Animation. He joined AWS in 2019 as a Pre-Production Supervisor on the FuzzyPixel creative team.