Filmmaking is a lot like solving a puzzle—there are a lot of pieces to put together and getting to the end result is as equally frustrating as it is rewarding. The first thing we work through in solving the “filmmaking puzzle” is the concept. After settling on what direction to head towards the question becomes, “How exactly do we pull this off?” And because we’re always striving to push our creative boundaries, the next question that immediately comes into play is, “Can we pull this off?” Oddly enough, when that feeling of the unknown arises, we’re usually onto something good. Now before we break down all of the pieces involved for our latest Uncommon production, we recommend watching the two finished commercial spots below. Bonus points if you can figure out how we made the magic happen on these pieces before reading on and watching the Behind the Scenes featurette at the end of this post!

The rest of this post can get a bit technical.  Fear not, you definitely don’t need to know all the gear terminology to follow along! :)

Phones moving in space. No CG. Those were the parameters we set for ourselves. The most obvious thought in bringing these parameters to life was to use a turntable. The limitations of that approach, however, were also just as obvious. A turntable can only rotate along a single axis and we would be severely limited in the number of angles if we didn’t want to include the surface of the turntable. Did we like this idea? No. Did we buy a turntable anyway? We sure did! It never hurts to have fallback plans.

Since the turntable option was, well, off the table, we turned our attention to camera motion control rigs. At the time, Kessler had just released their CineDrive motion control system. If we could program the camera to move, tilt, and pan, then we could potentially get the shots we needed.

Meet the Kessler CineDrive

The pan and tilt head paired with a slider and a focus motor gave us 4 axis/variables of control. The challenge was that the slider was only capable of lateral movement, unless we rigged the slider to a motion controlled crane to give it vertical movement. But even then, it still wouldn’t be capable of circular movement. Did we think the idea would pan out? We weren’t sure, but it was worth a shot. It helped that LensProToGo was/is renting them out. We snatched it up along with a Kessler CineSlider.

The choice of camera was a fairly straightforward one as we wanted the most resolution we could get so that we could crop into the image for creative framing. The RED Epic, capable of 5K images, made a lot of sense to use on this production. The ability to shoot high frame rates was an added bonus in case we needed to smooth out any movement with slow motion. When everything came in, we mounted the RED Epic to the Kessler Cinedrive, threw on a macro lens, and took it for a spin!

RED Epic camera on the CineDrive

Using a combination of a slider and a panning head to simulate a camera circling the phone meant that we had to program the slider to move fairly quickly. Though the CineDrive system handled the quick moves like a champ, the problem with shooting in macro was that every little movement is amplified. We found that the combination of the speed and the large payload of a RED Epic caused too much vibration in the image. Furthermore, the restriction in terms of the types of shots we were able to achieve was still a problem.

The next idea we had was placing the camera on a tripod and somehow using motion control for the phone. This sounded promising and was something we had considered before. Would it work? We had no idea since we couldn’t find anyone who had done anything similar before! 

The real challenge here was finding a solution to clamp the phone to the pan and tilt head in a way that was solid enough to minimize vibrations in the system, yet also flexible enough to allow for a wide variety of moves while having necessary clearance for free movement. After head scratching and headache inducing few hours, we started with a crazy system of Mafer clamps, pipe, Cardellini clamps, and a grip head.

Mafer clamps, pipe, Cardellini clamps, and a grip head

Mafer clamps, pipe, Cardellini clamps, grip heads, oh my!

The system was solid, if bulky, and very difficult to adjust. Moving the phone became a complex puzzle of adjusting this clamp and that clamp, and all in all, it was a bit of a nightmare. The breakthrough for us was using a Noga cinema arm. Suddenly, we had infinite adjustability at the twist of a single knob!

Enter the Noga cinema arm

Enter the Noga cinema arm!

If only we were done here! The next real challenge was centering the phone exactly along the X and Z rotation axis so that the phone rotated instead of moving in a circle like a Merry Go Round.

Rotation axis

Rotation axis

From the diagram, you can see that the intersection of the two axis is floating in free space. It turns out that judging it by eye was surprisingly difficult! The infinite adjustability of the Noga arm was a blessing and a curse here—the flexibility was great but if we only wanted to move the phone over by a hair, once we loosened the knob the whole system went limp. It became a tedious process of re-adjusting and testing. Clearly, this wasn’t acceptable! ENTER LASERS.

Laser guides

Laser beams!

We used a system of lasers to be able to visibly see where the center was and our success rate went up considerably from there. Finally, at this point, the motion portion of the puzzle was solved! The next arena to tackle was lighting. We knew that as the phone turned, we wanted to get a lot of specular reflections to help the shots come alive. We also knew that we wanted the lights to be shaped like bars.  We did some rough calculations regarding family of angles, angular velocity… wait, no we didn’t.  We just figured we needed lights, and we needed lots of them!  Cheap fluorescent fixtures fit the bill perfectly.

We picked up eight fixtures from Home Depot and wired them so that they could be plugged into a standard AC outlet. Fluorescent lights aren’t exactly known for having a smooth, continuous color spectrum so we were a little worried about the footage looking a bit green. Thankfully, because this was for a product shoot, it wasn’t too big of an issue since we didn’t have a lot of skin tones to worry about.

Flourescent Lights

Building a time machine

The glorious light.

Vibrations aside, due to the macro nature of the shoot, we had to work very hard to manage unwanted reflections. We used copious amounts of black gaffers tape to tape over all the shiny metallic C-stands, clamps, and anything else that decided to show up. Reflections came from the most surprising places and sometimes we would have no idea where they were coming from at all!

Gaff tape everywhere!

Random black gaff tape everywhere!

The other challenge was managing tiny dust particles and scratches barely visible to the naked eye. We solved most of our dust problems by wiping things down with lens cloths and removing dust particles with medical tape. Other problems like scratches, blemishes, and even digital text were just too persistent and required composting work in post production.

Digital cloning - Mom calling at 6AM?

Before and after some digital painting – Mom calling at 6AM?

If you’ve made it this far and you’ve correctly predicted how the two commercial spots were produced, then congratulations! You’re either a master puzzle solver, a genius, or you were hiding in the shadows during our production! If you cheated, and just scrolled all the way down from the top just to watch the BTS featurette, that’s cool—BUT NO BONUS POINTS FOR YOU! At any rate, we hope you enjoyed this inside look at how much work went into all of the moves, turns, drops, and splashes. And, though frustrating and confounding at times, it was certainly rewarding finding out that the answer to our questions was YES. We could indeed pull off these concepts.