Custom solutions behind the cinematography of F1 The Movie

With F1 The Movie, director Joseph Kosinski felt the need for speed. More specifically, the need to capture the speed of Formula 1 racing on camera and showcase it on screens up to and including IMAX scale, effectively placing audiences in the cockpit of F1 cars and immersing them in real race environments with unprecedented authenticity. Naturally, Kosinski turned to cinematographer Claudio Miranda ASC, who’s been behind the camera for all of the director’s features to date, to put the wheels in motion. Miranda quickly brought in 1st AC Dan Ming — a veteran of Kosinski’s Top Gun: Maverick whose credits with Miranda also include Tomorrowland and Life of Pi — and began reaching out to equipment and service providers who would be crucial to realizing the unique solutions the production required.

Among those solutions, Sony was tasked with developing a new, super-compact prototype camera - nicknamed "Carmen" - that would be mounted via compact pan-heads that were custom manufactured by Panavision, who also provided a brand-new Small Array plate, custom electronics, and more. The Mercedes-AMG F1 team were also critical to the collaboration, and RF Film, Wireless Wizards and Hornets Tech were instrumental in adapting active F1 racetracks into workable locations for the production crew.

Panavision recently caught up with Miranda and Ming for their detailed breakdown of the components that enabled F1 The Movie’s thrilling cinematography to take audiences on an unforgettable ride.

Panavision: Claudio, how did Joseph Kosinski first describe the movie?

Claudio Miranda ASC: He described it as wanting to put cameras onto cars and record the whole thing at speed, at 180-plus miles an hour. We watched Grand Prix together and loved some of the footage in there. There were some interesting high angles over the car that I really responded to — it's interesting because usually you think the lower angle will look faster, but because the car is going 200 miles an hour, even at that height, it still looks fast. So that influenced some of the angles that we wanted to do. 

I said, 'Joe, as soon as you know this thing is green-lit, let's talk to Mercedes about where we can put the mounts. Let's get cameras engineered.' And then it was a conversation with Panavision and Sony - you guys built the panners and Sony built the cameras. It was interesting because you were separated by thousands of miles, so you didn't actually see what they were designing, you just had to trust the measurements and the data to build the panners. I remember when I was at Panavision, and both slid together, and it was perfect. It was great to see that come together.

Dan, when did Claudio first mention the project to you?

Dan Ming: It was around December of 2022 when he said, ‘We're doing this film, and we have some things to build, starting with the camera, and then we need the heads. And we want to be able to pan and focus, and do it anywhere on the track, and maybe have up to four cars going at a time.’ 

Panavision was always at the beginning of the conversation because we knew we were going to be going all around the world - we needed a company that had that infrastructure to cover us wherever we went. We also needed to do some pretty major manufacturing, and we thought we could leverage Panavision's manufacturing expertise to help fill the holes in all the things we were trying to put together.

How did the idea for the compact cameras and pan heads initially come about?

Miranda: I talked to Joe, and we were like, 'F1's a race where you don't really need to tilt.' So we figured, 'Let's build a head, and to keep it small and light and simple, all it needs to do is pan. And we don't need to do it everywhere; we just do it at a couple locations on the car.’

Mercedes sent me the model of what the car was going to be, and then, working in 3D, I placed the cameras, obviously with Joe's approval. It was months of going back and forth - 'This is what we can do. You can't go here. This is a good angle here. The mount has to go a little bit more to the left.' It was a critical collaboration between Mercedes, Panavision and Sony to get all the data together and make the whole system work.

We had to make sure the cameras were going to work, that the pans were going to work, and then we integrated the wireless. I had the idea early on to make it a Preston-based system. I drew something for Dan, roughing-in a version of what I thought it would be, and then Dan had an idea, so I drew something else, and then we had Panavision build it from there.

Ming: We knew RF Film and ‘Noodles’ [Greg Johnson, RF Film’s president] had built serial-to-IP with Prestons so they could extend the coverage of Preston handsets. So we thought, ‘All right, if we use a Preston system to do focus, then since we have those motor drivers, why don't we also use that to operate the pan so we don't have to re-engineer another thing to drive all the motors?’ Based on that, Claudio then sent me a drawing of a pan head with a Preston motor on it, and we started with that as a prototype.

Once Claudio and Sony locked down the form factor of the camera, then we were talking to Sony about making sure to have two 1/4-20 holes on either side of the center of the sensor as mounting points. That way, we could put the camera right into the bail [of the pan head] and use that as our tilt adjustment mechanism without having to put a camera in a cage. So they built that in, and we locked that shape.

Based on that, they sent us a 3D model that Panavision 3D-printed, and then we started building the bail around that. Charlie [Hagmeier, Vice President of R&D and Engineering at Panavision Woodland Hills] and Steve [Boyce, Mechanical Engineering Manager] and Vince [Catlin, Senior Mechanical Engineer] took the lead on that and put together something solid.

Our car package was 20 Carmens, and we had to build all the heads [for each]. With [Panavision Woodland Hills Electronics and FA Manager] Juan Mosqueda and his group, Panavision also built the electronics for power distribution. The Carmens took 19.5 volts, so they had to build brand new boards to power them. We also built bridges so we could link the plates together, so that if one went down or you had a change of battery, they would automatically shift and cross-feed each other so the whole car would stay hot.

There were a lot of little details, like the lenses we were using [for the Carmen cameras] weren't really built to have focus gears, so Panavision manufactured our own focus rings as well as brackets to mount the recorders in the car. Pretty much all the things we needed to put everything together, Panavision helped manufacture.

What lenses did you use with the Carmen cameras?

Miranda: Because of the nature of us using our smaller cameras, which had to be really tiny, we used Voigtländers. I couldn't use any sort of anamorphic — it would just be too big — so we stayed spherical for the whole movie, and we used Master Primes and Fuji zooms for the large part of it with the Venice 2. We were also tying our footage with broadcast footage, so I couldn't get too crazy with lenses.

Ming: We had the main-unit package, which was Venice 2s, and then we had [DJI] Ronin 4Ds, and we also had the F1 footage unit that DIT Chris Nunn ran, which went to every race and mounted the wing cameras that Apple built, which was another thing that early on we were involved with. The F1 footage team was also capturing certain camera angles, uncompressed, at each race from the broadcast side. Those were all the units that were working at the same time.

When did you know that everything you were planning was going to work?

Ming: We were testing at Willow Springs in April of 2023, and that was when we had the first prototypes of everything. So it was between December and the beginning of April where we had to pull everything together, and that first test was when we were like, ‘Okay, this could actually work.’ 

Then it was about, ‘How do we make this work,’ because the test in April was just one car and two prototypes. We were testing various platforms, testing how the communication protocol would work. Willow Springs is a pretty open track, so it was pretty easy to maintain connection, but every country has its own challenges and rules about what frequency you can use. So we still had a lot to do, but at least we were like, ‘Okay, the bones of this are going to work.’

What were some of the further refinements that happened after that initial test?

Miranda: Once we got the mount, there was an issue with vibration. We wanted a little bit of vibration, but we also wanted to be able to see the actors, so the grips found this material called Sorbothane, and they would use different levels underneath the cameras to kill some of the harmonics. That was a great material. We had different levels for different positions and different tracks. We had to run the track with the mounts, and if there was a little bit of vibration, we had to figure out, ‘Are we too stiff or too soft?’

Ming: The Mercedes-AMG F1 team also said, ‘You should put shrouds on everything and protect your gears. There's so much rock and debris, you have a good chance of something getting lodged in your gears and jamming up your heads.’ So based on that feedback, the design was updated to cover the gears and protect the motor.

Because it's a Preston motor, you set your limits, and initially it was a 360 [degree] limit. When we were testing it, Claudio was like, 'No matter where we are, I always seem to hit the limit.' So we came up with a rocker inside the head that allowed us to do two revolutions, 720 [degrees]. So the Preston motor would calibrate for two spins each way. You still had to be careful how you oriented the head when you installed it so your sweet spot was forward, but you had a lot of extra space to spin in case you needed to go a little farther than originally planned.

How many of these panning cameras did you ultimately have on a car at one time?

Miranda: We mainly only had two at a time, but sometimes we'd do two pan heads and a lock-off, or we would have them on other cars. There are some angles from the top, those are lock-off. There's a high-front looking back, that's a lock-off. There's a front nose mount that was originally supposed to be a lock-off, but then we liked panning it, so we made that a panner. There's a rear mount that was also a panner — that was tricky because it gets really hot back there, and there's a lot of vibration.

You also have to work with the frequency you're allowed to put in an F1 race, because you don't want to step on Lewis Hamilton's frequency and jam all his telemetry. So everything had to be approved, and then we had to get coverage around the track. With RF Film, we put up a whole mesh network around each track, and then we would control it from our garage, our 'mission control.' We would have two to four operators operating the pans, and we would have focus pullers on some of them, because sometimes you're close on Brad [Pitt], and then you have to pan out and pull focus.

Were there ever moments where you thought, ‘I don't know if we're going to pull this off’?

Ming: It's kind of not an option, right? But everyone was so collaborative. Sony had the cameras. Panavision was doing the heads, the electronics. RF Film was working on the network and the mesh system for the Preston interface and network, so that took care of pan and focus. Wireless Wizards and Hornets Tech in the U.K. were dealing with the picture side, with the 4K transmission so we could see the image. Sam Phillips, our key grip, and his car riggers were in close contact with the Mercedes-AMG F1 team to make sure the heads would interface with the vehicle mounting points, and they were sorting out vibration dampening. Meanwhile, Team Mercedes was getting component parts and sizes of the system to make sure the electronics would all fit internally and we would be able to run our cables as well. So all the pieces were in place, and everyone was doing their part and was aware of how their part needed to plug into everyone else’s system. It was complicated, but it didn't seem impossible.

The thing with big projects is breaking it into sizable chunks and then making sure people who know what they're doing are working on each chunk, but in a cohesive way. It's the same when you're at a race and you have all these sequences to capture that are happening at the same time in different places. You put together a very detailed schedule, and each team has got their column in the schedule, and you just see how you fit in the overall picture. You show up where you need to show up with the gear that you need, and you just trust the schedule. And then you look at your watch and go, ‘Okay, I have five minutes, 26 seconds to get to the next spot,’ and you just go. And then occasionally, within that structure, there'll be pivots, like a team principal's available for a quick cameo. When that would pop up, we would look at the schedule and see who could shoot off and do that and come back and continue on with the original schedule.

I understand that during prep, you saw one of Panavision’s Large Array plates, which led to the development of the Small Array plate specifically for F1.

Ming: Yeah, we had an array unit, and we wanted to build our own array system that we could break down and set up ourselves and ship anywhere in the world. We were trying to make it as tiny as possible because the cars we needed to put them on had to be really fast and low. And so, with Guy [McVicker, Technical Marketing Director at Panavision Woodland Hills] and the engineering team, we put together the mini plates, and [camera assistants] Phill Hardy and Rich Daws put together the rest of the system in the U.K.

We built two plates, and we had three RED Komodos forward, four in the back and one looking straight up as a lighting reference. It was a great little array system that we could just put on with the camera grips and drive it on the track. We would shoot as much as we could during the races, but once the race is done, they’re breaking everything down and the crowds are gone, and we would still have more to shoot on the tracks. With the arrays, we could capture the signage and the crowd, and VFX could put it all back in later when we had to drive on emptier tracks.

Miranda: We tried to get all our major shots between slots for qualifying and then the race, where we had these little 10-minute windows, but we also did a lot of stunt work, and sometimes we weren’t on normal F1 times, so VFX used some of those scans for crowd replacements. It was a lot of coordination. There was a lot going on.

Was it a relief when the schedule would give you an interior dialogue scene? Or did you thrive in the pressure of those 10-minute windows on a race day?

Miranda: Sometimes it is nice to take a break and have a little bit of a quiet moment, doing a nice intimate scene with Brad and Kerry [Condon], and they were great. It's nice to have those moments and do normal dolly work, less mission control, and you don't have to be small and out of everyone's way. You could have a crane if you wanted to.

Ming: The big race environment is fun, but the scenes that were calmer gave us time to do other things, like our manifests or cleaning gear that had gotten dirty from a race. There was always maintenance to do on top of the actual shooting. Also, we were shipping all around the world, so Maggie [Stanaszek] and Georgina [Cook], our central loaders, were always working on a carnet. It was a lot to constantly keep an eye on to make sure the whole thing kept going smoothly. Nathan [Weiner, Logistics & RTR Storeroom Manager at Panavision Woodland Hills] and Carson [McCabe, Logistics Lead at Panavision Woodland Hills] really helped us out with that as well.

It was also great working with [Sales & Marketing Representative] Amanda Sasaki and [Sales & Marketing Assistant] Jose Mosqueda here in the States, and then [Account Executive] Harriet Cannon in the U.K. and [Technical Director] Kevin Greene and his team. Wherever we went, we could always call someone with Panavision.

Claudio, you've worked with Joseph Kosinski on all of his movies to date. How has that collaboration evolved over the years that you've been working together?

Miranda: Well, a lot of times we're talking about the next movie on the previous movie, so I kind of have a leg up. We have a really great shorthand, and we just get each other in a certain way. What I love about working with Joe is each of our movies is different. Tron's different than Oblivion, which is different than Only the Brave, which is different than Top Gun. Maybe there's a similarity in Top Gun to F1 — that ‘speed’ kind of movie — but I like that we're always trying to do something different, even though it might have similar tones. 

And Dan, how has your collaboration with Claudio evolved across the projects you’ve done together?

Ming: I love it because Claudio trusts me so much, and I love how we think. We're very compatible because he'll come up with the idea, but it's not just pie in the sky. It's like, ‘I came up with this idea, I went in Studio 4D and built the whole thing in a virtual environment, and here's the model of the thing that we need and the exact dimensions it has to be.’ He comes up with big ideas, but he's very specific with the ideas, so you have something solid to work with, which really suits anyone. He's great with communicating complicated ideas — not just to me, but to other departments. The way he thinks could be complicated, but the result usually is to make something as realistic and natural as possible.

You each keep things exciting with the projects you work on and the challenges they present. What keeps you inspired from one job to the next?

Ming: I actually have a degree in engineering from UCLA, but my best friend, Aaron, went to USC Film School, and I helped him out on all his student films. On the weekends, I would do commercials and music videos, and over the summer I would do three six-day-week movies where I would load for free for the experience because it was just way more fun than being an engineer. So by the time I graduated, I was already in the union and working, and I never looked back on the engineering side. But the truth is, I use it every day. It influences the way I think and the way I approach problems or any situation. 

The reason why I love this business is that every job is different. You're in a different place, different environment. It's not always pleasant, but it's always a challenge, and it always keeps you thinking, and you never stop learning. I'm naturally curious, and it's very hard to not be interested if you're curious and open to learning new things. So that's what I love the most, and also all the people you get to work with from all different walks of life. The variety just makes it so interesting every day.

Miranda: I like solving problems, and maybe that's where my strengths are. Someone tells me there's a problem, and I go, ‘Let's see how many people I can get around to help me solve this problem.’ So I think one of my strengths — and with my team all included, it's not just me — is getting everyone on board about, ‘Hey, let's try this new approach, and let's see how we can make this.’ And then people like saying ‘No, you can't do that,’ or ‘That'll never happen.’ And then it’s like, ‘Well, why not?’ I think it's fun.

Unit photography by Scott Garfield. All images courtesy of Warner Bros. Pictures / Apple Original Films.