Thoughts on the Cinematic Image

Hey folks - Freddie here. This Patreon affords me the opportunity to pour myself a tall glass of iced tea and just straight up pontificate upon my cinematic thoughts, which is very exciting (as usually these thoughts are reserved for long-winded conversations with Matt). I hope this will provide elucidating insight into the thinking behind our creative choices.

An early consideration when thinking about a film of any length is, "What is the look?" This can start out fairly broad - is the camera handheld or supported? Color or black and white? Wide angle like Malick or tight and long-lensy like Heat? And so on.

These properties are, I think, best understood in the service of the story you're trying to tell, and whether the qualities of the image feel like they add to or detract from the story. I use the phrase "feels like" purposefully, as this analysis starts to get pretty theoretical pretty fast as there is no hard scientific evidence (usually) for the efficacy of certain cinematic techniques over the other.

For example, I've heard it argued both ways whether shooting "dirty overs" (i.e. a shot focused on one of the interlocutors in a dialog that is "dirtied" by the occlusion of the body of the person they're yapping to) makes the dialogue scene feel more intimate or more distancing. Do you feel "closer" when some part of your brain is aware of the proximal physicality of the players in a space? Or does that occlusion instead remind us subconsciously that we are outside of these characters?

How does the presence of the body on the right inform this shot from The Shape of Water? What's the difference if the shot was clean?

The answer is, more often than not, "it depends."

I am reminded that, in music, the knowledge taught to children is that major harmonies sound "happy" and minor harmonies "sad." Yet Hava Nagila is in minor and the original Jewish dance floor banger, and Taps, the lonesome funeral melody in cartoons and Oscar-winning features alike, is blown in major. The shot, of course, doesn't exist in a vacuum, and many aspects (maybe most salient - what the ding dang actors are doing) inform how we, the audience, feel.

But regardless, it's fun to think about what a thing should look like. If making Spotify playlists for your fictional character is the ultimate procrastination tactic of choice for the screenwriter, then surely "Browsing ShotDeck for cool image inspiration" must be for the director.

Eventually, you do need to answer some pertinent technical questions, namely: What camera and what lenses are you going to shoot on?

When we interviewed DPs for We're All Gonna Die, we would meet up at a coffee shop after sending along the script. The DPs would all have a nice little presentation they put together of image comps (an insight from Jon Salmon, a frequent collaborator - as a DP, the process of building these decks is always a bit of a pain because you're trying to guess what the directors want to see), and the conversation would invariably steer to us asking "what do you, after reading the script, think we should shoot the picture on?"

The answer was nearly always "The Alexa or Alexa Mini," with the occasional hedged "Obviously, in a perfect scenario, we'd shoot on the Alexa." Bongani Mlambo was the only one who took a serious, considered beat to think about and suggest alternatives to the camera platform I suggested - the (considerably cheaper) Canon C300/C70. He got the job and shot the hell out of that movie.

The Canon line was the subject of my focus owing to a late-night bike ride I took with Niko, of Corridor Digital, to the top of a mountain trail in Glendale (we rode e-bikes, as I'm not doing off-road hill climbs at night in my mid 30s). I wanted to ask his opinion on camera platforms, as his words have a lot of weight because I figure the folks at Corridor are about the only people out there building honest work-a-day RIGS to accommodate their intense production needs, and as such, these rigs are not designed to look cool in a thumbnail for some gear recommendation video.

Looking out across the glittering lights of the Glendale/Burbank expanse, I said "We're probably going to have limited ability to light at night, so I need good lowlight sensitivity, and we're probably going to have limited support, so I need good battery life." Niko said that he was a fan of the Canon series.

I admit I did not expect that. I found that I, too, resorted to the common industry understanding of Canons in a cinematic context - i.e. they are NOT cameras for "Real Movies," but instead are looked upon as reliable documentary workhorse cameras, owing to their phenomenal battery life, their second-only-to-Sony autofocus speed and accuracy, and relative portability and friendly solo operator ergonomics. Canon's imaging also has the reputation of " having good skin tone rendition." But rarely is the C300 mentioned in the context of narrative cinema work, aside from that Saulnier shot Blue Ruin on one.

But the more I stewed on it, the more Niko's suggestion made sense for WAGD.

After all, everyone and their mom wants to shoot on the Alexa. But what is the Alexa? Dynamic range that reaches deep into the lowlights and highlights equally (from a distantly remembered technical demo that Abel Cinetech did: The Alexa had the most reach into lows and highs, Blackmagic's cameras did just about as well in the highlights but suffered in lowlight sensitivity, and the inverse was true for the REDs (This was an old test, I should note).

If the Alexa represents the industry consensus "best" image (or at the very least, the image that all these DPs seem to love), it would follow that in the professional world, all of Alexa's competitors would be competing to deliver as-close-to-Alexa performance at a lower cost. That is to say, everyone in the pro camera industry is probably converging on, if not exceeding, the same technical benchmarks of the Alexa - dynamic range, rolling shutter, resolution, bandwidth, etc.

And on paper, what the C300 (specifically the Mk. 3) was giving us was plenty - 4:2:2 10-bit, dual gain, plenty of dynamic range, built-in ND filters, etc. But most importantly, for We're All Gonna Die, its reputation as a documentary camera was a benefit. Our story is set, according to our script, in a "familiar, but changed America" post-arrival of a world-altering alien object in our upper stratosphere. Our goal was to shoot it loose, and to have the kinds of images you might get on a road trip are harder to capture if you're fumbling around with a hefty cinema rig. Our images, in other words, would benefit from the physical form factor and advantages of the C300. Sure, you can argue that you may run into scenarios where the Alexa would excel, but in doing so, you ignore all the shots and moments you might miss if your camera is a pain in the ass to get running.

In the end, I was extremely satisfied with our choice. When talking about the movie with film folks, I'll often ask them what camera do they think we shot it on. To date, nobody's ever guessed the C300; most people assume we used the Alexa Mini. The C300 also had the benefit of the same image sensor in the C70 (which we bought as a company, allowing us to not rely on camera rentals for when we went to shoot pickups). Moreover, our ability to B-Cam on an identical image sensor in a smaller package meant no headaches in matching our two cameras later (a headache we experienced on VGHS Season 1). And here, a lesson we were able to reapply - a lightweight handheld B-Cam meant we could snipe opportune angles and shots within our setups.

Here I am in a river with waders to grab a crucial side-profile shot of Jordan Rodrigues as Kai. The blocking demanded this camera be positioned a foot or so above the water line to get the angle.

See also this Bongani-helmed B-Cam angle that gave us, in the edit, a subtle humorous textural shift when Kai's flirty joke fails to land.

I think we also converted Bongani, who expressed a newfound appreciation for the C300. He has used it on further projects, and has also reported that the bias against these cameras by "real" filmmakers is still going strong.

Which takes us now to Nail House. Here, I want to dive a little deeper. Like WAGD, there are questions of ergonomics when you're shooting a highly choreographed, physical, hand-to-hand beat-ass movie, as portability and weight are important factors. Years back, I remember hearing that Gareth Evans employed the Manfrotto Fig Rig on The Raid films after seeing it on our behind-the-scenes YouTube videos.

The Fig Rig removed the high-frequency jitter you get when your hands are tight on the camera body. You're able to get rapid and accurate pans by punching one of your arms out while holding it. The trade-off is it's annoying as hell to set down, as the rig suspends the camera inside an aluminum circle (the rig, which hangs in my entry hallway, is referred to as "the steering wheel" on shoots).

Unsupported handheld rigs like the Fig Rig are also a major pain in the ass the heavier your camera gets (the same is true of gimbals, of which I have a lot more to say). When we ran RocketJump's RED Epic with a fat prime and follow focus, the rig became more unwieldy. It needs to be set down between takes, and consequently, you add time and energy expenditure just picking it up and looking for a spot to set it down. The solution that the industry seems to gravitate towards is to mech out the camera operator with an EasyRig (which having now used it on a number of occasions with different rigs, I am not a fan of).

I have a more radical solution - make the camera package smaller and lighter.

And when it comes to that, we now have to consider the Sony line of cameras. Much has been made of Gareth Edwards’s use of the FX3 on The Creator (lotta Gareths in this post huh) and while the interviews are a good read, these articles also have a strong whiff of "technological gee whiz PR thing to promote the movie" (see also: the infrared camera rig they used for day-for-night on Nope, which I am thoroughly unconvinced was actually useful).

Very few cameras in my life have actually impressed me. Most recently it's been the Sony ZV-E1, which I picked up knowing we would be doing more video for Dungeons and Daddies. I had completely missed out on the A7 train, having used a first-gen A7s and been impressed if not annoyed by Sony's menu system. Sony billed the ZV-E1 as a next-gen "vlogging" camera, but a glance at the specs revealed a hint of professional pedigree - full frame sensor shared with the A7s III and FX3 and 4:2:2 10-bit recording.

And once again, what was on paper held up to my real-world experience. The Sony slog footage grades extremely well and the larger battery pack solved my issues with the early A7 batteries. Lack of the internal ND was maybe the only thing missing from my wishlist (the discretely stepped C300/C70 NDs were immensely useful, but I would imagine the variable NDs on the Sony cameras, where you can park an f-stop and shutter speed and let the electronic ND take care of the rest, would be even more convenient).

On the pro end, Sony continues to push the Venice (and Burano) and has been, at least from informal conversations with DPs, making some inroads against Alexa's superiority. This gives Sony a nice reputation boost, which means the DPs we talk to now at least won't think I'm a giant moron for even suggesting a Canon camera. On the image end, most recently I watched Wim Wender's Perfect Days, which was shot on the Sony Venice 2, and loved the look which, for me, seemed to embrace a clean digital aesthetic and reject the retro film look.

Preliminary, off the strength of the ZV-E1, the FX3/FX30/FX6 has elevated to the top of my list, but there's still a wide field to explore. One of the things we're planning, as research (which we will show!), is visiting actual Chinese restaurants and rolling the ZV-E1 to get a sense of how the camera behaves in the space, and importantly - what kind of lensing we should be considering.

In summary - I believe the creative latitude afforded by post-color correction and the technological convergence on the salient aspects of the Alexa image by all pro camera manufacturers means that the question of camera body should be focused on ergonomics rather than on image quality (because in my experience, unless you're A/Bing the images, you're going to be hard pressed to identify one camera over the other, and if us pros have difficulty doing it, then assuredly the viewing public, the majority of which do not care about motion smoothing on their flatscreens, doesn't stand a chance).

Lenses are the other crucial component here, and I have much to say, but this post is already getting long in the tooth. Next time!

-fw