Canon Cinema Primes vs. Canon L Series from Jonathan Yi on Vimeo. Sooner or…
Editors Note: Here we have the second installment in how a professional production is conducted using the Atomos Ninja 2 field recorder, told from the first hand experience of Sean Ware, the DP as he shot it. It is rare to get a look “inside” another’s head and hear the thoughts that pass through as a new piece of gear is put through the paces of a professional workflow. This week Sean and his crew have been kind enough to provide us with some stills pulled directly from the ungraded footage. I’ve resized them and inserted them between the paragraphs of Sean’s text. I hope you will find these useful reference for your own projects, as this is how exposure and tonal range are controlled by professionals for each of this type of shot setup. Once again, may thanks to Sean and his crew for allowing us to share in this exciting project!
More on the Technicolor Cinestyle color profile. It is designed to shoot the footage specifically for post production editing. It is NOT representative of how the final will look in any way. Saturation and contrast are both flattened for recording, and then the dynamic range is reconstructed and expanded on both in post. For further information, and to download the Canon EOS Technicolor Cinestyle profile for yourself, it can be found here: https://www.technicolorcinestyle.com/download/
– Chuck Jones, Editor
Shoot Day 2 –
Our second day of shooting took us to a couple of spots in the deep valley. For those of you who aren’t familiar with the San Fernando Valley, or Los Angeles for that matter, the “deep valley” refers to a part of Los Angeles County that has an incredibly diverse populous of ethnicities and culture. It’s a place where you can find both cheap (and amazing) mexican food and hand-crafted furniture, sometimes within the same building. It’s also a place that has an interesting aesthetic to it once the sun sets and the street lamps turn on. It’s beautiful, really. The smells and sounds of the city come to life and the sprawling network of streets is dotted with store-front neon and food truck patrons. Not the gourmet, grilled-cheese food truck variety, but the trucks that feed the families of parents that work several jobs a day.
The locations we picked to shoot the next couple of scenes were right off one of the main boulevards, which posed some obvious concerns. Most notably, it meant that the Director and I had to be very much in sync as to what we were shooting, as the first scene on the schedule took place at magic hour, and we’d only have about 20 minutes to get the scene covered. Due to scheduling conflicts, I had not been able to join the Director on the location scout prior to the shooting day, and while I had seen location photos, I knew that I would have to be quick on my feet with understanding the blocking of the scene. We arrived to the location a few minutes early so that the Director and I could walk through the shot list, so we broke out the Ninja 2 from its [awesome] case and went to work.
As I had previously mentioned, I use my Zacuto viewfinder regularly when operating. And while the Ninja 2 is ultimately rather light for a monitor / recorder unit, it stills adds uncomfortable weight onto my handheld rig. The most successful configuration for viewing the Ninja 2 unit (for this particular project) was briefly noted in my last post. It was the moment I didn’t have time to mount it to my rig but rather gave it my Director to use as a handheld monitor. It took the strain off of both of us as we were both able to focus solely on our respective tasks: me, operating and the Director, well, directing. With one simple task added to his workflow (pressing record on the Ninja 2 unit), it was a configuration that just kind of made sense for this project.
I am particularly pleased with how quickly I can have the Ninja 2 hooked up and running now. The only bummer about the new configuration are the dangling cables that give every Assistant and Operator a small aneurism. However, in the case of the Ninja 2, it’s only two thin cables that need to be accounted for and I was easily able to tie them together to create a low-profile umbilical between the camera and the recorder. It really wasn’t an issue. Besides, how could that be done away with? Wireless recording to the unit? Wait a minute… er, nevermind.
Remember that part in my last post where i said that I was interested in using the unit more as a monitor for this night shoot? Yeah, so that didn’t happen. The rig I’ve built is truly optimized for operating with the aid of a viewfinder. Whether it’s day or night, my rig lends itself to be best utilized as an extension of the eyecup. Focusing becomes “relatively” easy at this point, and I’m able to limit my field of view to what’s actually being recorded, rather than seeing any outside distractions happening off the borders of a field monitor. It has a lot to do with my style of handheld operating. And to be honest, it’s how I learned. I learned how to operate on cameras like the ARRI 2C and the Bolex H16R. Even though I’m surrounded by LCD flip-outs and “LiveView,’ I’d much rather bury my eye in a viewfinder and not be distracted by any outside influence. And I suppose that’s why I’ve never jumped at the chance to start operating off of a monitor. That said, any time I’m on sticks, you can sure as hell find me looking at and using an on-board monitor. It’s especially helpful to check lighting, composition, etc. And it’s always undoubtedly important to look at a monitor with the Director and talk about the shot. And that’s exactly what I’ve loved about having the Ninja 2 unit on set, whether we’re shooting on sticks or handheld. The Director is seeing the shot as it happens, uninterrupted. And when we’ve cut, he’s able to give me instant feedback. “But Sean, that’s not any different than any other film set!” That’s true, and it’s an important point to make. But here’s where it applies to this situation; The Director is able to hold / view the monitor. No big deal, right? But he’s also holding the recorder! And it’s compact, light and he/she can easily operate the monitor if needed as well! Still not a big deal, right? But the recorder lets you record 10-bit, pro-res 422 HQ, nle-ready files on the fly! If you’re still doubting this system, you’re missing the point.
For guys like me, who like to shoot things quickly and efficiently, the Ninja 2 is a very helpful tool that makes it possible for smaller crews to capture higher quality content. It’s an increase in efficiency all around.
I’m bummed that I still didn’t get a chance to really test-drive the focus assist, zebra, false color, etc but I’m positive that the opportunity will present itself shortly, as we have a few more shoot days lined up with different shooting scenarios. So I’ll have to report back on that at a later time. My instinct tells me they are going to work just fine and as advertised. However, it wouldn’t be the first (nor last) time I was proven wrong at the hands of camera technology.
The Edit begins –
After the second day of shooting, we decided to start editing the scenes we’d shot. The film lends itself to being done in stages, and this was yet another way we could make sure we were getting exactly what we wanted. It was also a good way to test the ingest of the footage from the Ninja 2.
So, where do I start? Well, I’ll start by saying there’s not much to talk about in terms of ingesting the footage. I’m editing in Adobe Premiere Pro CC. I set up the project, I import the footage and voila. No transcoding, no trouble, it just works. Everything about ingesting the footage makes me smile. Especially the fact that I know I have an image that’s prepped for timing. I haven’t tested this part of the process out yet and I’m still debating which workflow to use for my timing needs, but it’s nice to know that Atomos has already done the heavy lifting in that department, which is to give me a higher quality file to start with. There is some debate whether or not recording an 8-bit, 420 image on a 10-bit, 422 recorder actually does anything. It’s a rather complex, yet ultimately simple process to explain, and the answer is as well. I’ll dive into the specifics a little later on.
I’ve been shooting the Technicolor Cinestyle color profile with my 5D mkIII for this film. I hardly ever shoot this profile, and it’s simple to explain why. Before the newest firmware update for the 5D and the introduction of the Ninja 2, the recorded files were compressed using the H.264 codec. While the codec is a good one, especially when paired with the 5D’s sensor and the high bit-rate, IT’S STILL COMPRESSED. So I tend to be in the camp of believers that ask “why would you want to flatten out an already compressed image to just bring it back later?” In my experience, this process usually introduces the ADDITION of noise [especially] into the mid-tones. By flattening the colors of an already compressed image, you’re setting yourself up for the addition of noise with any aggressive color-timing, in my humble opinion of course. Normally, I use the “Prolost” color profile, which is really just a modified neutral profile that you can set-up yourself as a custom picture profile on your camera. This is also especially helpful in the event that you need to match settings between multiple 5D’s on the fly.
So why have I been shooting Cinestyle for this project? To be honest, I’m not really sure. I kid, I kid… sort of. I kind of figured it would be a good time to test out the functionality and theory of using a flattened, Log-C type color profile on an image that was being recorded at 10-bit, 422. I mean, why not, right? I’ve always loved the IDEA of Cinestyle, it’s just never really made sense to me as a real-world, practical solution. But now it kind of does. The idea of getting a little extra range from a flattened color profile is extremely intriguing now that I’m able to get a pro-res 422 HQ file out of my 5D. I’ve yet to get into timing the short, but the footage I’ve seen thus far looks pretty damn good. Better than what I’m used to seeing? That’s debatable, and to the average eye, the answer is probably no. But to a set of eyes that’s always looking to get as much detail as possible out of every shadow and highlight, maybe. I’m very anxious to get into the guts of the footage and start playing around with color. My instinct tells me that I’ll see a [very] slight difference. But isn’t it the slight differences that make the largest overall changes? If I can get a little something extra out of the Ninja 2 with the use of the Cinestyle color profile, than why wouldn’t I use it? For me, it’s only a theory at this point. Mind you, it’s a theory I’ve fully committed to in the image acquisition of this project, so I’m REALLY hoping it pans out.