Canon 5D Mark III & Atomos Ninja 2 ...finally from Stefan Czech on Vimeo. After…
Editors Note: One of the best things about our move to the Studio City, CA area last fall was the proximity to the action “where the nerds live” in the movie business. They don’t call it Studio City for nothing…. our apartment is surrounded by all major studios within a couple of miles, six major TV network production studios & countless smaller production and post-production houses. Our neighbors all work in the TV or film industry in various creative jobs of one type or another. This is the real deal, the land of multi-million dollar production budgets where dreams are made and hopes are dashed on a daily basis. This is also where the rubber meets the road, where the best of the best teamwork is done by some of the best craftsmen and women in the world, using state-of-the-art gear creating state-of-the-art entertainment. This is why we relocated here, to be as close as possible learning daily by osmosis from all the talented artists.
One such talented artist is a great young guy by the name of Sean Ware. After untold hours of pleading, sniveling, groveling, and generally making a nuisance of myself, Sean has finally lost his mind and has decided to join our rag-tag creative team here at TheCameraForum.Com. Sean’s task will be to help keep us all abreast of the latest technology and to share his insights on how independent Hollywood producers and directors are using tools with technology in crafting their artistic productions. Sean’s stories are going to be more “in-depth” and meaty perspectives of gear, talent, and techniques important to the creation of great content, but told from the perspective of a user, not a product reviewer. Far more than our usual “first glance” or “field reports” these stories will be much longer form content. For the most part these will be multi part stories where my editorial direction is to really dig into the how and why these various tools are best used and do they in fact hold up in real world professional production conditions. My intent is to bring you the view from the other side, the professional movie makers and shakers who have had the budgets to use anything and everything made. No tool is ever perfect for every use, but “the best tool for the purpose” does exist at every budget level. We just need to identify them.
There should be more depth to major production tool reviews than just a couple of hours of artificial testing with a manufacturer loaner. There is a need to answer the the more subtle questions like how does something feel in use over time, how do the first impressions change, does this shiny new tool grow on you, and do the workflow changes required yield sufficient long term benefits to warrant the cost? What are the pros and what are the cons? A long term Torture Test, if you will. We’ve chosen the Atomos Ninja 2 field recorder to use with our Canon 5D Mark III as two of our long form evaluation projects. I’ll let Sean pickup the story from here below. Thank you for reading! – Chuck Jones, Managing Editor.
A little back-story first –
Like many others, I have used Canon DSLR’s for a long time. It’s not that I have anything against Nikon, or any other brand for that matter. But my first real DSLR was the Canon 5D, and the rest was history. I bought some new Canon lenses, some Canon DSLR accessories and it just started to make sense that I would continue down the Canon path. And I must say, I can’t say that I’ve been particularly let down at any step of the way. Sure, every camera manufacturer stumbles at certain times, but can you really blame them? Well, I certainly can’t. There’s no perfection within this guy, just ask my girlfriend.
When I purchased the Canon 5D mkIII last year, I couldn’t have been more excited. No more aliasing? No more moire? Reduced rolling shutter? An even better sensitivity rating from the sensor? All too good to be true, right? And then came the suck… still no uncompressed HDMI output. I mean, seriously? Why had Canon gone through such research and development to produce such a great camera and leave out what many had thought to be the single most asked for, and important feature of a new 5D? There are many theories, and while most of them make complete sense, I don’t spend my time hypothesizing on the “why’s” and “if’s,” for I find it to be too exhausting and well, pointless at the end of the day. The fact is, Canon isn’t going to put everything everyone wants into one single camera. No company ever will, it’s bad business. And while some moan and complain, I work. And more importantly, I find work-arounds. Was I frustrated that I still couldn’t shoot uncompressed video out of my full-frame, Canon DSLR? Sure, but I also had to look at what I was using my DSLR for in the first place.
My background has been primarily spent in post-production and the camera department. I have worked for many years in both areas and it has put me in a very advantageous position for independent filmmaking, as I am able to strategize, develop and adapt workflows that are time and cost-effiecient to the filmmaking process. I have spent the better portion of the last decade working with the newest digital cine-cameras as a Camera Assistant, Operator, and a Director of Photography. I come from a world branded by companies like Arri, Sony, Red, etc.. I was a part of the beta-testing team for the Arri D-20 before it was first released, one of the first users of the Sony F900, and I was one of the first people who got to put their hands on the Red to give feedback on the ergonomics of the design from the perspective of a working Camera Assistant. By no means am I inferring that I’m some sort of digital cinema guru, or even remotely close to it. There is so much to know within this realm of science, and like so many others, I can’t help but feel like I’m constantly grasping to stay afloat with the release of new technology. I have however, taken away a very important lesson from working on the bleeding edge of digital cinema technology for a fair amount of time; “It’s always only almost there.”
Coming from a background of working with such high-end digital cinema tools and accessories, it didn’t take long for the concept of being able to shoot full HD on a DSLR to turn my idea of digital filmmaking upside-down. A camera less than a third the size of a Red? A camera that was already most likely in my camera bag? A digital film camera that I already had lenses for, or could easily (and inexpensively) acquire more of? There was never a question of whether or not I was going to have a vested interest in this revolution, but only to what degree.
My biggest frustration with the DSLR filmmaking world is the camera support infrastructure. Since it’s still a new, and relatively inexpensive world, you inherently get a slew of starter companies (and I use the phrase “starter companies” loosely) making gear that you wouldn’t normally use under the most dire of situations. Companies like Red Rock and Zacuto have made a literal fortune selling DSLR gear that is [in my humble opinion] for the most part, just good enough to get the job done, and nothing more. To be fair, I own some gear by both afore-mentioned companies, and while it has served its purpose dutifully, it still falls into the category of gear that’s usually too expensive for what it is. These companies have definitely found the fine line of pricing that gets you just the tiniest bit aggravated, but not upset enough to not throw down the plastic to pay for it. Anyway, I digress. I’m just angry at myself for recently going in to my camera store to buy a $50 filter and leaving with $1000 worth of new gear. Eh, what’s a camera nerd to do, right? Right?! Somebody please answer me!
Every once in a while, something happens in the DSLR filmmaking world that gets everyone excited. Like, really excited. And that’s exactly what happened when Canon released the news of a firmware update that would [finally] allow uncompressed HD video to be output from the 5D mkIII. The internets went crazy, skeptics were beside themselves with new material and I, I was very intrigued. But only moments after I had gotten caught up in the frenzied news that my camera was finally going to become a real filmmaking tool, I panicked. How in the hell was I going to record this material? I had never really needed an external recorder for my DSLR filmmaking needs, and I was at a loss. I mean, one of the cool things about the compact nature of a DSLR camera is that I could record the material onto a CF card, right? And did I really want to introduce a new field recorder to an already wanky support system, thus negating the appeal of shooting on a DSLR in the first place? Well, my answer if you haven’t guessed already, was no.
But hold on, you say there’s a company that worked closely with Canon (really?) to make recording the new uncompressed material an easy venture? AND it is relatively inexpensively priced? AND it’s a monitor? AND the monitor has focus peaking, zebra’s and other features usually only found in the higher-end monitors? AND it makes the footage stream-lined for NLE’s? It was about this time that I thought to myself, “This is too much [potential] good news for now, I better go shoot something on an ALEXA to get my head straight.”
A friend of mine, Chuck Jones, is a firm believer of getting his hands on testing anything and everything as soon as it comes out. A fellow “bleeding-edge believer,” if you will. Chuck (a long-established photographer / filmmaker) and I have bonded over conversations ranging from the practicality of 4K production to the joy of tilt-shift photography. A recent convert to the 5D mkIII family himself, I knew Chuck and I would be sharing the same level of excitement and curiosity about this flurry of good news regarding the uncompressed mkIII video output and the new digital recorder / monitor made by Atomos. A short time after the Atomos Ninja 2 was released for sale, Chuck rushed out and bought a kit (which have been subsequently discontinued because they were TOO good of a deal, what a surprise). I ran into him the same day he purchased the unit and I can still remember the ear-to-ear grin he had when he told me he had just bought it and wanted to test it out. You don’t have to twist my arm too much when it comes to testing out new camera equipment, especially when it’s something I am extremely anxious to get my hands on, as was the case with the Atomos Ninja 2 recorder / monitor.
Before any real world testing, I like to get my hands on the equipment. I like to feel the weight of it in my hands. I like running my fingers along the lines and edges of the casing. If the item can’t pass those first few initial tests, then I don’t bother with it. Remember, I come from a Camera Assistant background, and there’s nothing worse than a piece of flimsy equipment. Especially one that you have to put so much trust in, like the piece of equipment that’s, you know, RECORDING ALL OF YOUR FOOTAGE. So yes, for me, making sure that the Ninja 2 felt solid and trustworthy was extremely important because once you have faith in the craftsmanship of the item, then you can focus on the operation of it.
In terms of craftsmanship and materials used in the building of the item, it’s top-notch. The piece feels sturdy and well-made. And at the same time, it’s lighter than I was expecting, which was promising (and extremely important) for mounting to a handheld rig. Being that the unit is both a monitor and a recorder, it’s advantageous for me to have all of my crucial gear located on my rig and easily accessible at any stage of any given shoot. And the Ninja 2 seemed to be exactly that.
I will say this, the screen size of the Ninja 2 is a bummer, especially upon second glance. Upon first glance I thought, this is a nice, compact unit that will integrate nicely and unobtrusively within my support system. But then (upon that dreaded second glance) I asked myself, “Is the size of this monitor that much larger, and more advantageous, than the LCD on the back of my camera?” Hmmm…
Chuck and I shot a couple of quick things in his office space, nothing groundbreaking by any means, and we knew that what this unit really needed was a real-world test.
The project –
A good friend of mine and I have recently started collaborating on shooting a series of short films. We both write, he directs and I shoot. They are meant to be mostly experimental in nature, and will hopefully serve as a showcase for our respective talents. One of the core, agreed-upon principles of these projects, is incorporating an extremely high production value and aesthetic to each short. Using the resources we have, along with our own ingenuity, we are aiming to integrate a look that’s usually only done at a much higher budget level. And to be clear, our budget level is somewhere around the $0 mark. With this benchmark established, I knew it was important to step up my game with the image acquisition. I knew that I wasn’t going to be able to rent an Epic or Alexa, and to be honest, I didn’t really think those cameras were right for this project anyway. I got excited that this project could be the perfect case-study for my 5D mkIII, uncompressed video output and the Ninja 2 recorder. So with permission granted from Chuck to borrow his Ninja 2 monitor / recorder, it was off to the races.
The look of this particular short film is very organic, edgy, and verite. Mostly hand-held. Also important to note, is that the 3 minute script takes place in the late-afternoon and magic-hour time frame. So we’re limited to more shooting days with fewer hours, sometimes even less than an hour to shoot. So it became very important to me in the pre-production phase to make sure I had a rig that was going to maximize my time spent shooting and being ready to find shots on the fly.
It’s also important for me to have a low-profile set-up when I’m shooting these guerrilla-type shoots. There’s nothing worse than getting a bunch of onlookers asking questions, slowing down your shoot. Or even worse, a law-enforcement official asking questions, which can lead to a very short shooting day.
So, the rig that I built for this shoot is as follows:
– Jag35 Austin v2 Hand-held Rig (with some various Red Rock 15mm rods mixed in)
– Red Rock Micro Follow Focus v2
– Zacuto Z-Finder Pro (w/ jag35 gorilla stand baseplate adaptor)
– Atomos Ninja 2 Monitor / Recorder (with articulating, 1/4-20 to hot shoe, israeli arm)
Shoot Day 1 –
The scene we shot on the first day takes place in the late afternoon. Hard light and shadows, mixed with soft, atmospheric lens-flare kinds of shots. Since the lighting was particularly harsh, I knew that I would be using my Zacuto viewfinder the entire time, and it was a good thing I did.
One of my first observations about the Atomos monitor, is that without any type of hoodman, or homemade contraption to keep the monitor shaded, it was almost impossible to see the image in that type of lighting condition. Shooting Technicolor’s Cinestyle color profile (more on that decision in a bit) didn’t make things any easier, for that color profile flattens out the image and makes the image appear even more washed out on the already sun-drenched monitor. In summary, bad news for my Director. One of the nice things that I had hoped for, was that since my eye was going to be stuck to the Zacuto, I could turn the monitor for the Director to see during the shot. But most of that theory was rendered useless because of the glare problem in the Ninja 2’s screen. To be fair, they did just release a shade that fits this monitor, so I’ll reserve final critique until I use either the Atomos shade or an improvised one on the next shoot day.
The weight that the Ninja 2 added on my rig, along with the two batts on the back of the unit, was fairly substantial. I wouldn’t say it affected any of my shots, but I was definitely sweating a bit on the longer takes. My lack of time spent at the gym lately could have been a contributing factor as well… but nah, I’ll stick to my first theory. One thing that I’ll mention, is that at the end of the day I started breaking down my rig. I had taken the monitor off of the Israeli arm and we had left the location. As we were driving home, the Director and I both saw a perfect location and a perfect sunset shot that we knew we had to get. So we jumped out of our cars, and I grabbed my rig, along with the monitor, which was now separate from the rig. The Director simply held the Ninja 2 and used it as a handheld monitor, as I was still using the Zacuto viewfinder. It was much better for that particular kind of set-up. It took the weight off of my rig and it gave the Director a very accessible unit to monitor from. Obviously not ideal for when I have to use the monitor for operating, but when I’m using a viewfinder to operate, this is a much better solution that will save me time and relieve some unnecessary back-pain.
A few takes in, the recorder froze, oh-no! The timecode stopped running, the image froze, and I had to reboot the unit. Upon reboot, a message came up asking me if I wanted to take the time to recover and rebuild the file that had been compromised. Not knowing how long that would take, especially with shooting in a limited amount of time, I chose to “rebuild later.” Here’s the danger of that, and it’s more my fault than anything. Moving quickly at the time, and working without a camera assistant / DIT, I completely forgot that the one clip had been compromised. I don’t know how long it takes for the Ninja 2 to rebuild the clip. I’m assuming it depends on the total running time of the clip before it froze up, but assuming it can’t possibly take that long (right?), I’d rather the unit not give me a choice and just do it first thing upon reboot. I like when machines take the guesswork out of things for me. The clip wasn’t a circle take, so it didn’t really matter at the end of the day, but it could’ve been…
Using the monitor to playback shots in the field was incredibly helpful and quick. The layout of the user-interface is very intuitive, which in turn makes for efficient operation. I hadn’t spent much time playing with the unit before this first shoot day and my fingers quickly learned the paths of the buttons and it became almost second nature to go from the recording menu to the playback menu, and everywhere in-between. The menus and button layouts have been thoughtfully positioned and they make sense. These are the sorts of things I love when it comes to camera equipment. I don’t want to think about these things in the field, I just want them to be where they make sense to be. This ultimately saves me time and provides less opportunity for things to become stressful.
I wouldn’t say that the touch-sensitivity of the unit is on par with say, an iPhone, but it isn’t bad by any means. There were times when I had to press something two or three times, but it wasn’t the end of the world.
All in all, the Ninja 2 recorder / monitor worked very well the first time out in the field. It performed more or less how I thought it would, which in this day and age of camera support equipment, is optimistic to say the least.
I definitely wish the screen was larger and that the screen was made out of a different material to help offset the glare from an afternoon sun. But again, maybe that will be drastically improved from the use of a shade or hoodman.
It’s a simple unit. It doesn’t have a ton of controls, but the ones it does have work as advertised and at industry standard levels. I only used the focus assist and zebra in prep and will definitely integrate them more in the second shoot day, as we will be shooting a couple of magic hour and night scenes. I’ll be interested in seeing how the focus assist holds up in the Cinestyle color profile, as that particular profile lacks the contrast that focus assist sometimes depends upon for accuracy.
Wait, digital dailies review? –
One of the coolest things about the Ninja 2 recorder / monitor is the ability to plug it into an HD television and watch dailies. After the shoot, I was simply able to plug a HDMI cable from the Ninja 2 unit’s HDMI “out” port into the HDMI input of the television and voila, dailies! On top of that, I am able to sit with the Director and make selects right there from the unit and flag circle takes for editing. All of that info gets input into a .txt file that I am able to export into the batch of .mov files for input into a nle like FCP X or Adobe Premiere Pro.