I’m rarely a first adopter of new technology, and sometimes I get burned, but the morning they announced the C500mkII I was actively kitting out a C200 package to replace my C100mkII. I stopped immediately, told FilmTools I was absolutely going to buy one and to hold it for me, and I ended up with serial number 008. The key here, the thing that got me excited, is ease and flexibility: with the C500 I’ve got tons of options and can scale up or down depending on the client, and personally it gives me everything I could want when making my own films. You’d be hard-pressed to find a situation that the C500mkII wouldn’t excel in, and I’ll attempt to lay that thought process below.
After practicality (which we’ll get to) the most important thing is, of course, the image. I weigh features very highly when considering a camera, and some are make or break (I really love internal NDs) but at the end of the day I do want the best image possible. After testing the VENICE, Alexa LF, FX9 and C500mkII I discovered that the C500 was the easiest to match to the LF, which I found to be the case with the C300mkII and original Alexa. Now, that doesn’t say anything specifically but it does lend itself to the idea that the C500mkII is “as good as” the big cameras on campus. If I’m getting most of the way to an LF for 20% of the price, I’m happy. I don’t have to worry about it.
I’m really loving the dynamic range on this camera, as demonstrated in my first “practical” camera tests (here and here), and in the grade it really is a treat. Under normal circumstances I’m not seeing any noise or artifacts, the colors are great (especially after I whipped up my own picture profile), and shooting the 2.1Gbps 12-bit RAW just unlocks so much more depth. Canon cameras famously will clip pretty fast compared to Arri, but when you keep everything under that threshold the rolloff is really pleasant. I still expose as bright as I can and then bring it down in post, but even so the C500mkII seems to record very low contrast so I advise using the shadow log wheel in Resolve to bring the darkest areas down a touch to give your image some more “bite”. The low light performance (as has been asked of me more than almost anything else) is good as long as you have a light on. Basically the way I see it, ISO800 looks like real life; if you can see it with your eyes it should work out. During my tests I found that yes, if it’s really dark it’s noisy. Resolves denoiser handled it find, but it is there. That being said, I was shooting in the dark. If you can turn on a light you should be fine. If you’re going to use or buy a camera at this level one would hope you have some lights around. A regular house light bulb can even do the trick, it’s great. Higher ISOs look fine, you just can’t have no light in the scene without encountering some noise. Find a bulb, use it.
Previously, CLOG3 was introduced to mitigate the less-than-ideal noise situation in the shadows on the C300 at the cost of some DR, but in the C500mkII that is no longer a problem. CLOG3 still looks great, but for the most DR and lowest noise, CLOG2 is definitely the move; I don’t foresee a reason for you to shoot Clog3. Some say Clog2 is “more difficult to grade” but using color space transforms in Resolve instead of LUTs or going full-manual makes this trivial.
With XF-AVC you’ve got 10-bit 4:2:2 footage at 410Mbps which is lovely and a broadcast standard that made the C300 so popular. You’ll also be best served shooting a touch over-exposed if you can, and bringing that down in post. The Canon sensors really love light. Shooting up to 4000ISO is totally acceptable in my opinion, although ISO800 seems to be sufficient for any normal circumstances, and you could likely lean on 6400ISO with some NR in post. In any case, either shoot CLOG2 or if you need to, 709 (the C500 OWLCOLOR profile includes a 709 version *wink*) for the best results.
A bigger sensor
The highlight feature of the C500mkII is the Full Frame C700 sensor, which is just a touch smaller than the RED Monstro. I got my C100mkII because it was an original C500 sensor in an indie package, and this is kind of the same situation just amped up to 11. I was never a 5D shooter, opting instead for the AF100 as an upgrade from my XL2, but Full Frame is so fun to shoot, and being able to really isolate things in the frame opens new options. You absolutely can “match” the characteristics of a S35 and FF sensor (to a degree) but the additional photosites do give you a cleaner, sharper image as well as the opportunity to more easily get a nice blurry background on (perceptually) wider lenses. To that end, shooting Full Frame may require you to get some new lenses, but if you’re getting hired on a gig that doesn’t care about Full Frame you can just switch on over to S35 mode and use the ones you’re used to or already own. You’ll also likely switch over to the S35 mode if you’re on most PL lenses.
Resolution is a tool, not a barometer of quality. I think everyone would rather shoot with an Alexa at HD than a iPhone at 4K right? A good sensor beats out a large output every day. That being said (and RED has kind of made this their thing) higher-than-average resolutions can give you a lot of options. When given the option, David Fincher shoots 6K/8K for a 5K/6.5K extraction respectively to give him the maximum amount of real estate for reframing and stabilization in post (a common refrain, but one that Fincher famously uses on nearly every clip along with split comps). Furthermore, their deliverables are either 4K or 2K, so when they kick out their final render the image is still super clean and sharp even though they haven’t used the whole image recorded.
If your delivery resolution is 1080p, 6K gives you a lot of room for interesting or helpful setups. As a classic example, I can set up a two-shot in a car and effectively have 3 camera angles in the form of a wide and two punch-in closeups. Coupled with my Odyssey 7Q+ and the Titan Extract function I can actually shoot that way (in 4K) and have all 3 shots saved individually, and can even live-switch between the three which could be big for shooting interviews. As a best practice you always aim to shoot “better” than your intended display, so if your deliverable is 1080 8-bit, you’d wanna shoot something like 4K 10-bit. In the same way, a 4K 10-bit deliverable calls for a 6K 12-bit recording format if you’re trying to get the best finished quality.
More than 4K+ resolutions, HDR is the dark horse in the camera race. Dolby Vision and formats like it are rapidly coming to televisions across the world and for good reason: HDR is markedly better looking (when done right) than an SDR image at the same resolution. Being able to see more deeply into the shadows and bright areas of an image can give your footage far more life and realism than a simple boost to the K-count. I personally am not mastering for HDR yet but it’s nice to know the camera is capable.
NOW, there’s a difference between “recording” in HDR and displaying it. RED’s HDRx and to a degree Canon’s new DGO (I think) are more akin to recording HDR as we would with multiple exposures in the stills world. In the case of the C500mkII, it’s the fact that it can record 15 stops of latitude that makes it “High Dynamic Range” not some new exposure method. Traditional monitors or TVs (8-bit 709 or sRGB) can show -at best- 8 stops so obviously there’s going to be some adjustments made in the grade. The difference is that new HDR displays can show more of that range. You don’t “shoot” in HDR, you master for it. There’s a few different standards but for those keeping score, HDR10 and Dolby Vision are the clear frontrunners.
As I mentioned, the practicality of the camera will come in to sharp relief when you’re using it on a daily basis. Small things will start to stand out and really grate on you over time so it’s important to really give any given camera hard scrutiny and see if it stands up to your use.
To use computers as an example: my PC is pretty fast. It handles editing most anything I throw at it and I almost never need to use proxies. It’s great. However, when I go and use stabilization on a clip, it takes a bit of time. When I reviewed that computer Puget let me borrow, I noticed that I was never waiting for renders or effects or anything and my workload could be increased by a significant margin because I wasn’t getting distracted. You see, when the stabilizer takes 30 seconds I’ll instinctively reach for my phone just to see if someone texted me, if I got an email, whatever. I’m sure we all do that, but it breaks your flow state and can make editing feel more like a chore than it needs to.
In the same way, little things on the camera that get in your way or make you look for external solutions can become cumbersome or even annoying over time. One thing I’ve loved about the Canon Cinema cameras is there are rarely “splinters” like that. I can only name two on my C100mkII (switching to slowmo mode and the center-only AF) and the C500mkII is similarly competent. I’ve yet to run in to something about it that really grinds my gears, although some could argue only being able to use HDMI or SDI at any given time, and the lack of proxy recording during XF-AVC filming are annoying. Hopefully they can be fixed in a firmware update.
Audio & NDs
At this point in my life I absolutely demand in-camera ND filters. I can’t think of a reason why I’d prefer to use mattebox NDs and I’m squarely anti-variable ND (the screw-on kind not the FX9 kind, that’s actually cool). This camera has 10 whole stops of ND, which is 4 more than I’m used to. It also has built-in XLRs which is great for smaller shoots where I don’t have a sound person, or said sound person is mixing directly into the camera. With the expansion unit you get an additional two XLR inputs bringing the total to 4 channels. I’ve seen folks grumble about the location of the gain wheels but I’d rather them be on the assistant side than on the top handle again.
If I’m a solo operator or working on something that doesn’t need the toppest of quality, I can easily just “detune” the C500 to shoot at S35 (for lenses) at 1080p (even in RAW if I want), and if I really want to min/max it, I can just use the 2K Proxies which are SuperSampled from the full 6K sensor and saved to 35mb 420 8-bit files on the SD card allowing for something like 10 hours of record time on 128GB SD card (although regular 2K XF-AVC comes in close at 7hrs on a 512GB CFexpress card). I’ve always said the C100mkII (and even C200) 420 8-bit footage is some of the best in the game, punching above its weight, and the C500mkII’s proxies (proxies!) are no different. I still get DPAF, SDI, ND’s, XLRs (and a 3.5mm mic), and even the built-in EIS but can shoot a more modest filetype. Tons of people use the Alexa Mini to shoot small-budget music videos or shorts and the like, but that camera is designed for a crew, meaning concessions would have to be made (such as with audio). With the C500mkII, there’s no such issue. In regards to the EIS, it does work but it’s better suited for smoothing out bumps in a track or standing still than full-on handheld. Face Detect AF, on the other hand, is absolutely amazing.
Film Set Features
Even though it works great at an indie level, the C500mkII is squarely built for larger productions. SDI and HDMI outs, PL/EF Swappable (even B4), Genlock, Timecode, XLRs, NDs, the RAW Workflow, 6K,4K, 15+ Stops DR, Anamorphic Modes, False Color, HDR, LUTs (in-camera and to different outputs), and all in a small form factor. It’s got nearly everything you could want in a cinema camera. Only the VENICE seems to have more going on. It’s also N E T F L I X A P P R O V E D and is ready for the ACES workflow. One thing, though, is you’ll likely want the V2 expansion pack if you’re on a bigger shoot, but it doesn’t make the camera unwieldy at all and is easy to attach. I would like to see some 12v power solutions for accessories but hopefully that can come from a new backpack (a “V3”). Another nice thing, as an aside, is the backlight side-buttons. I’ve noticed that the columns of buttons are on raised planes so you can find them with pure dexterity but if you’re the operator and you look down, they’ll be illuminated for you anyway (which you can turn off). Another nice thing is that it turns on fast (like a couple seconds), formats fast (also a couple seconds), and black balances quickly (again, seconds).
DPAF is the truth, and it’s not just for autofocus. It also allows for “AF Boosted MF” (which I’ve yet to test) but also the fantastic focus assist box which visually shows you if you’ve missed or come up short on your focus pull with some arrows. It only works with EF lenses but it’s way way better than having to squint and get your face an inch away from the monitor. Focus peakin
The stabilization is decent. It’s EIS so it’s not a gimbal replacement, clearly, but it sure helps with shoulder-mounted or monopod situations. Even handheld it smooths things out a touch without getting too jerky or take on that “warp stabilizer” blur. Otherwise, I personally don’t use gimbals (or haven’t really, I should say) but since the C500 is so small it easily can be run on even a simple one like the Ronin S, which for some people is a game changer.
One thing I’ve loved about shooting on the Cinema EOS cameras is the body itself, and this new version shared with the C300mkIII is my absolute favorite iteration. Some people have gripes about the LCD placement on the C300 (which I get) but overall the cameras crush in the ergonomics department. In the C500’s case, it’s super small and light, the backlit side-buttons are already set up with your most used features but are all easily re-mappable. With the Cinema EOS cameras it really is as simple as “battery, media, lens” and you’re shooting. The Alexa Mini uses CFast cards, which are more ubiquitous than the CFexpress cards in the C500, but as more cameras switch over that’ll be less of an issue. For now, though, media is a bit of a struggle for the C500. Not as bad as it is for RED though. Also it’s SUPER easy to use slowmo now, with a dedicated button on the side that switches over the shooting mode instantaneously. On top of that the new Touch Screen and UI are nice and simple, and you can even have the data is around the edges of the frame (or not) more like the ALEXA.
I’ve had to do much less customization to get the image I want than I did with the C100mkII, but after using it more I will likely find little things to tweak.
First thing I did was make my own Picture Profile; I set up some color charts and did my best to match the C500 to an Alexa using a vectorscope. It’s not dead-on, but it’s close. Normally I’ll design my own monitoring LUT but after making the picture profile I’ve learned that I’m plenty happy with using the K1S1 LUT from ARRI, so I just loaded that in to the C500. That’s definitely something I’ll work on in the future but for now it’s just fine and can make grading simple things or quick turnarounds easy. Luckily Clog2 and Log-C are nearly identical gamma curves so getting the color close was a more valuable use of time than messing with the black point and whatnot like I did with the C100mkII version of OWLCOLOR.
Generally I’m shooting the supersampled 4K but will switch over to 6K RAW when I’m on a gig that I’d like to flex on, a commercial or something. Honestly I’ll use the 2K proxies when it’s just something quick for myself and I’d rather save HDD space, they easily out-perform the files out of my C100mkII and functionally (saving the files to an SD card) is the same experience. I really do love that.
I haven’t used it yet, but when I use the S35 crop I’ll simply go back to using my trusty Sigma 18-35mm or whatever PL lenses are called for, but at the moment I’m using a set of Nikkor primes when shooting FF (a 24/28/35/50/85/105mm set) that I’ve had for my Nikon F2 for many years. Using my 3D Printer I ran off a set of focus gears that pressure fit on to the rings of the lenses so that I have nice, low-profile and seam-free gears that work on any focus pulling system and are incredibly compact, only adding a couple inches to the front of the camera. I’ve also got an RZ to EF adapter if I really want to feel fancy and use my Medium Format lenses, but I’m thinking a set of 645 lenses might be more apropos (and cheaper). One day if I’m lucky I’ll have Zero Optik rehouse the Nikkors and then we’ll be in a whole new world.
The C500 is a bit power hungry so I’ve got a DTap to XLR cable to run my cinema batteries off-camera instead of the usual BPA60 or having to use the V2 expansion unit. On a tripod I have a little v-lock clamp that holds the battery, but when handheld I’ll either use a MOLLE pouch or even just the pouch of my hoodie. That being said, for a quick shoot (under 4 hours) I’ve had no issue using the BPA60 and just turning the camera off when I’m not using it. Probably get an hour and a half of run time.
I’ve also got a Bright Tangerine full cage for rigging purposes. With it I’ve got top and bottom rails, ⅜” and ¼” taps all over the place (with security pin holes), a top handle, a place for my monitor (although I tend to use a Noga arm instead of the attachment from Canon) and BT’s nice “Lift-Off” feature that lets you quickly detach the cage from any Arri-style dovetail or the second “top plate” flipped over to attach to a regular tripod plate. I’ve also got a Redrock Micro handle and a spud attached to the left side so I can go handheld very comfortably.
WRAP IT UP
Oddly enough, I feel like the C500mkII’s only true competitor is the Sony VENICE in terms of specs and options. I’d love to see an LF Amira though, if I’m honest. The image of the C500 may be Alexa-like (to a degree), but aside from that only the VENICE has a similar set of features and usability. Then again, you’re also paying way more for it and it’s a big ol’ boy too. I struggle to think of a single gig you couldn’t use this camera on, it covers all the bases from web to doc to cinema work. You can check out the full nitty gritty specs here, and the user manual can be found here.
You can watch me read this entire article to you in this video if you’d like:
Canon EOS C500 Mark II Cinema Camera