By Devin Forbes

Was it everything I hoped? YES. The images I got were spectacular, and I truly believed that it would revolutionize filmmaking. And it absolutely did, putting cameras capable of beautiful images in the hands of younger filmmakers that lacked the resources to buy or rent the more traditional cinema cameras like the Arri Alexa or the newly released Red One. It made the film industry more accessible.

I am eternally grateful for my trusty 5D – it was the camera I used in the early days of Process Pictures, shooting countless short films, commercials, and live events with it.
It is well known that there are limitations to the video produced by the 5D Mark II. Its H.264 files are highly compressed, leaving little room to push it very hard in color correction, and break down quickly depending on how carefully you plan your post workflow. The 5D is not, nor was ever intended to be the ‘pro camera killer’ that some professed.
But this is not what I’m talking about when I say my camera developed serious problems in mid-2012. What I experienced was catastrophic image degradation over the span of only four years in the field.
By 2012, roughly 25% of the clips that I recorded were crushed in the highlights and shadows, had a dull brown color shift regardless of color profiles, and had vertical banding in the severest sense you can imagine. It was hard to believe that the clips even came from the same camera.
The camera was sent in three separate times for a top-to-bottom service, and each time Canon reported that nothing was out of the ordinary. For the longest time I believed it was the sensor failing, but then there was a clip I shot for our “Miracle on the Hudson: Coming Home” documentary that made me rethink this conclusion. The clip began looking exactly like I had intended – evenly exposed with the Technicolor Superflat color profile. Then about a second into it, it’s like I had flipped on an ND filter, and all those horrible image traits appeared.

Around this same time, the camera developed a severe buffer problem where it would stop rolling within a couple seconds of hitting record when filming after playing back a previous clip in-camera. We couldn’t tell if it was the camera or the cards we were using, but the cards were fast enough, and we had been using them for months prior to this issue developing. All of this got me thinking that it must be a board (or chip) failure elsewhere in the camera that was causing these intermittent problems.

The 2.5K BMCC, while still entry level, was a monumental leap from DSLR filmmaking. The ability to shoot in a raw format and color correct with the bundled DaVinci Resolve software was a godsend for us, and we couldn’t be happier with the decision to upgrade.
Now we use the DSLRs to do what they were originally intended to do – take photos.
Comments 7
I agree very much with your findings here. I also prefer real video cams for real video and DSLRS for photographs. I see ergonomic and other tech issues involved. DSLRS and their lenses are not made for extended handheld shooting, and need extensive accessories such as Zacuto brackets and external audio recorders to alleviate this. I can often tell if a DSLR was used to originate video from its appearance like your example.
Its like saying; I wasnt happy with my Honda Civic but now that I got money I’m better off with my Porsche!!!! Frugal means shoestring budgets and dslr have opened the big leagues to all; that s y 4k and 5k are storming! Gotta make things obsolete so we can see the difference between me and Spielberg!!!
Author
The BMCC is hardly a Porsche, Tony. It is a stepping stone to bigger and better things.
The 5D was never intended to be a cinema workhorse. With that in mind, it is interesting to note that Laika uses only 5D cameras to shoot their Academy Award nominated films, so these cameras are not exactly a horrible choice.
thanks for the article and comparison. I would have loved to see the same lens used on both cameras for a 1 to 1 comparison ~ using a Canon 5D Mark III with 50mm, f1.2 and Cinema Camera with 24mm, f1.4 seems like you would expect to get different results, no?
Author
We are now working on a BMCC RAW / Canon 5D Mark III ML RAW comparison, so stay tuned!
It’s still a tool, there are plenty of cameras/tools to choose from and unless you’re using a high end camera there will always be trade offs in capability that includes your camera of choice, although not a dslr barely a notch above one. Hey if it works for you great but not a compelling reason to dump dslr’s . I’ve had and used several 5D’s and never had any of the issues you report but I don’t use only one camera, it’s generally different every day and even for the same projects it’s usually a minimum of three different cameras for most things…just my 2 cents