Bob Primes, ASC produced a documentary that evaluates the new wave of single large chip cameras.
This documentary was showed at the Paramount theaters during the Cine Gear Expo of 2011.
- Bob Primes, Image Quality Geeks
I was there for the screening and also could watch a Q&A with the producer Bob, cinematographer Stephen Lighthill, and post supervisor of the documentary Mike Curtis.
This documentary is a carefully Single Chip Camera Evaluation, shot on 12 cameras . There was an extensive battery of tests that evaluate the following:
- over and underexposure latitude,
- low light performance,
- compression artifacts,
- color reproduction,
- skin tones,
- shutter artifact
The cameras being evaluated were:
- Arri Alexa,
- Sony F35,
- Red One MX,
- Panasonic AF100,
- Sony F3,
- Phantom Flex,
- Weisscam HS-2,
- Canon 1D MkIV,
- Canon 7D,
- Canon 5D Mk II,
- Nikon D7000
- and two film stocks: Kodak 5219 and 5213
Before we go over the details of this evaluation, let me explain how they did the evaluation which I think is very relevant. All cameras were recorded in their best output formats, which means that if a camera had as its best output a 4:4:4 12 bit to a external device, that was the way they shot this particular camera in all tests. Also each and every camera had a person called “camera master” that would be responsible for tweaking the camera for optimum quality. There was no color correction on the footage that would disregard the evaluation. The editing was soundless, so we could focus on the image. Last but not least, they had total control over the whole process from conception to final cut, with no interference from any of the cameras manufactures.
With that being said, lets show some graphics taken from the documentary that explain in details the camera performances.
Maximum resolution is the amount of data in one pixel. As you can see in the graphic below, the camera that performs better is the film camera. Surprisingly enough, the Red camera is as good as the film cameras in this category. The worst cameras are the DSLRs, which is expected because they have the worst compression, huge aliasing/moire and terrible recording outputs (they record only on internal cards).
- Maximum resolution
Sensitivity is how well the camera (sensor/film) will capture light. In this regard the film cameras were the worst ones, but I did some research and I saw that these 2 film stocks (Kodak 5213/5219) were actually pretty slow. Correct me if I am wrong, but it seems that they are 200 and 500 ISO. Which means that it is expected the low sensitivity results. The best camera, or the most sensitive, was the Canon 1D, that also has the biggest sensor. Of course that everything comes at a price, and in this case is the amount of noise that the Canon 1D has.
The Sony F35 was the one that I was amazed by the sensitivity performance. It is not bad, but the low light performance resembles what the film cameras capture. I would even dare to say that the F35 would be similar to a 1000 ISO Kodak film stock (if that’s even a possible consideration).
The Arri Alexa is a very good camera overall, and I also like the Sony F3 and the Red One.
Latitude measures the amount of light a camera/film can process and still creates an acceptable image. For example: a good camera has a greater latitude and it will be able to capture in one frame the highlights and the shadows without any lost of information/detail.
Of course that the film cameras would perform better in this category because they are film, and there is no digital sensor capable of capturing so much information than a film. What I was amazed is how close the Arri Alexa is from the latitude of the film stocks. In the latitude category, we could divide it in three groups. First is film and Alexa, then is Red and F35, and then is the rest.
Over and Under Exposure Latitude
This category evaluates the latitude but having a middle point where the image is properly exposed and then the exposure is stretched over and under.
We can see now that the sensitivity evaluation meets with the latitude evaluation at this point. The film cameras perform better at over exposure and lack at under exposure. The Arri Alexa is again a very reliable camera, and the rest is pretty much the same.
- Over and under exposion
A good visual comparison would be the following video from Nick Paton, ACS comparing both 7D and Alexa. Check it out!
Rolling Shutter and Compression Artifacts
In my opinion, for the evaluation in this set of test, we can divide the cameras into 3 categories. The good ones, the not so bad ones, and the ones that are terrible.
In the good ones category there are:
- Arri Alexa
- Sony F35
- Kodak 5213/5219
in the not so bad ones there are:
- Phantom Flex
- Red One
In the terrible ones there are:
This was a very subjective evaluation and in my opinion there is no way I can even give my impression. I think that for this evaluation, everyone has to look at the image and decide by themselves.
Well, lets get to the bottom of the discussion. There is simply no better camera that does everything. It all depends on questions that are not related to technology but external factors like:
- budget or “how much is being reserved for the camera and/or camera department?”
- aesthetic or “what is the feeling being sought by the cinematographer?”
- final medium delivery or “where this video/film will be seen?”
- expertise or “how well the cinematographer uses this tool?”
- personal taste or “I don’t care what other say, I love the camera I use!”
What makes a good picture is definitely not the camera itself, but the good use of objective and subjective visual components.