A few days ago the guys at Film Cyfrowy shared Part IV of their “12 Cameras Test” which features comments by a professional colourist. Their 12 Cameras Test compares 12 of the the most popular digital cameras in an attempt to find out how each camera captured reality in relation to the ARRI Alexa. The purpose of their test wasn’t to see which camera is a winner or a loser, but rather compare how each one captured the same scene and then compare the final images when colour matched and evaluate the differences.
As a reminder, the cameras used in the test were: the ARRI Alexa, RED Epic, RED Dragon 6K, Blackmagic Production Camera 4K, BMPCC, KineRaw Mini, GH4, 5D Mark III, Canon 1DC, Canon C500, Sony F55 and FS700. The ARRI Alexa was used as a benchmark for colour accuracy.
Colourist Wiktor Sasin (Chimney, Poland) shared some comments from his perspective about how each camera performed in the test. We’ve summarised his findings in the next couple of paragraphs.
In order to match the colour from the 12 cameras, his goal was to even out and match skin tones, not necessarily going by the grey card. Establishing comparable skin tones from all cameras showed him how different each camera was in the way it captures a particular scene. The adjustment of the contrast in the skin tones when trying to match them renders a different reproduction of colour in the rest of the image shot by each respective camera. This difference can be clearly seen on the grey card.
- The ARRI Alexa was used as reference, thus the footage from the Alexa had only an ARRI Alexa LUT applied and no other adjustments. Thus his goal was to match the rest of the cameras to the Alexa look.
- The RED Epic and Dragon performed well requiring little adjustments to arrive at a comparable image to the Alexa.
- The BMPC 4K has a lot of potential possibly attributed to its 4K resolution. The images from the BMPCC (Pocket Cinema Camera) were also rather easy to work with a both Speed Booster and non-Speed Booster test, they were able to achieve an acceptable image in a short period of time.
- The C500 required more work to get it to align with that of the Alexa, and so did the 5D Mark III, on both of which they did not use the Alexa LUT as it complicated matters further.
- With the 1DC they had some problems as the footage was shot with the wrong colour balance, and due to not having the ability to record Raw on the 1DC, the white balance issue affected the images from the camera permanently.
- The Sony F55 and the FS700 according to him “exaggerate the colour a bit, and have a Television look to them, rather than a more filmic look”. On both Sony cameras they did not use LUT’s but matched colour manually.
- The KineRAW Mini surprised him as it matched the Alexa and RED Epic/Dragon much easier than most of the other cameras in the test, a produces an images similar to the original RED One camera.
- They pushed the GH4 to its limits but found it rather difficult to “escape the TV look” and match it to the Alexa.
12 CAMERAS TEST – PART IV – “THE COMMENT” from Film Cyfrowy on Vimeo.
In summary, according to Wiktor, the Alexa is still a benchmark in acquisition, and the RED Epic/Dragon coming in close. He was very pleasantly surprised with the performance of the Blackmagic cameras (the 4K and the Pocket Cinema Camera) as well as the KineRaw Mini.
In my opinion, I think if nothing else this test proves that the Blackmagic cameras have a what it takes to be used on any film set. As evidenced recently by the fact that the BMCC was used to film complex action sequences in The Expendables 3. I am a but surprised as to his findings regarding the F55 and the GH4, but this is his perspective and findings from the test.
The F55 /FS700 and GH4 are fully capable of producing cinematic images, as evidenced by a multitude of cinematic content shot with them, which one can find with a simple google search. And of course, the fact they had the GH4 for only a day and not being able to fully get into the best settings for video has contributed to their findings. The F55 is capable of 16-bit 4K raw, however for this test they used the internal 4K XAVC in MXF, and I am sure the results would have been different in raw.
Regardless, I think this test can be helpful to a lot of folks out there who want to see how a specific camera compares to the Alexa.
[Links: 12 Cameras Test Parts I & II / Part III]
Follow Film Cyfrowy on Twitter here.
Claim your copy of DAVINCI RESOLVE - SIMPLIFIED COURSE with 50% off! Get Instant Access!
Not sure on the validity of this test.
There are obvious moving reflections in the test chart. Glare has to be completely eliminated on the chart for this test to be proper.
In addition, they evidently screwed-up the 1D C test.
What exactly is a TV look? And what is film look? When I hear this I realize we are dealing with “professionals” that think cinema starts with a cinema camera .