The folks at Cinema5D put the Sony A7s low-light monster through a thorough dynamic range test against industry heavyweights – ARRI Amira, newcomer – Panasonic GH4, and field-proven – Canon EOS C300, Canon 5D mark III and Canon 1DC DSLR. They call this test – the Cinema5D Lab test.
By dynamic range what is meant is generally the latitude, or range of shades of grey at a given exposure that the sensor can capture. In other words, it measures the levels of luminance any given camera can capture. The bigger the number, usually measured in stops, the more detail is retained in shadows and highlights, making for a more robust, natural-looking and visually pleasing image.
What is the cinema5D test lab?
cinema5D has established their own scientific testing facility to accurately measure and evaluate the performance of cameras. As a platform for reviews about cinematic cameras we strive to provide objective comparisons and share insights to help you choose the right camera for your projects.
The cinema5D test lab has been developed over the past 6 months. We are using precise imaging tools, techniques and software to evaluate each camera’s performance precisely. The following test indicates one of several attributes we test when looking at a camera sensor.
The Sony A7S is definitely an incredible low-light performer, but not only that, it’s strength lies in it’s sensor’s ability to deliver very detailed 1080p down-sampled image without line-skipping from a 4K sensor in a robust 4:2:2 XAVC-S codec. Yes, it is 8bit 4:2:2, but coming from a monster full-frame sensor with superior low-light capabilities, like the Bionz X, the images from the Sony A7s can rival some 10bit images coming from competitor cameras in a heart-beat.
The A7s also benefits from a high-quality OLED viewfinder, slow-motion 50/60p mode, as well as a much often neglected APC-S crop-mode, which approximates a Super 35mm gate in the world of film cameras – the standard for feature film acquisition of the last 50 or so years.
Here’s the dynamic range test that Cinema5D did:
Surprised by the results? We were a bit. Definitely surprised to see the A7s get so close to the Amira, which has the same sensor as the ALEXA, which is rated at 14 stops of DR. Now, as the chart shows above, the A7s was shot in S-Log2 Cine Gamma for this DR test in order to get the most out of the sensor.
Sure, Sony rate it thereabouts, but that’s usually just a marketing gimmick, and the real usable DR is usually at least 1 or 2 stops below what the manufacturer claim, as in their testing maybe the A7s was rated at 13.5 stops, and of course they will round up for marketing purposes. Cinema5D used the Zeiss 50mm CP.2 Macro below for their test as well as DSC Labs Xyla 21 charts for this test.
As a rule of thumb, anytime I look at DR tests, I’d always take-off 1 stop of DR just to be safe, a more conservative approach. After all, if 13 stops of Dynamic Range is not to your liking, or does not allow you to get a pleasing image, maybe you should consider a different line of work.
Ok, so with that said, back to the chart above.
The Canon 5D Mark III does not have a log option, but the Canon C300 does – Canon Log. I am a bit startled that the 5D Mark III did better than the C300 and the Canon 1DC, and here is what Cinema5D say about this:
As mentioned before we’re measuring usable dynamic range. This means we’re observing actual dynamic range relative to the noise ratio of the signal. In other words, we only measure dynamic range where the signal still upholds a certain quality which is measured in noise. This is where the bad resolution and codec of the 5D mark III gets a few extra points of dynamic range rating, because its softness blurs the visible and measurable noise. While this might be deemed unfair, the worse quality also gives the viewer an impression of a cleaner image in view of its actual resolution.
It is important to note that they did this test in a complete blacked out room, with a very sharp Zeiss CP2. 50mm 2.1 Macro lens, and set each camera at their respective native ISO, while only adjusting the aperture to get the proper exposure.
This is very helpful test, which puts a lot of things into perspective for people who are not only considering the A7s as their next purchase, but also, for those who want to see how other 4K cameras like the GH4 stack against some of he big boys like the ARRI Amira and Canon C300.
Let us know in the comment section what you thought of this test, and head over to Cinema5D to check out the full test on the link below.