Camera Tests, Pt. 2: The Practice of Camera Testing
As I blogged about in my last entry, this week was all about the first round of applying my theoretical knowledge and putting it into action – both in terms of filming as well as editing – as well as my learning accompanying it.
So let’s dive right into it!
Monday, 4 November 2019
Monday started with Hannah and me taking out the equipment and setting up for the camera tests. Thus, we started with the setup for the colour and exposure tests. Since time was pressing (with only about 2h per person and week in the studio), I settled for a setup that allowed me to film both the colour tests as well as the exposure/latitude tests simultaneously. Thus, following the setup of Shane Hurlbut as delineated in the video below:
I not only included the mandatory stand-in and a SpyderCheckr24 colour chart, but also a multi-coloured background consisting of plain white, black, and green screen (since they are the colours we will most likely encounter filming in a studio and using green screen) to include these as exposure reference points in relation to our stand-in. We furthermore included a sheet of the white leather material we currently intend to use for the interior of the spaceship to give us a feel of how we could and/or should be lighting.
While that setup was quite easy to implement, finding the correct exposure to start with was a bit difficult to achieve. Since I wanted to have an equal scale in both overexposure and underexposure, and because our lens kits cover an aperture range of 9 stops, I needed to set the lighting in a way that would result in our subject being correctly exposed at f5.6, allowing me four full stops of exposure in both directions.
After a couple of rounds of metering and adjusting the lighting, I finally found the sweet spot of exposure for my subject and noted down the reference values of the background objects in there. This was the setup we ended up using:
PLACEHOLDER LIGHTING PLAN FOR CAMERA TESTS - TECHNICAL ISSUES - PLEASE BEAR WITH ME UNTIL UPLOAD
Following this, we set up both cameras – the Blackmagic Cinema Camera and the Blackmagic URSA Mini Pro 4.6k. As such, and for the initial setup, we used the exact same settings on both cameras:
1080p
25fps
Native ISO of 800
Shutter speed of 1/50 (URSA) and shutter angle of 180° (BMCC)
24mm 1.5 Samyang lens, set to f5.6
ProRes 422
Aspect ratio of 2.39:1
Dynamic range set to film
And put the cameras next to each other, at the exact same distance away from our stand-in. However, the individual results looked decidedly different, as this image from the edit below illustrates:
These two images illustrate two different things: Firstly, a difference in colour space, secondly, a difference in frame size. While colour space and colour rendition is the matter of another blog entry entirely, it should still be noted for the time being.
The reason for this is as follows: The difference in frame size, however, is worth looking at. For this is a prime example how the different sensor sizes used for the individual cameras affect the effective focal length of a lens.
As I blogged about in my blog entry My Test Subjects: The Cameras I Will Be Testing, the Blackmagic Cinema Camera (released 2011) is effectively close to a 16mm sensor, whereas the Blackmagic URSA Mini Pro has a Super35mm sensor.
As you can tell from the images above, this difference in sensor size also results in a difference in field of view. While 24mm on the BMCC (with a crop factor of 2.4x) results in an effective focal length of 57,6mm, 24mm on the URSA create an effective focal length of 33,6mm. Because of this, I had to mitigate this difference by swapping the lens on the URSA for the ensuing exposure tests.
Thus, I used a 50mm lens for the URSA (effective focal length of 70mm) to approximate the difference between both cameras and just tweaked the physical distance of the BMCC to make both shots match up framing-wise. A rough approximation, I know, but these were the limitations of the kit at hand. As such, the setup was conducted with the following settings:
1080p
25fps
Native ISO of 800
Shutter speed of 1/50 (URSA) and shutter angle of 180° (BMCC)
50mm 1.5 Samyang lens, set to f5.6 (URSA) and 24mm 1.5 Samyang lens, set to f5.6 (BMCC)
ProRes 422
Aspect ratio of 2.39:1
Dynamic range set to film
After this little demonstration of the different affect in sensor size and swapping the BMCC to a longer lens (plus physical re-jigging) that approximated the effective focal length of the URSA camera, we finally started the exposure/latitude tests filming mid shots of our stand-in Eline; first the overexposures, then the underexposures, by stopping up/down half a stop for each take.
After the mid shots were done, Eline had to leave, which is why Hannah jumped in for the MCUs. This however allowed us to film another set of over- and underexposures with a different skin type.
Much like with Eline’s mid shots, we started with simultaneous overexposures of the BMCC and the URSA, then moved on to underexposures with the same framing. After we conducted these tests, my time for this week was up and we moved on to Hannah’s setup and camera and lighting tests:
Wednesday 6 November 2019
After having transferred the footage to the Editshare on Tuesday, Wednesday was the time to putting the footage through the edit and analysing it, as well as assessing any pickup shots that need to be done in order to complete my experimentation.
Having had a quick look at my footage I was pleased to find out that all went fine for the colour and exposure/latitude tests and that I could proceed onto editing and analysing them for my purposes.
A First Round of Analysis
Having watched back the footage, I not only learned about the different colour spaces of the BMCC sensor and the URSA sensor; I was also able to prove the claim that the URSA underexposes its footage by two stops. As such, the footage of the BMCC at the correct exposure of f5.6 – both in terms of colour as well as exposure – looks approximately the same as the footage of the URSA at f2.8.
The Editing Process
After setting up my workspace and my project settings for colour tests, overexposure tests and underexposure tests each, I imported the relevant footage into the 2019-version of Avid and put the relevant metadata about each file to the relevant files.
I then pieced together the relevant clips in sequence and added the required information to it as shown below using the Avid Titler+ title tool. This took me quite a while since I needed to make sure that the information provided was correct at every step of the way to ensure that my analysis would not be mistaken due to an error earlier in the chain. Also, the process of editing titles, although meant to be more versatile with the new version of Avid (now also including titles for resolutions higher than HD), also became a lot more complicated, taking away a lot of valuable time in the process.
Lastly, I then set about polishing the clips from a rough cut to a first edit. However, the individual exposure reference points for the background will still need to be included in the videos in the upcoming week.
Thus, the ungraded versions of my exposure tests and colour tests were almost done. The next step next week will be to test the ISO settings of both cameras using the lighting scenario I devised early in the module and testing for grain/noise, as well for aesthetic effects. Furthermore, a corrected version of the exposure/latitude tests and colour tests, as well as the sequencing of the ISO-tests will have to ensue next week.
Still a lot left to do – better get going!
References
Hurlbut Academy (2016) Blackmagic URSA Mini 4.6K Latitude Test | Over Exposure at ISO 800 [online] Available at: https://www.youtube.com/watch?v=GRNoSHysSxM&list=PLHyU9PnnTe1eK6TDFuJkhUrsRpZB86cKh&index=94 [Accessed on 7 November 2019]
VMI (n.d.) A Filmmaker's Guide to Sensor Sizes and Lens Formats [online] Image taken from: https://vmi.tv/training/useful-stuff/Guide-to-Sensor-Sizes-and-Lens-Formats [Accessed on 7 November 2019]