The shootout consists of a series of scenes shot over the course of 3 shooting days. These scenes are designed to push the cameras to the limits of their abilities. Image quality will be compared between the two cameras by assessing dynamic range, exposure latitude, MTF, resolving power, noise, and color reproduction.

These tests should show how close digital cameras have come to achieving the “film look” (remember this project took place in 2015). Alternatively, the 4K film scan should prove a worthy adversary to the RAW workflow and demonstrate why film is still a popular choice amongst cinematographers in the film industry.

These scenes were created and lighted in a way so that all six cameras from the three groups could easily switch off. Some scenes were adjusted slightly to accommodate for an individual camera. Having a group shooting at high frame rates requires a lot of light; so the other cameras adjusted their normal exposure to compensate for the increased amount of light, if necessary.

Note: We are limited to an ISO of 500 for the film stock we have selected. While the Arri D-21 is capable of the same ISO, its native ISO is 200 and raising the ISO would have a consequence on noise. Therefore, each shot will use an ISO of 500 for the BL-III and an ISO of 200 for the D-21. Various filters will be used to compensate for lighting, and f-stops will be adjusted for underexposures only. The BL-III is shot using Kodak Vision3 500T film, and the D-21 will be shot in ARRIRAW.

The Scenes

#1 Charts

This scene contains the capture of 4 different image assessment charts, and a group of actors that span a skin tone range. The setup for each of the three charts is the same. The chart of choice will be raised to a height of 6 feet from the ground with the camera at the same height and be placed 6 feet away from the chart with two 300W lights beside and equidistant from the camera, and 6 feet from the camera on the axis parallel to the wall.

With the assistance of a light meter, each chart should be uniformly lit to a f-stop of f/4 at an ISO of 200.

The f-stop of f/4 will be the baseline exposure for each scene. The ISO12233 chart, OECF chart, MacBeth chart, and a gray card will all be exposed: normally, underexposed by 2 stops, and overexposed by 2 stops.

#2 Dynamic Range

Interrogation Scene

System dynamic range and exposure latitude will be analyzed through the use of a high dynamic range scene. Dynamic range is an important aspect of image quality because it dictates the extent to which the very dark and very light components in a scene can both be exposed properly without losing information.

To accomplish a high dynamic range, the set will be dressed to resemble a jail scene, comprised of a dark cell contrasted with a bright spotlight in an interrogation-like setup. Inside the cell, an actor will be sitting in a shadowy region , about 4-5 stops darker than normal. Outside the cell at the interrogation table, a bright light will illuminate objects on a table to about 8-9 stops brighter than normal exposure using multiple lights. The objects on the table will include a mug and papers, both with text on them with the purpose of determining whether either of the cameras can capture these details at such overexposed conditions.

Two additional actors will be stationed at the interrogation table, one acting as the investigator and the other as the suspect being interrogated. Their interaction will add some interest and movement to the scene. From an aesthetic perspective, the harsh lighting should be indicative of the discomfort felt by suspects during interrogation while the dim jail cell should represent the bleak outlook of a prisoner. We initially planned to over and underexpose this scene to examine the implications on exposure latitude, but the amount of light required to overexpose this scene for the 200 ISO of the D-21 is not practical.

Therefore, the exposure latitude will only be tested on the skin tone reproduction in the sharpness scene. This scene will allow us to compare the dynamic range of each camera. We will then make qualitative comparisons towards how much detail information is contained in both the highlights and the shadows, and how much information is crushed or clipped. We will also asses the noise content in the shadows using the uniform areas in the walls that surround the scene.

#3 Skin Tone, & Aliasing

A portrait scene will be used to assess the extent of aliasing in both cameras’ imagery. The scene will consist of a simple, plain background with a group of 3 people sitting on a sofa talking. The talent will be wearing shirts with very high frequency patterns which should result in aliasing in the imagery captured by the two cameras at normal exposure.

In addition, the shot will be over and underexposed by 2 stops to test the exposure latitude with the varying skin tones. This will allow us to examine how well the scene can be recovered from mis-exposure.

Implementation of a typical 3 point lighting setup will be used. This will provide the overall image with properly balanced shadows, and will represent the most common lighting setup to represent skin tones.

The primary purpose of this scene is to provide a scenario in which there is a decent color tone representation of various skin types. An additional factor to this scene will be the assessment of sharpness within the scene created by the selection of bright and high frequency patterned shirts.

Note: Unfortunately a lot of this scene has been lost over time and there are few photos on it.

#4 Color Reproduction

Cooking Scene

To investigate system color reproduction, a cooking scene will feature a “chef” (actor dressed in a white chef coat and hat) standing behind a table lined with a variety of colorful, recognizable, fruits and vegetables. The chef will be using a large knife to chop ingredients for a salad, which will be placed into a bowl.

Each camera will be dollied into the scene, slowly, to allow us to get both a wide shot of the whole scene and a close-up without having to do multiple takes, thereby wasting film. These shots will be shown side-by-side in the final video to compare the color reproduction abilities of the cameras.

The scene will also include a potted plant (indicative of vegetation color reproduction) and recognizable packaging (such as a Cheerios cereal box) in the background. The reason for shooting these particular items is because the colors should be easily recognizable to the majority of people, acting as “memory colors,” which will be instantly indicative of the color reproduction capabilities of each camera.

The lighting setup for this scene involves five lights, two 2Ks, two 5Ks , and a 1K. Both 2Ks and the 1 K are used as backlights with the 5Ks being used as key and fill lights

Note: Unfortunately the digital footage corrupted somewhere along the pipeline and was not recoverable.

#5 Green Screen

Now that digital cinema has been around for a few years, there is a unique workflow for VFX that differs from film. This part of the shootout experiment will compare some of the VFX elements between film and digital cinema.

The green screen will be lit with the lights that are already hanging from the grid above, and will be controlled by the dimmer board. The actor will be lit with a key light, fill light, and a back light. This will give the actor, or objects in the scene, a good separation from the green screen allowing for a proper key. A soft and more professional look will allow the keyed footage to composite well with background source footage.

Takes place in front of the Studio A Green Screen Wall. The actor will stand in front of the green screen, holding a Captain America Shield and Thor’s hammer. This scene will be processed in post-production for the use of assessing the ability of cameras to acquire an image that will create a good composite.

**Scene 1A Green Screen**

Takes place in front of the Studio A Green Screen Wall. The actor will stand in front of the green screen, holding a Captain America Shield and Thor’s hammer. This scene will be processed in post-production for the use of assessing the ability of cameras to acquire an image that will create a good composite.

**Scene 1B Green Screen vs Blue Screen**

Takes place in front of the Studio A Green Screen Wall, with a blue screen set up on the green screen. The actor will move in front of each screen, holding a Captain America Shield and Thor’s hammer. This scene will be a mini shootout in itself, comparing the green screen with the blue screen of each camera system; and what sort of artifacts can be seen.

**Scene 1C Difficult Composite**

Objects will be placed around the scene to give a difficult time for the compositor. Objects include: a MacBeth Color Checker Chart, a comb, a net, and a clear plastic shower curtain liner. The comb will be used in the scene in order to assess the noise associated with keying high-frequency content.

**Scene 2 Tracking**

Tracking points are setup within the scene and several tracker balls are tossed on camera to test the ability of each camera to track these objects. The question to answer is, how does the noise in each system affect the tracking ability; and to asses how much the MTF will affect tracking in post. This will be broken down into each color channel, since the noise present in each channel is different and will have varying results.

The purpose of this tracking test is to combine what we know about MTF degradation, noise, and color reproduction and see how it affects the tracked footage. This test is more on the post production side, but is very important for compositors because sharpness and noise can seriously affect a compositor’s track.