Image Comparisons

 

 

 

The BSC Film & Digital Image Evaluation 2009

Since the first evaluation in 2007 there have been many changes and advancements in the technology of image acquisition with new film stocks and data recording cameras being added to the mix, and this latest presentation is designed to shed some light on the reality of those changes.
Not only have digital cameras come on leaps and bounds but so have the film stocks and the tools associated with them.

There is so much misinformation about the capabilities of some formats, and such hype from manufacturers, that we felt we had to show what each camera was actually capable of. The results are intended to be of use to all departments concerned with film production, not only cinematographers but also hair and make-up, costume, production design, producers, directors and post production.

All the cameras and film stocks available today are just tools, they are not machines capable of creative choice, they are there to be used, manipulated, and pushed to their limits by creative cinematographers.

There is, perhaps, a misconception that HD is always cheaper than film. Yes it can be, if you talk in terms of tape stock versus film costs, but a producer must now consider the time taken to download the information from the new data recording cameras, add in the digital processing costs involved along with all the extra equipment required and additional crew such as Digital Imaging Technicians, and then add hard drives for storage and secure archive and you have a different picture. In addition to this, some of the processes that used to be purely the domain of post-production have now become an integral part of the work of the shooting crew and that also has to be taken into account when budgeting a production.

However we are NOT going to tell you which format is better than another. This is purely an investigation, and the results are intended to help you to make up your own mind. This is a camera evaluation rather than a post-production evaluation and as such we have tried to create a level playing field for the presentation of the material.

The shoot took place over 5 days on B Stage at Pinewood Studios, who very generously gave us the stage for the best part of a month.

The Dops were Gavin Finney BSC, Robin Vidgeon BSC, and John Daly BSC.

With 10 digital cameras and 16 film stocks involved, one of the considerations was to present the material in a way that will be as informative as possible. Each shot is presented in 3 parts, with groups of large, medium and small formats so you can compare like with like.

VIEWING RECOMMENDATION

The differences between the various formats are often quite subtle and will therefore not be visible on small or low quality screens or projectors, this includes all laptops and computer monitors. Viewing of the BSC Film & Digital Image Evaluation Blu-ray disc should ideally be on a high quality HD projector in a darkened room or on a large full HD (1080p) Plasma or LCD screen (not less than 42” preferably more than 50”) with the curtains closed.

THE CAMERAS

Arriflex 435 with

• 35mm Kodak 200T, 500T, 250D & 50D
• 35mm Fuji 250T, 500T, 250D & 64D
• Panavision Genesis
• Arriflex D-21
• Sony F35
• Red One
• Arriflex 416 with
• 16mm Kodak 200T, 500T, 250D & 50D
• 16mm Fuji 250T, 500T, 250D & 64D

S I 2k
Panasonic 3700
Sony F900R
Thomson Viper
Sony EX 3
Panasonic HVX 201
Canon EOS 5D Mk 2

As a control, every camera shot a grey scale and a Macbeth Colour Chart which were carefully lit to exactly 3200k for tungsten balanced cameras and 5600k for the daylight formats. A waveform monitor and histogram was used to verify the sensitivity of the digital cameras according to the manufacturers specifications. To ensure impartiality, technicians from the Camera Rental companies set up the digital cameras, not the DoPs.

The shots were lit as a technical exercise with extreme highlights and shadow areas, strong colours and a variety of skin tones, this was not biased in any way towards film or digital formats. Some differences are easy to see, such as the differences in depth of field, but others such as colorimetry can only really be assessed by use of a split screen presentation.

SET UP

The wide shots on the set used the Mo-sys Motion Control System to ensure the camera movement was exactly the same each time, the only variable being the movement of the actors, which was very consistent over the period of 2 days that it took to shoot each shot.

All the digital cameras that could record Raw data were invited to do so and we selected only the best takes, choosing whether that was the Raw data or tape recording. Generally, we chose the Raw data.

THE SHOTS

Interior/Exterior DAY on the set
Interior/Exterior NIGHT on the set
The Close Up
Exterior day in the Garden at Pinewood

POST PRODUCTION

It has taken 5 months in post-production to present all the material shot in 5 days. The multitude of formats all required different processing to bring them into line as 10 bit log DPX files. To provide a level playing field, the images were kept at the highest possible quality throughout post-production.

Initial Post-production, was mainly done at Ascent 142, supervised by Paul Collard.

The 35mm film was scanned on an Arriscan at 6k resolution, creating 4k DPX 10 bit log files and the 16mm film at 3k, creating 2k files to provide the highest quality images with no de-graining.

Whilst Grading we used the Grey scale & Macbeth charts that were shot for each format, with continual reference to the picture to ensure that the blacks were not crushed or the highlights clipped.

Only primary grading was allowed and every attempt was made to achieve a true rendition of the Macbeth Chart within this constraint. This was a purely technical grade and we made no subjective or secondary corrections to the colour.

Gwyn Evans was the colourist at Hat Factory Post.

The Editor was Torquil Dearden

The final format output is a DCP Digital Cinema package, in XYZ Colour Space.

The parameters of the process we employed in post-production were designed to show up and highlight differences between the formats, and whilst we acknowledge that a more tailored workflow may have produced better results from some cameras, it would not have been a fair comparison.
.