The Differences Between Film & Digital Video

Found this article while searching for some informaton on video preservation. Bloke form Kodak has some intersting things to sat about Digital vs Analogue mediums. Remember that the film the shot motion pictures is the same 35mm format ands pretty much the same negative emulsions as the film shot in 35mm still cameras (“cameras”).

The Differences Between Film & Digital Video

The following is a conversation with Tom Wallis, chief technical officer, Kodak Entertainment Imaging Division, who explains how images are recorded on film and digital video media, how the processes are alike and how they are different, and the possibilities for the future.

QUESTION: What happens when light strikes a frame of motion picture film?

WALLIS: Every frame of film has a thin emulsion layer composed of randomly distributed microscopic silver halide crystals that are suspended in a three-dimensional pattern. When photons of light strike the emulsion, a photo-electrical reaction takes place on these microscopic “solid state sensors.” These sensors react logarithmically to light and form a latent image consisting of tiny silver atoms embedded within the crystal structure.

QUESTION: How does the latent image become a picture?

WALLIS: When a laboratory processes the negative, the latent image is chemically amplified. This amplification chemically produces the dye images that form the picture in a color negative film. Since the eye reacts logarithmically to light, the resulting dye image is very close to what you see through the viewfinder.

QUESTION: How much control does the cinematographer have?

WALLIS: They have enormous creative freedom to interpret images by their choice of negative, the way they expose the film, the absence or presence of light, the color of light, the use of filters and diffusion, and how the film is handled by the lab. Films can be “pushed” or “pulled” and there are various other proprietary lab processes. Cinematographers can also manipulate colors, contrast and sharpness during digital postproduction and they can alter colors optically during timing in the lab.

QUESTION: What are the differences between various camera films?

WALLIS: Every negative is designed to provide specific imaging characteristics, including speed or sensitivity to light, grain, contrast and colors.

QUESTION: What happens when light strikes a CCD sensor?

WALLIS: With a CCD array, you have a rigid two-dimensional grid of pixels (picture elements) on the sensor. As photons strike each pixel, a photo-electrical reaction takes place kicking off electrons in a linear, rather than logarithmic, fashion proportional to the amount of light received. These electrons are stored in a “well” for subsequent amplification, electrical processing and storage on magnetic media. This fixed grid format is a key difference between the imaging systems. The random nature of the silver halide grains and resulting differentiation frame to frame is why you do not see such artifacts as aliasing with film.

QUESTION: What are the other differences between film and digital video?

WALLIS: Another key difference is dynamic range or in other words the capacity for recording shades of tonality and colors from darkest shadow to the brightest highlight. Film has a very broad dynamic range of 1,000:1 compared with about 100:1 with digital video cameras. The HD 24P camera performs well in the shadow areas when used at its rated speed but it still lacks the nuances you get with the “film look”. At the other end of the scale, however, digital video will blow out or “clip” highlights in higher contrast scenes.

QUESTION: What about picture resolution?

WALLIS: When you scan film on state-of-the-art equipment, you can get 4,000 pixels of information across the horizontal axis of 35 mm color negative. This 4K image compares quite favorably to the 1920 pixels across the horizontal axis of the new 24P HD digital cameras. Super 16 mm color negative film has 2000 pixels of information across its horizontal axis.

QUESTION: Does that mean that Super 16 film and digital HD are equivalent?

WALLIS: No way! While resolution may be about the same in terms of pixel count, pictures are more than pixels. With film you have the dramatic dynamic range, the full gamut of color, the texture of the image and the complete “film look” which is not offered by the digital HD system.

QUESTION: Some video vendors claim Super 16 film is too grainy for HD display and they have displayed tests that seem to indicate that.

WALLIS: Kodak has done tests with well-exposed Super 16 film transferred on the Philips Spirit DataCine and subsequently broadcast through an HD delivery system. The results are stunning! My point is this: different tests by different testers will produce different results. For example, at a recent SMPTE conference, a video vendor compared a new HD scene to an older film scene using a stock that Kodak manufactured years ago. They also did not choose the best scanning technology to convert the film image to digital video format. Ultimately you convince yourself on the relative merits with your own tests done in a fashion that is meaningful for the project at hand.

QUESTION: Do you agree with claims that digital video needs less light for exposure?

WALLIS: We believe that the “fastest” digital video cameras have a relative exposure index of 400 to 500. The HD 24P cameras used with the new Panavision lenses seem to be more like 200 to 300 at the maximum. Our film speeds range from an exposure index of 50 to 800. You can “boost” video and gain speed by sacrificing other imaging characteristics. You can also “push” film and trade speed for grain. Here again, personal experience will go a long way to sort out the speed issues. It is very important not to fall into the trap of thinking that any of these technologies reduces the creative aspect of lighting. Images are made of light and the cinematographer uses light and shadows to create the image. The notion that you can save time by not lighting is ludicrous.

QUESTION: Is there a way to define the “film look”?

WALLIS: We try to define it but all we can do is describe it. The word that many people use is organic. I believe at least, in part, that’s because the random nature of the silver halide crystals is similar to the way our eyes work. Everyone has a certain number of rods and cones in their retinas but they aren’t on a fixed grid. They are random, so all of us see the same thing a little differently. Part of the “film look” is also how color is recorded. Our estimate is that our current negative films can resolve about 20 million different colors. Theoretically you can do this with digital video but today the technology has not advanced to that point either in gamut or performance in various light sources.

QUESTION: Can you predict when digital video might catch up with film?

WALLIS: I can envision building a digital imaging system that is comparable to today’s film system once we solve some technical problems and also figure out how to make it economically feasible. The problem is that digital video is aiming at a moving target. We are making improvements in the entire film system. We are designing new films with specialized imaging characteristics. We are also seeing improvements in lenses that are very important. If you can limit the scattering and flaring of light in lenses, you can make film more efficient. In addition, there will be new intermediate and print films that will retain more of the original image quality captured on the negative. Film labs are also making advances in processing and printing technologies and we are working to improve both optical and digital projection.

QUESTION: What about hybrid technologies?

WALLIS: During recent years Kodak has earned several ATAS awards and an AMPAS award recognizing our contributions toward advancing the convergence of film and digital technologies. We earned the Emmy Awards for developing the imaging system for the Philips Spirit DataCine and the AMPAS SciTech award for the Kodak Lightning Film Recorder. Our Cinesite subsidiary has also done some pioneering work with digital intermediate technology that has allowed cinematographers to create “looks” in a digital suite.

QUESTION: What about claims that film technology is mature?

WALLIS: Our scientists believe we have the capacity to improve film speed, or sensitivity to light, by a factor ranging from two to seven stops of efficiency. I’m not sure what we would do with a film that is seven stops faster. But the point is that we can make dramatic progress without making anyone’s investment in cameras or other hardware obsolete. Similar advances in video technology tend to make hardware obsolete.

QUESTION: Is speed the only possibility for progress in camera films?

WALLIS: By no means! You can apply advances in the efficient capture of light to improve other aspects of image structure, including grain and sharpness, or maybe you want a film that has a different spectral response to colors. Maybe we’ll design films with a tonal scale designed for scanning devices that are optimized for hybrid applications, such as television programming or commercials. The possibilities are almost limitless.

QUESTION: It has been said that HD images are inherently sharper than film.

WALLIS: That’s a complicated subject because a lot of things contribute to the perception of sharpness, including contrast and depth of field. One of the things that happens in electronic image capture is that you’re amplifying the edges of objects and people, which makes the images seem sharper. That’s actually a problem when a cinematographer wants to use selective focus to draw attention to something in a frame. As an alternative, if you want a totally sharp picture, you can begin with film and enhance the image during digital postproduction.

QUESTION: How is today’s digital imaging technology different from the old analog systems that have been around since 1956?

WALLIS: With the old video systems, when light hit a photo-multiplier tube it was converted into an analog signal which was amplified, processed and then stored on magnetic media. Today, you use a digital camera to record a scene in much the same way, except you have a more tightly spaced sensor grid instead of a photo-multiplier tube. That basically allows you to sample at a higher density. Digital cameras have also eliminated image-smearing problems and there are a lot more controls on the camera. Real progress has been made.

QUESTION: Some people have speculated that Kodak is scaling down its investment in film technology. Is that accurate?

WALLIS: I suspect that may be wishful thinking by our competitors. We are investing a very significant part of a very large research budget in advancing silver halide technology. We are also investing in modernizing and building high-tech film manufacturing plants. We believe that film and digital imaging will co-exist during the foreseeable future with each offering its own benefits.

QUESTION: What about archivability? Is that an issue?

WALLIS: That’s an important question, because most people produce films for posterity. They want to be able to re-release and re-purpose the original film. I’ll give you an example. DVDs are usually longer than the original films because you want to intercut out-takes that the audience hasn’t seen before. Today, film is, by far, the best solution for archiving. Black and white separations will last for up to 500 years and color negative and intermediate stocks will last for hundreds of years, presuming they are stored correctly. Digital video is an improvement over analog video for archiving purposes because it consists of ones and zeros that enable lossless duplication. However, both digital and analog video are stored on comparatively volatile magnetic tape and disk media, and once a digital signal is gone, it’s gone forever. The other problem is obsolescence of formats. Some 74 video formats have been introduced since 1956, and even if the media survived, in many or most cases, there is no equipment for playback.

QUESTION: Is there a solution?

WALLIS: You can transfer important video programs to film for archiving, however most people who do that don’t invest in converting out-takes that might be needed in the future. In the longer term, we envision a digital recording medium that is more archival than magnetic tape. We are doing the research but we are probably a few years away from delivering a commercial product and then we will need standards and hardware before it becomes practical.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s